UEU-co logo

AccessSurgery – Print

  Print  |  Close Window

Note: Large images and tables on this page may necessitate printing in landscape mode.

Skandalakis’ Surgical Anatomy > Chapter 15. Stomach >


The anatomic and surgical history of the stomach is shown in Table 15-1. Some additional details are found later in this chapter under the heading “Surgery of the Stomach.”

Table 15-1. Anatomic and Surgical History of the Stomach

Matthis/Croll 1609 First gastrotomy: removal of knife; performed by Matthis; reported by Croll
Schwabe/Becker 1635 Schwabe removed knife from stomach; reported by Becker
Nolleson le fils 1767 First gastrorrhaphy
Gavard (1753-1802) ca. 1792 First to describe the oblique muscle fibers of the stomach wall
Merrem 1810 Pylorectomy successfully performed on dogs
Brodie 1814 Showed that vagal section inhibits gastric secretion in dogs
Delpech, Cayroche/Huard, Imbault-Huart 1819 Delpech advised gastrotomy for fork in a woman’s stomach; Cayroche performed gastrotomy; reported by Huard and Imbault-Huart
Lembert 1826 For intestinal anastomosis, employed interrupted sutures that included entire thickness of intestinal wall
Beaumont 1834 Observed gastric secretion via gastric fistula
Blondlot 1843 Performed gastrostomies in dogs
Sédillot 1849 First gastrostomy in human
Billroth, Gussenbauer, Winiwarter 1874-6 Billroth I – Distal gastric resection with anastomosis of stomach and duodenum (developed in dogs)
Billroth II – Anastomosis of stomach to jejunum through transverse mesocolon
Verneuil 1876 First successful gastrostomy in human
Péan 1879 First gastric resection for cancer (patient died fifth day)
Rydygier 1880 Second unsuccessful gastric resection
Billroth, Wölfler 1881 First successful resection on human patient
Wölfler and Nicoladoni 1881 Performed antecolic gastrojejunostomy
Loreta 1882 Described antral incision and finger dilatation of pylorus
Courvoisier 1883 Reported first use of retrocolic gastrojejunostomy
Heineke 1886 First successful pyloroplasty
Credé 1886 Reported 26 gastrotomies
Mikulicz-Radecki 1887 Performed successful pylorectomy for pyloric stenosis
Mikulicz-Radecki 1888 Performed pyloroplasty for bleeding duodenal ulcer
Billroth Clinic 1890 41 gastric resections for cancer, 19 successful (46.5%)
Kriege 1892 First successful closure of perforated gastric ulcer
Braun 1893 Described jejunojejunostomy as routine addition to gastrojejunostomy
Eiselsberg 1895 Divided stomach proximal to cancer and performed gastrojejunal anastomosis (antral exclusion)
Becher 1896 Showed the outline of the guinea pig stomach with lead contrast and X-rays
Pavlov 1897 Studied the mechanism of gastric secretion
Mikulicz-Radecki 1897 Initiated segmental resection. Indication for segmental resection was peptic ulcer of lesser curvature.
Schlatter 1898 Performed total excision of stomach for cancer
Wendel 1907 Performed esophagogastrectomy
Fredet 1907 Performed extramucosal myotomy for hypertrophic pyloric stenosis
Dufour, Fredet 1908 Performed successful extramucosal myotomy. Longitudinal incision with transverse closure.
Halsted 1910 Performed intestinal suture and anastomosis including strong submucous coat, but without having suture needle enter intestinal lumen
Rammstedt 1911-12 Performed successful extramucosal pyloromyotomy
Carlson 1912 Recorded stomach movements in animals with a balloon inserted through a fistula
Cannon 1912 Demonstrated with X-rays that hunger pains are caused by contractions of the stomach
Haberer 1919 Reported 2 successful gastric resections for perforated peptic ulcer
Latarjet 1922-23 In recognition of delayed gastric emptying, he added gastrojejunostomy to his vagotomy
Bircher 1925 Performed first selective vagotomy in humans
Lewisohn 1925 Reported the incidence of neostomal ulcer after gastrojejunostomy for duodenal ulcer to be 34%
Odelberg 1927 Reported 20 cases of perforated peptic ulcer treated by gastric resection
Judd 1930 Excised duodenal ulcer and performed transverse pyloroplasty
Reichel/Pólya (Pólya 1876-1944)   Performed anastomosis of entire opening of divided stomach to jejunum
Moynihan 1932 Reported 1000 cases of gastroduodenostomy or gastrojejunostomy with only 1 death and 1.8% anastomotic ulcer
Dragstedt 1943 Reintroduced truncal vagotomy with gastrojejunostomy for duodenal ulcer
Zollinger, Ellison 1955 Studied islet cell tumors and peptic ulcers of jejunum
Griffith and Harkins 1957 Performed and studied superselective vagotomy
Harkins (1905-1967)   Strong proponent of vagotomy plus antrectomy ad modum Billroth I. Reported recurrence rate of only 0.5% for duodenal ulcer.
Skandalakis et al. 1962 Reported 795 cases of smooth muscle tumors of stomach
Rhea 1965 Highest reported recurrence (24%) with Billroth II
Johnston & Wilkinson 1969 Performed selective vagotomy without drainage for ulcer
Debas 1974-1983 Very interesting publications about gastric physiology
Cheung 1976-1982 Studied the pathophysiology of gastric mucosa
Longmire 1981 Performed radical treatment of gastric carcinoma
Miller 1988 Published excellent book about physiologic basis of modern surgery
George et al. 1990 Demonstrated a link between Helicobacter pylori and ulcer 
Weerts et al. 1994 Review of laparoscopic gastric vagotomy for recurrent duodenal ulcers or pyloric ulcers

History table compiled by David A. McClusky III and John E. Skandalakis.


Garrison FH. History of Medicine (4th ed). Philadelphia: WB Saunders, 1960.

George LL, Borody TJ, Andrews P, Devine M, Moore-Jones D, Walton M, Brandl S. Cure of duodenal ulcer after eradication of Helicobacter pylori. Med J Aust 1990;153:145-149.

Holle F, Andersson S (eds). Vagotomy: Latest Advances. New York: Springer-Verlag, 1974.

Johnston D, Wilkinson A. Selective vagotomy with innervated antrum without drainage for duodenal ulcer. Br J Surg 1969;56:626.

Schmidt JE. Medical Discoveries: Who and When. Springfield, IL: Charles C. Thomas, 1959.

Skandalakis JE, Gray SW, Shepard D, Bourne GH. Smooth Muscle Tumors of the Alimentary Canal: Leiomyomas and Leiomyosarcomas, a Review of 2525 Cases. Springfield, IL: Charles C. Thomas, 1962.

Wangensteen OH, Wangensteen SD. The Rise of Surgery. Minneapolis: University of Minnesota Press, 1978.

Weerts JM, Dallemagne B, Jehaes C, Markiewicz S. Laparoscopic gastric vagotomies. Ann Chir Gynaecol 1994;83:118-123.


Normal Development

The primordia of the esophagus, stomach, and proximal duodenum are formed by the elongation of the embryonic foregut. During the fourth or fifth week of embryonic life dilatation starts at the area of the future stomach at the level of C3-C5. At the end of the seventh week the stomach may be found at T5-T10, not by “descent,” but by the cephalad growth of other embryonic entities. Growth of the trunk causes the stomach to locate between T10 and L3, its normal final position.

Does the stomach rotate and bend around its own longitudinal and anteroposterior axis? This is a controversial question.

Perhaps 90° clockwise rotation takes place around the longitudinal axis (Figs. 15-1A, 15-1B), pulling the dorsal mesogastrium to the left. The formation of the omental bursa (Fig. 15-1D) may be appreciated at this point. Because of this rotation the topographic anatomy of the vagal trunks changes: the left trunk innervates the anterior gastric wall and the right the posterior gastric wall (remember the mnemonic larp).

Fig. 15-1.

Schematic representation of positional changes of stomach. A, B, Rotation of stomach along longitudinal axis (anterior view). C, D, Effect of rotation on peritoneal attachments (transverse section). (Based on Sadler TW. Langman’s Medical Embryology, 5th Ed. Baltimore: Williams & Wilkins, 1985.)

The anteroposterior axis rotation (Fig. 15-2) changes the position of the gastric cardia and fundus, as well as the position of the pylorus and gastroduodenal junction. Therefore, the dorsal midline becomes the greater curvature and the ventral midline becomes the lesser curvature (Figs. 15-1C and 15-3). At the fourth fetal month the concavity of the lesser curvature is obvious and, at the eighth month, the fundic outgrowth (Fig. 15-4). This is because the ventral border (lesser curvature) of the future stomach grows more slowly than its dorsal border (greater curvature).

Fig. 15-2.

Schematic drawings of rotation of stomach around anteroposterior axis. Note change in position of pylorus and cardia. (Based on Sadler TW. Langman’s Medical Embryology, 5th Ed. Baltimore: Williams & Wilkins, 1985.)

Fig. 15-3.

Cross-sectional diagrams through the embryonic stomach during the seventh week illustrating two hypotheses for the rotation of the stomach. A, Classic concept: dorsal border rotates to left to become greater curvature. B, Concept of Dankmeijer and Miete.352 Left side becomes greater curvature as result only of increased growth. No actual rotation occurs.

Fig. 15-4.

Shape of stomach in prenatal stages and in adult. (Redrawn from Lewis FT. The development of the stomach. In: Keibel WP, Mall FP (eds). Manual of Human Embryology. Philadelphia: JB Lippincott, 1912; with permission.)

The human hand is used frequently as a mnemonic aid to instruct students about specific anatomic entities. To teach the embryology of the stomach, we use the left hand held in such a way that the hypothenar and fifth finger touch the epigastric area (Figs. 15-5, 15-6). With rotation, the palm touches the epigastrium; the hypothenar and fifth finger touch the greater curvature; and the thenar and thumb touch the lesser curvature. Remember that with further development, the proximal part of the stomach moves to the left producing the permanent fundus, and the distal part moves to the right producing the pylorus.

Fig. 15-5.

Location of left hand at embryonic foregut (beginning of dilated stomach). (Modified from Skandalakis JE, Gray SW (eds). Embryology for Surgeons, 2nd Ed. Baltimore: Williams & Wilkins, 1994; with permission.)

Fig. 15-6.

Use of left hand to demonstrate embryogenesis of stomach. Hypothenar and fifth finger touch the epigastrium vertically. If the same hand now turns so that the fifth finger and the hypothenar touch in a transverse way, the epigastrium and the palm in toto touch the epigastrium; then the fifth finger and hypothenar represent the greater curvature, and the thumb and index represent the lesser curvature. (Modified from Skandalakis JE, Gray SW (eds). Embryology for Surgeons, 2nd Ed. Baltimore: Williams & Wilkins, 1994; with permission.)

Some authors believe that positional rotational changes of the stomach do not take place. The gastroesophageal junction may be seen after augmentation of the fundus from the body of the stomach. O’Rahilly and Müller2 write about intrinsic growth changes in both gastric curvatures. They report that there is no evidence of gastric rotation. The embryologic enigma continues. But most likely O’Rahilly and Müller are right.

The rugae appear by the eighth week. The muscularis emerges between the eighth and 14th weeks.

The first glandular pits appear on the lesser curvature between the sixth and ninth weeks and by the tenth week are all over the stomach (Fig. 15-7). At birth approximately 200,000 pits and 500,000 glands may be found in the entire stomach. Parietal cells may be recognized at approximately the 11th week, and chief cells around the 12th week. At the same time mucous cells are found; pepsin is present in the mucosa by the last half of the sixth month.

Fig. 15-7.

A, Pattern of appearance of gastric pits by weeks of embryonic development. B, Pattern of maturation of gastric surface epithelium by weeks of development. (Data from Salenius P. On the ontogenesis of the human gastric epithelial cells: a histologic and histochemical study. Acta Anat [Basel] 1952;50 [Suppl 46]:1-76.)

Congenital Anomalies

The anomalies of the stomach are presented in Table 15-2. In Figure 15-8A their chief locations are shown; Figures 15-8B through 15-8K depict the anomalies. We describe here only some of the most important anomalies. The interested student will find more information in Embryology for Surgeons by Skandalakis and Gray.3

Table 15-2. Anomalies of the Stomach

Anomaly Prenatal Age at Onset First Appearance Sex Chiefly Affected Relative Frequency Remarks
Agastria and microgastria Week 4 Infancy ? Very rare Many associated gastrointestinal tract and splenic anomalies
Malposition Week 10 Any age Males Rare May be associated with diaphragmatic eventration or hernia
Atresia and stenosis ? Infancy to adulthood ? Rare Maternal hydramnios familial occurrence (?)—autosomal recessive associated with epidermolysis bullosa
  Membranous partial   Infancy to adulthood    
  Membranous entire ? Infancy to adulthood    
    Complete (solid) Weeks 6-7      
    Gastroduodenal discontinuity ?      
    Luminal with microscopic canal ?      
    Typical stenosis with perforated membrane(s)        
Hourglass stomach ? Any age ? Rare Usually not congenital
Congenital pyloric stenosis Postnatal week 2 2-4 weeks Male Very common Not a true anomaly
Congenital muscular defect Weeks 8-10 Infancy ? Rare May not be congenital
Diverticula ? 40-70 yr Equal Rare May not be congenital
Duplication—double pylorus Week 3 Any age Female Rare Faulty separation of endoderm and notochordal plate
Mucosal heterotopia Weeks 4-5       Adhesion or metaplasia
  From other organs to stomach
    Small bowel
  From stomach to other organs
Congenital arteriovenous malformations ? Adult Male Rare May not be congenital
Teratoma   Infancy to adulthood Male Rare Benign (usually)
Gastroduodenal adhesions ? ? ? ? ?

Source: Skandalakis JE, Gray SW (eds). Embryology for Surgeons, 2nd Ed. Baltimore: Williams & Wilkins, 1994; with permission.

Fig. 15-8.

A, Chief locations of congenital anomalies of stomach. B, Five types of gastric atresia. (a), Membranous atresia; (b), Perforated membrane (stenosis); (c), Luminal atresia with microscopic endodermal canal; (d), Complete solid atresia; (e), Complete atresia with discontinuity. C, Two explanations of gastric atresia. (a), Redundant endodermal lining. Caudal slippage of endoderm forms circular fold. Fold may or may not retain a central opening. (b), Attenuation of endoderm. Proliferation of endoderm fails to keep up with elongating foregut. Atresia or even complete discontinuity results. D, Rammstedt pyloromyotomy for infantile pyloric stenosis. E, Locations of 342 diverticula of stomach. Most are located on posterior wall of upper part of stomach. F, Duplication of the stomach. Those in top row are probably more common. G, Location of spontaneous rupture through congenital defects of gastric musculature. Numbers indicate order of frequency; over half occur on greater curvature. H, Inversion of stomach. Tracings from radiographs of 65-year-old woman. A, Antrum; B, Body; C, Cardia; F, Fundus. I, Sources of heterotopic tissues found in stomach. Thickness of arrows indicates relative frequency. J, Distribution of heterotopic pancreatic tissue in stomach. K, Location of heterotopic gastric mucosa in other organs of body. Thickness of arrows indicates relative frequency. (A-G, I-K modified from Skandalakis JE, Gray SW (eds). Embryology for Surgeons, 2nd Ed. Baltimore: Williams & Wilkins, 1994; with permission. E, Data from Palmer ED. Gastric diverticula. Surg Obstet Gynecol 1951; 43:432-443. H, Modified from Rhinehart BA, Rhinehart DA. Congenital abnormalities of the stomach. Radiology 1926;7:492-497. J, Data from Palmer ED. Benign intramural tumors of the stomach: a review with special reference to gross pathology. Medicine 1951;30:81-86; with permission.)

Microgastria and Agastria

Only in nonviable monsters is the stomach ever completely absent. Microgastria is an extremely rare congenital anomaly (fewer than 30 cases reported)4 and may be present as a small, tubular stomach which is associated with incomplete gastric rotation and megaesophagus. If conservative medical attempts such as total parenteral nutrition or continuous nasogastric feedings are not successful, the treatment is surgical. Jejunal reservoir was reported by Gerbeaux;5 gastrojejunostomy by Blank and Chisholm.6

The surgeon should be aware of further anomalies associated with microgastria and agastria: these include cardiac defects, asplenia, megaesophagus, and others.

Atresia and Stenosis

Gastric atresia (Fig. 15-8B and Fig. 15-8C) is limited to the antrum and pyloric region and, in most cases, is usually produced by a membranous mucosal diaphragm. This diaphragm may be perforated. If an antral or pyloric ulcer is present we conclude that atresia and stenosis is not congenital, but acquired.

The three cardinal symptoms in infants are persistent bile-free vomiting, upper abdominal distention, and decreasing quantity of stools.

The treatment of choice is surgery (excision of the diaphragm or gastroduodenostomy) before the 10th day of life.

Hourglass Stomach

Hourglass stomach is a division of the two gastric chambers by a constricting ring that may be of congenital origin. If symptomatic, surgery includes removal of the ring and gastrogastrostomy or gastrojejunostomy of the proximal chamber.

Congenital (Infantile Hypertrophic) Pyloric Stenosis

The clinical entity of hypertrophy and hyperplasia of the muscularis of the pyloric canal appears early in postnatal life. This may be why it is considered a congenital anomaly. According to McMullin,7 the incidence of infantile hypertrophic pyloric stenosis is approximately 2.4 per 1000 live births, and does not appear to be a seasonal variation by month of birth.

Pyloromyotomy (Rammstedt’s procedure) (Fig. 15-8D) is the operation of choice. It is a simple longitudinal incision of the pyloric musculature leaving the mucosa intact. The muscularis must be incised without injury to the mucosa. The incision is not sutured. Balloon dilatation of the pyloric canal is also reported to be successful.8

Congenital True Gastric Diverticula

Most cases of congenital true gastric diverticula are located in the upper posterior wall (Fig. 15-8E). All layers of the gastric wall are present, but there is no evidence of organic disease. If symptomatic, diverticulectomy or segmental gastric resection is performed.

Duplication of the Stomach

Duplications of the alimentary tract are named for the structures with which they are associated rather than for the mucosal lining within them. These rare lesions differ widely, so that few generalizations are possible. Most gastric duplications are located in the greater curvature of the stomach (Fig. 15-8F). If symptomatic, removal of the entire duplication is the ideal treatment.

Congenital Defects of the Gastric Musculature

Congenital defects of the gastric musculature are caused by failure of myoblast formation. The greater curvature is most frequently affected (Fig. 15-8G). With gastric perforation, the treatment of choice is surgery.

Malposition of the Stomach

The most dramatic form of malposition of the stomach is associated with situs inversus viscerum (Fig. 15-8H). Gastric volvulus associated with wandering spleen secondary to an absence of ligamentous connections of stomach, spleen and posterior abdominal wall has been reported.9 If gastric volvulus occurs, emergency surgery is necessary.

Gastric Mucosal Heterotopia: Recipient and Donor

The stomach can be the recipient and the donor of gastric mucosa, which is the most peripatetic entity of the GI tract. Figs. 15-8I and 15-8J illustrate the stomach as the recipient; Fig. 15-8E, as the donor. If symptomatic, conservative surgery should be performed.

Gastric Teratoma

Gastric teratoma is a very rare benign lesion. In 1964 Skandalakis, one of the authors of this chapter, and his colleagues10 reported the 13th case of gastric teratoma. They reported that in all cases in which specific data had been available the patients were males and the teratomas were benign. Subtotal gastrectomy or excision is the treatment of choice.

Lingawi and Filipenko11 reported gastric outlet syndrome secondary to a benign hamartoma of Brunner’s gland.

Surgical Anatomy

The stomach is a pouch connecting the abdominal esophagus and the first part of the duodenum. Gahagan12 speaks of the “termination of a tube, the esophagus, and the beginning of a pouch, the stomach.” From a surgical standpoint, however, we will include the gastroesophageal junction and the first portion of the duodenum with the stomach. These are the proximal and distal gastric connections. They are part of the “surgical stomach” which includes the cardiac orifice and the pyloric canal.

Topography and Relations

Few abdominal organs are as mobile as the stomach. Its position depends on the position of the individual, the degree to which the stomach is filled, the degree to which the intestine is filled, the tone of the abdominal wall, and the habitus of the patient. The stomach is higher in broad, stocky individuals than in slender ones. Anatomic characteristics (e.g., shape and size), pathologic disorders (e.g., diaphragmatic hernia), and excessively large meal consumption are factors that can play a role in the genesis of gastric volvulus.

The only fixed point of reference is the gastroesophageal junction, which lies to the left of the midline behind the seventh costal cartilage at the level of the tenth thoracic vertebra. Some authors consider the retroperitoneal duodenum a second fixation because it has lost its primitive dorsal mesentery. However, 1 inch (2.54 cm) of the first part of the duodenum retains both primitive ventral and dorsal mesenteries; this may represent a second fixation. For the remainder of the stomach there is no single “normal” projection.

Anteriorly, the stomach is in contact with the liver, diaphragm, and anterior body wall. Posteriorly, in the recumbent cadaver, the stomach lies in a bed formed by the diaphragm, spleen, left kidney and adrenal gland, pancreas, transverse mesocolon, and splenic flexure of the colon. The position and extent of the areas of contact of the stomach with these organs will vary in life.

The close proximity of the stomach to other organs, the sharing of its blood supply with these organs, and its lymphatic system all dictate rules and regulations to the surgeon so that anatomic complications can be avoided during gastric surgery. For example, the inflammatory process of a benign gastric ulcer can be responsible for fixation of the posterior wall of the stomach to the transverse mesocolon. A careless separation of the two structures could jeopardize the blood supply of the transverse colon by injuring the middle colic artery. Another example of possible danger is injury to the spleen from avulsion of its capsule during gastric surgery.

Maingot13 studied the direct invasion of gastric cancer to the neighboring organs. The most frequently involved organs are the colon, pancreas, liver, gallbladder, spleen, duodenum, and upper proximal jejunum. The authors of this chapter have seen direct invasion of the left hemidiaphragm, left adrenal, left kidney, and falciform ligament. Also, other organs such as the pancreas (Table 15-3) are responsible for direct invasion or lymphatic spread to the stomach as well as to other anatomic entities.

Table 15-3. Pancreatic Carcinoma


Tr, Transverse; Reg LN, Regional nodes.

*Data from Cubilla AL, Fitzgerald PJ. Metastasis in pancreatic duct carcinoma. In: Day SB, Meyers WPL, Stanley P (eds). Cancer Invasion and Metastasis: Biologic Mechanisms and Therapy. New York: Raven Press, 1977, pp. 81-94.

**Data from Howard JM, Jordan JL Jr. Cancer of the pancreas. Curr Probl Cancer 1977;2:25.

It is our opinion that esophageal and duodenal involvement from gastric cancer is an established fact, and the distal esophagus and proximal duodenum should be treated accordingly (Fig. 15-9). In 1967 Paramanandhan14 stated that gastric lesions invade the duodenum by direct extension, or lymphatic spread, or both. We agree.

Fig. 15-9.

Direct extension of gastric cancer.

Stomach and Peritoneum

The stomach is covered completely by the peritoneum, except in two small areas at the posterior surface of the cardia and at the proximal half of the first part of the duodenum. If we want to be more anatomically correct, the lesser and greater curvatures of the stomach are not covered by peritoneum. We see this during superselective vagotomy at the lesser curvature and during the ligation of the short gastric vessels at splenectomy. The two peritoneal layers are present anterior and posterior to the vessels.

Peritoneal Reflections

Derivatives of the Ventral Mesentery

Only the cranial portion of the embryonic ventral mesentery persists in the adult. The anterior part of this mesentery is represented by the falciform ligament (Fig. 15-10) between the liver and the anterior body wall. The posterior part of the ventral mesentery becomes the hepatogastric and hepatoduodenal mesenteries that form the lesser omentum.

Fig. 15-10.

Peritoneal reflections and ligaments of stomach. (Modified from Skandalakis JE, Gray SW, Rowe JS Jr. Anatomical Complications in General Surgery. New York: McGraw-Hill, 1983; with permission.)

Falciform and Coronary Ligaments

The falciform ligament, a remnant of the embryonic ventral mesentery, extends from the abdominal wall to the diaphragm and anterior surface of the liver. The free edge of the ligament contains the round ligament (the obliterated left umbilical vein).

The leaves of the falciform ligament separate as they reach the liver to form the anterior, superior, and posterior (inferior) layers of the coronary ligament. Laterally, these layers reunite to form the right and left triangular ligaments. On the left, the anterior and posterior layers of the coronary ligament are almost in apposition until they reach the abdominal esophagus. On the right, the layers diverge as they approach the inferior vena cava. Their wide separation is characteristic of the “right coronary ligament.”

In reality the so-called right and left coronary ligaments do not exist as separate entities, but for descriptive purposes it is convenient for the surgeon to use these terms for naming the left and right parts of the coronary ligament. From a surgical standpoint, the division of the left triangular and left coronary ligaments as a unit is anatomically justified for the exposure of the gastroesophageal junction. The surgeon must be careful to avoid the left hepatic vein, which is situated in the upper part of the left segmental fissure. (Our advice: After the left triangular ligament and lateral half of the left coronary ligament are cut by scissors, approach the gastroesophageal junction by finger dissection.)

Hepatogastric Ligament (Lesser Omentum)

The hepatogastric ligament is the proximal part of the lesser omentum. It extends from the porta hepatis to the lesser curvature of the stomach and upward as the ventral mesentery of the abdominal esophagus. The ligament regularly contains the left gastric artery and vein, hepatic division of the anterior vagal trunk, anterior and posterior gastric divisions of the vagal trunks (nerves of Latarjet), and lymph nodes and vessels. Also, in about one-quarter of individuals, one can encounter an aberrant left hepatic artery in the proximal part of the hepatogastric ligament. Distally, and to the right, the hepatogastric ligament contains branches of the right gastric artery and vein. In this region also are the common hepatic artery and portal vein; here they rise ventrally to gain their positions in the hepatoduodenal segment of the lesser omentum.

Hepatoduodenal Ligament

The hepatoduodenal ligament is the dextral part of the lesser omentum, extending from the liver to the first 2.5 cm of the duodenum. The free edge envelops the hepatic triad, which includes the proper hepatic artery, portal vein, and extrahepatic biliary ducts, in addition to the hepatic plexus and lymph nodes. By definition, the proper hepatic artery is the part of the artery distal to the origin of the gastroduodenal artery. An aberrant right hepatic artery, arising most commonly from the superior mesenteric artery (18-20%),15,16 may ascend toward the liver on the deep, or posterior, aspect of the structures within the hepatoduodenal ligament.

Weiglein17 studied the topography of the arteries in the lesser omentum in 138 cadavers. Only 9% had “normal” anatomy; the rest exhibited one or more of the following variations (Figs. 15-11, 15-12, 15-13 and Tables 15-4, 15-5, 15-6):


Aberrant hepatic arteries, 37%

Artery in border of hepatoduodenal ligament, 19%

Right hepatic artery crossing portal vein posteriorly, 4%

Right hepatic artery entering triangle of Calot anteriorly, 29%

Right hepatic artery entering triangle of Calot posteriorly, 7%

Accessory left gastric artery branching off left hepatic artery, 2%

Table 15-4. Variations of Arteries in the Lesser Omentum


NORM = incidence of normal anatomy, AGAS = incidence of accessory left gastric artery, CALOT = incidence of right hepatic artery crossing the common hepatic duct arteriorly or never, LIG = incidence of an artery in the free border of the hepatoduodenal ligament, AHA = incidence of aberrant hepatic arteries.

Source: Weiglein AH. Variations and topography of the arteries in the lesser omentum in humans. Clin Anat 9:143–150, 1996; with permission.

Table 15-5. Incidence of Aberrant Hepatic Arteries


NORM = right and left hepatic artery branching off the proper hepatic artery, AHAD = aberrant right hepatic artery, AHAS = abberant left hepatic artery, AHAD + AHAS = aberrant right and left hepatic arteries.

Source: Weiglein AH. Variations and topography of the arteries in the lesser omentum in humans. Clin Anat 9:143–150, 1996; with permission.

Table 15-6. Variations of the Right Hepatic Artery in the Triangle of Calot


POST = right hepatic artery crossing the common hepatic duct posteriorly (normal anatomy), ANT = right hepatic artery crossing the common hepatic duct anteriorly, NEVER = right hepatic artery crossing the common hepatic duct never.

Source: Weiglein AH. Variations and topography of the arteries in the lesser omentum in humans. Clin Anat 9:143–150, 1996; with permission.

Fig. 15-11.

Variations of arteries in lesser omentum. A, Normal pattern. B, Aberrant right hepatic artery (replacing type). C, Aberrant right hepatic artery (accessory type). D, Aberrant right hepatic artery (anastomosis with common/left hepatic artery). E, Right hepatic artery to right of bile duct. F, Cystic artery to right of bile duct. G, Right hepatic artery branching off gastroduodenal artery to right of bile duct. H, Right hepatic artery crossing portal vein posteriorly. I, Aberrant left hepatic artery (replacing type): hepatosplenic trunk and isolated left gastric artery. J, Aberrant left hepatic artery (accessory type). K, Aberrant left and aberrant right hepatic artery: gastrosplenic trunk and hepatic artery branching off superior mesenteric artery. L, Triple vascularization: aberrant left, aberrant right, and proper hepatic artery. M, Accessory left gastric artery branching off left hepatic artery. N, Triangle of Calot: right hepatic artery dorsal to hepatic duct (64%). O, Triangle of Calot: right hepatic artery ventral to hepatic duct (29%). (Modified from Weiglein AH. Variations and topography of the arteries in the lesser omentum in humans. Clin Anat 9:143-150, 1996; with permission.)

Fig. 15-12.

Aberrant left hepatic artery (L) branching off top of left gastric artery (G). Aberrant left hepatic artery gives rise to esophageal branches, celiac trunk, splenic artery (S), and common hepatic artery (H). (From Weiglein AH. Variations and topography of the arteries in the lesser omentum in humans. Clin Anat 9:143-150, 1996; with permission.)

Fig. 15-13.

Triple vascularization of the liver. Parts of the liver supported by a separate liver artery are broken apart. A, Aberrant right hepatic artery branching off superior mesenteric artery. B, Proper hepatic artery branching off common hepatic artery. C, Aberrant left hepatic artery branching off left gastric artery. (From Weiglein AH. Variations and topography of the arteries in the lesser omentum in humans. Clin Anat 9:143-150, 1996; with permission.)

Derivatives of the Dorsal Mesentery

The primitive dorsal mesentery gives rise to three structures in the upper part of the abdomen (Fig. 15-10): the gastrocolic ligament (not shown), gastrosplenic ligament, and gastrophrenic ligament.

Occasionally in the operating room and in the anatomic laboratory – especially with fresh cadavers – one of the authors of this chapter (JES) has seen several inconstant, or anomalous, peritoneal folds (derived from lesser omentum) which extended between the gallbladder and the duodenum, colon, or stomach in that order of frequency (Fig. 15-14). It is uncertain whether there is a relationship between these folds and the corresponding fistulous tracts of gallstone ileus; however, it is possible that these folds predispose to gastric, duodenal, and right colonic fistulae. This is hypothetical, of course, and reminds us of the gubernaculum testis with its tails (which of course do not exist) and its physiologic (normal or abnormal) destiny during the testicular descent.

Fig. 15-14.

Inconstant peritoneal folds of gallbladder to duodenum, colon, or stomach (in order of frequency). (Modified from Skandalakis JE, Gray SW. Surgical anatomy of intestinal obstruction. In: Fielding LP, Welch J (eds). Intestinal Obstruction. New York: Churchill Livingstone, 1987; with permission.)

Gastrocolic Ligament

The gastrocolic ligament is a portion of the greater omentum. It passes from the greater curvature of the stomach and the first part of the duodenum to the transverse colon, to which it is fused. On the left and craniad, it is continuous with the gastrosplenic ligament.

The gastrocolic ligament contains the right and left gastroepiploic (gastroomental) arteries and veins. If the recess of the omental bursa passes downward into the sac of the apronlike omentum, the gastrocolic ligament is said to be absent.

Gastrosplenic Ligament

The double-layered gastrosplenic ligament (Fig. 15-10) attaches to the greater curvature of the stomach. It is a downward continuation of the gastrophrenic ligament, or perhaps the gastrophrenic ligament is a continuation of the gastrosplenic ligament. Both are parts of the dorsal mesentery.

Michels15 showed that only the anterior leaf of the dorsal mesentery becomes entirely free to form the gastrosplenic ligament. The posterior leaf does not reach the gastroesophageal junction, so that there is a small “bare area” on the posterior wall of the stomach. The bare area lies over the left crus of the diaphragm, and is easily separated from it by the surgeon’s index finger. The left adrenal gland and the left gastric artery and vein lie close to this area.

Within the gastrosplenic ligament are:


Upper part: Short gastric arteries and veins and lymph nodes

Lower part: Left gastroepiploic artery and vein, terminal branches of the splenic artery, and lymph nodes

Gastrophrenic Ligament

The gastrophrenic ligament (Fig. 15-10) is continuous with the hepatogastric ligament to the left, or perhaps, opposite the esophagus. It has an avascular area through which the surgeon’s finger can safely pass and through which a Penrose drain can be inserted around the cardia to pull down the esophagus. This is a useful maneuver in vagotomy.

Within the lower part of the gastrophrenic ligament are short gastric arteries and veins and lymph nodes. The upper part is avascular.

Phrenoesophageal Ligament

The esophagus is attached to the diaphragm at the hiatus by a strong, flexible, “airtight” seal known as the phrenoesophageal ligament (Fig. 15-15). The pleura and peritoneum ensure that the seal is “airtight,” while collagen and elastic connective-tissue fibers provide strength and flexibility.

Fig. 15-15.

Coronal section through gastroesophageal junction and esophageal hiatus of diaphragm. (Modified from Skandalakis JE, Gray SW, Rowe JS Jr. Anatomical Complications in General Surgery. New York: McGraw-Hill, 1983; with permission.)

The fibers arise from the endoabdominal fascia of the diaphragm, pass through the hiatus, and insert into the adventitia and intermuscular connective tissue of the esophagus 1 or 2 cm above the hiatus.18,19 A second leaf of the fascia turns downward and inserts into the adventitia of the abdominal esophagus and stomach. A much weaker and less constant component arises from the endothoracic fascia and passes upward to join the fibers of the endoabdominal fascia.20,21 The fibers allow about 2 cm of vertical movement of the esophagus.22

The phrenoesophageal ligament is best seen in infants and children. With age, the esophagus is less firmly fixed in the hiatus, the ligament becomes less definitive, and fat appears between the surviving fibers. In our experience, by middle age the ligament has lost much of its identity. For all practical purposes, it does not exist in patients with longstanding hiatal hernia.23

Greater Omentum

From a surgical standpoint the greater omentum can be divided into two parts: an upper part, the gastrocolic ligament (described previously), and a lower part, the true “greater omentum” or fat apron. The greater omentum is formed by four peritoneal layers, two anterior and two posterior, which usually fuse together during development and growth (Fig. 15-16). Another late development is the fusion of the posterior layer to the transverse colon and the mesocolon, forming the “transverse mesocolon,” which therefore also has four peritoneal layers.

Fig. 15-16.

Formation of greater omentum. A, After rotation of stomach and formation of omental bursa. B, First fusion of posterior wall of omental bursa to transverse colon and mesocolon. C-D, Second fusion or attachment of anterior and posterior walls of omental bursa to form adult omentum. S, Stomach. P, Pancreas. C, Colon.

In infancy and childhood, the omentum is small and not well developed. The width and length of the adult omentum differ from individual to individual.

The greater omentum serves as the “policeman” of the peritoneal cavity, exhibiting protective action in inflammatory processes. It also serves as a depository for tumor metastases in ovarian carcinoma. Fukatsu et al.24 stated that the greater omentum may play an important role in host defense as a source of exudative neutrophils.

Blood Supply of the Greater Omentum

The blood supply of the greater omentum is presented very well by our late respected friend, Charles A. Griffith, in Surgery of the Esophagus, Stomach, and Small Intestine.25 We highly recommend this chapter.

The right gastroepiploic (gastroomental) branch of the gastroduodenal artery (Fig. 15-17) and the left gastroepiploic branch of the splenic artery usually anastomose along the greater curvature of the stomach. The right and left gastroepiploic arteries form the arc of Barkow through their right and left epiploic (omental) branches in the posterior omental layer. The arc of Barkow is reinforced by anterior epiploic arches which spring from the right and left gastroepiploic arteries and from posterior epiploic branches from the pancreatic vessels.

Fig. 15-17.

Arterial supply to greater omentum.

To maintain the viability of the omentum in surgery, the arc of Barkow can be preserved by several techniques.


For mobilization of the right side of the omentum:

1) transect the gastrocolic ligament inferior to the right gastroepiploic artery

2) ligate the anterior epiploics or the right gastroepiploic artery distal to the origin of the right epiploic vessels


For mobilization of the left side of the omentum:

1) ligate only the gastric branches of the left gastroepiploic arches

2) do not ligate the left epiploic arteries

Surgical Applications

Be aware of the relationships among the greater omentum, gastrocolic ligament, and transverse mesocolon. Remember the following:


The two-layer gastrocolic ligament (see “Embryogenesis” in chapter on peritoneum and omenta), whether short or long, is the best pathway for lesser sac exploration, especially close to the greater curvature.

Be careful with the 6-layer gastrocolic ligament. All the layers are fused and the middle colic artery as well as other omental vessels may be the cause of complications.

Gastric Divisions

The two well-defined borders of the stomach are the lesser and greater curvatures. Five arbitrarily defined regions are the:







In many anatomy texts, the pylorus is subdivided into the pyloric antrum, pyloric canal, and pyloric sphincter. The antrum is the funnel shaped subdivision of the pylorus which leads into the pyloric canal. The canal leads to the pyloric sphincter, which is more uniform in diameter. Often, “pylorus” is used synonymously to indicate that which is referred to in textbooks as the pyloric canal, or to the area of the pyloric sphincter.

The stomach is subdivided into its parts by lines (Fig. 15-18A) that very often are more imaginary than real:


A horizontal line passes from the cardiac orifice to the greater curvature, separating the fundus from the body, or corpus

An oblique line begins at the angular notch, and traverses the stomach roughly perpendicular to the greater curvature, thereby separating the corpus from the pyloric antrum

An oblique line extends superiorly from the sulcus intermedius of the greater curvature to the lesser curvature, separating the pyloric antrum from the pyloric canal

Fig. 15-18.

Morphologic and physiologic presentations of stomach. A, Three imaginary lines subdivide stomach. The first separates fundus from body, second separates body from pyloric antrum, third separates pyloric antrum from pyloric canal. B, Distribution of parietal (acid-secreting) cells in the areas of the stomach, each relative to the other. Percentages indicate that the maximal concentration of cells is in the body; in the fundus the concentration is about half that of the body; along the lesser curvature, about three-fourths. C, The area in which gastrin is produced by the pyloric glands; and the areas (cardia, pylorus, and proximal duodenum) in which mucus is produced. In the proximal duodenum, the mucus is secreted by the Brunner’s glands. (Data in B and C from Berger EH. The distribution of parietal cells in the stomach: a histotopographic study. Am J Anat 54:87, 1934.)

Neither in the operating room nor in the anatomy laboratory do we regularly find the angular notch and the sulcus intermedius. The gastroesophageal junction, lesser and greater curvatures of the stomach, and gastroduodenal junction – with the prepyloric vein of Mayo as an additional landmark if present – are the only visible landmarks. Therefore, for all practical purposes, there are no clear external landmarks to help the surgeon or anatomist accurately define the functional subdivisions of the stomach. We therefore agree with Griffith26 that histologic and physiologic studies are necessary for the delineation of the several parts of the stomach.

Morphology of the Stomach

In the ancient fable, several blind men encountered and touched an elephant, each touching a different part. Each then proceeded to describe the form of the elephant as he perceived it. The one who had felt the tail thought the elephant to be much like a snake; the second, who had felt the belly of the huge beast, described the animal as seeming like a wall; to the third, who had felt a leg, it seemed like a tree; and so on. In somewhat similar fashion, the stomach has been described from different points of view in textbooks and other literature.

The embryologist describes the stomach as a fusiform dilatation of the foregut, beginning at the gastroesophageal junction and ending at the gastroduodenal junction.

The anatomist describes the stomach in terms of its several parts, such as the gastroesophageal junction, cardia, fundus, body, antrum, pyloric canal, and sphincter, but also accepts the stomach as a distinct entity, a well-defined organ that is easily visualized, dissected, and demonstrated.

The physiologist and gastroenterologist describe the endocrine and exocrine stomach, treating the organ as two units, proximal and distal:


Proximal stomach. The proximal stomach (fundus and body) (Fig. 15-18A) receives and temporarily stores gastric contents. It is the home of the parietal cells (Fig. 15-18B) which secrete acid and intrinsic factors, as well as the home of the cells that produce Group 1 pepsinogen.

Distal stomach. The distal stomach (antrum and pylorus) (Fig. 15-18A) mixes and propels gastric contents. It also contains the area of the pyloric glands which produce the hormones gastrin and somatostatin (Fig. 15-18C).

Chang et al.27 reported that the distal stomach plus pylorus are likely to exert an important inhibitory mechanism in the regulation of gastrointestinal movement; vasoactive intestinal peptide was not found to be a major mediator of motility.

Relative to the maximal concentration of acid cells existing in the corpus (Fig. 15-18B), the concentration of acid cells in the fundus is approximately one-half; along the lesser curvature, about three-fourths.28 Both units – proximal and distal – work to produce Group II pepsinogens and bicarbonate. Debas29 states, “In truth, the entire stomach functions as both an endocrine and exocrine organ.”

The pathologist recognizes three subdivisions of the stomach: the fundus, body, and antrum. These provide the basis for descriptions of localized gross pathology.

The radiologist, in practical terms, locates the gastroesophageal junction just “below the junction,” and refers to the first part of the duodenum as the duodenal bulb.

From the surgeon’s viewpoint, the stomach is part of two almost-separate organ systems, each with its special pathology and surgical approach. The first of these systems can be called the “proximal gastric unit” (Fig. 15-19), and contains the proximal stomach, distal esophagus, and esophageal hiatus of the diaphragm. The second is the “distal gastric unit” (Fig. 15-20), and includes the gastric antrum and pylorus, together with the first part of the duodenum.

Fig. 15-19.

Proximal gastric surgical unit (shaded area). (Modified from Skandalakis JE, Gray SW, Rowe JS Jr. Anatomical Complications in General Surgery. New York: McGraw-Hill, 1983; with permission.)

Fig. 15-20.

Distal gastric surgical unit (shaded area). Most gastric surgery takes place in this area. (Modified from Skandalakis JE, Gray SW, Rowe JS Jr. Anatomical Complications in General Surgery. New York: McGraw-Hill, 1983; with permission.)

Proximal Gastric Surgical Unit

As noted above, the proximal gastric unit (Fig. 15-19) consists of the distal esophagus, esophageal hiatus, and proximal stomach. The esophagus joins the stomach in the abdomen, just below the diaphragm. The length of the abdominal esophagus is from 0.5 to 2.5 cm.30,31 Allison32 has pointed out that, by taking the level of the lowest connective tissue fibers attaching the esophagus to the diaphragm as the inferior limit of the mediastinum, there is technically no “abdominal” esophagus. In spite of this view, the surgeon has access to an appreciable length of esophagus below the diaphragm.

The abdominal esophagus lies at the level of the 11th or 12th thoracic vertebra, perhaps lower in tall, asthenic subjects, and higher in short subjects.31 Its relationships to surrounding structures are:


Anterior: Posterior surface of left lobe of liver

Posterior: Right crus of diaphragm, aorta

Right: Caudate (spigelian) lobe of liver

Left: Fundus of stomach

The histologic junction between the esophagus and stomach is marked by an irregular boundary between stratified squamous epithelium and simple columnar epithelium (Fig. 15-15).33 It is not coincident with the external junction. In the cadaver, this epithelial junction lies about 1 cm above the external gross junction.19 Above the boundary, islands of columnar gastric epithelium may be present at all levels of the esophagus.34 Such heterotopic patches are not pathologic.

In the living patient, identification of the line of transition between esophagus and stomach is more complicated than in the cadaver. The submucosal tissue is so loose that the mucosa moves freely over the underlying muscularis, bulging in folds into the stomach at each swallow.20 Even at rest, the junctional level can change. Palmer,35 using silver markers on the epithelial boundary, found that the junction was lower in the full stomach than in the empty one.

The mucosa that lines the body of the stomach is distinct from the rest of the stomach. Its glands are mucus-secreting, without chief or parietal cells. These are the cardiac glands of the histologist. Hayward18 opposed the use of the word cardia, and characterizes terms derived from it as “insufferably vague.” He suggested the term “junctional epithelium” for this area between the typical esophageal and typical gastric mucosae.

In Barrett’s syndrome (Barrett’s esophagus), metaplastic changes can, presumably, occur in the lower esophagus in response to persistent gastroesophageal reflux, resulting in a change in the epithelium from stratified squamous to columnar. Subsequently, in such sites characteristic peptic ulcerations may occur and, in some cases, adenocarcinoma.36

It has been hypothesized that the presence of cardiac mucosa at the gastroesophageal junction represents an early histologic manifestation of gastroesophageal reflux.37

Cardiac Sphincter

Because standing on one’s head with a full stomach does not cause uncontrollable regurgitation into the esophagus, there must be a closure of the cardiac orifice of the stomach that normally permits swallowing but not reflux. No true anatomic sphincter guards the orifice, although a slight thickening of the circular musculature of the distal esophagus has been described.33,38,39 Bombeck et al.19 found such a thickening grossly visible in 21 of 33 cadavers (64%). In most cadavers, as in most living subjects, the lower esophageal opening is normally constricted.

We endorse Botha’s statement that “there is no reason why the cardiac and pyloric sphincters should be compared, or why the presence of the former should be denied because it does not resemble the latter.”20 It should be remembered that except when a person is upside down, the cardiac sphincter does not have to bear the weight of the stomach contents and thus does not need to be as strong as the pyloric sphincter.

Various other structures have been held responsible for closing the cardia and preventing reflux from the stomach into the esophagus:


the angle of His at which the esophagus enters the stomach

the pinchcock action of the diaphragm

a plug of loose esophageal mucosa (mucosal rosette)

the phrenoesophageal membrane

the sling of oblique fibers of the gastric musculature

Further information about these structures can be found in a publication by Mann and colleagues.40 Some or all of the structures may play a part in cardiac closure, but the importance attributed to the sling of oblique gastric muscle has received the greatest support. Jackson41 has proposed the term “spiral constrictor” for the muscle arrangement at the gastroesophageal junction.

Regardless of the mechanism involved in closing the cardia, the lower esophageal sphincter, whose normal pressure ranges from 14.5 mm Hg to 34 mm Hg,42,43 resists esophageal reflux. Not every patient with hiatal hernia demonstrates reflux; the sphincter may function satisfactorily in the thorax. Conversely, in a few patients, esophageal reflux can exist without evidence of hiatal hernia. Fisher44 believed that pressure of the lower esophageal sphincter is the most important factor in preventing reflux in the unoperated patient. Following hiatal herniorrhaphy, other factors contribute to competence of the cardiac opening (see chapter on esophagus).

Relations of the Proximal Gastric Surgical Unit

(Fig. 15-21)


The proximal gastric surgical unit has relationships with the lesser and greater curvatures, the upper part of the lesser sac (omental bursa), and the gastroesophageal (G-E) junction.

The lesser curvature of the body, the G-E junction, and the abdominal esophagus are attached to the hepatogastric ligament and its contents.

The greater curvature attaches to the upper part of the greater omentum and the several related splenic ligaments. The anatomic entities of the G-E junction will be found in the chapter on the esophagus.

Fig. 15-21.

Relation of stomach to other organs in cadaver. A, Anterior relations. B, Posterior relations. Remember: in a living patient, these relations are highly variable. (After Anson BJ (ed). Morris’ Human Anatomy, 12th Ed. New York: Blakiston Division, McGraw-Hill, 1966, Fig. XI-52.)

Posterior relations are with the upper part of the lesser sac, the aorta and celiac axis and its branches, and the celiac ganglion and plexus.

Anteriorly the proximal unit is related to the anterior abdominal wall, left hepatic lobe and anterior segment of the right lobe, and diaphragm.



Anterior and posterior relations are quite variable.

The most important relation of the proximal gastric unit is with the left gastric artery, which should be ligated at its origin to include the lymph nodes associated with it and the celiac axis.

Distal Gastric Surgical Unit

From an embryologic, physiologic, and certainly a surgical viewpoint, the gastric antrum, pylorus, and first portion of the duodenum form a unit which is referred to as the distal gastric surgical unit (Fig. 15-20). Most gastric surgery takes place in this area, so it is much better known (at least to the general surgeon) than the proximal gastric unit.

Gastric Antrum

In the opened stomach, the antrum is easily distinguished from the body of the stomach by its mucosa, which is flatter and without rugae. The antrum (Fig. 15-18A) begins just distal to the termination of the gastric canal.45 It is also histologically distinct, being without chief or parietal (acid-producing) cells.

The margin of the antrum is irregular, but definite. Externally, the antrum is difficult to demarcate. The boundary on the lesser curvature usually lies at the incisura angularis, usually found in textbook drawings, but inconstant and often absent in the operating room. We agree with Griffith25 that “in the operating room this landmark either cannot be found or can be only vaguely located.”

If the surgeon is not planning a gastrotomy to locate the antral margin, he or she can use the “crow’s foot” of the anterior descending vagal trunk as a landmark (see “Anterior Gastric Division” under “Gastric Innervation” below). The antrum can be expected to begin 3 to 4 cm cranial to the “crow’s foot,” about 8 to 10 cm proximal to the pylorus. There is no good landmark on the greater curvature. In most cases the boundary extends from a point on the lesser curvature two-fifths of the way from the pylorus to the esophagus to a point on the greater curvature one-eighth of the distance from the pylorus to the esophagus.


The pylorus (Fig. 15-18A) is a region of the stomach variously called the pyloric canal, pyloric ring, and pyloric valve. Proximally, it merges into the gastric antrum without a definite external boundary; distally, it ends abruptly at the thin-walled duodenum. At its narrowest point, the luminal diameter never exceeds 19 mm.46,47 The size is important in estimating the optimal size of artificial openings, such as in gastrojejunostomies or pyloroplasties.

At the pyloroduodenal junction, the continuity of the circular musculature is interrupted by an anular septum that arises from the connective tissue of the submucosa. Proximal to this ring, the circular muscle layer is thickened to form the pyloric sphincter (Fig. 15-22). Distal to the ring, the circular muscle coat at the duodenum is thinner. The sudden decrease in wall thickness as one passes from the pylorus to duodenum results in an “os pylorus,” surrounded by a duodenal “fornix.”48 The existence of this fornix must be kept in mind when performing pyloromyotomy.

Fig. 15-22.

Pyloric sphincter. A, Normal. B, Infantile hypertrophic pyloric stenosis. Hypertrophy involves only circular muscle layer. (Modified from Skandalakis JE, Gray SW, Rowe JS Jr. Anatomical Complications in General Surgery. New York: McGraw-Hill, 1983; with permission.)

The longitudinal muscle layer is similarly interrupted, except on the side of the lesser curvature, where the most peripheral fibers of the pylorus maintain continuity with the fibers of the duodenum. The greater part of the longitudinal fibers turn toward the luminal surface and terminate in the thickened circular musculature. Torgerson49 described this arrangement in detail, concluding that, in addition to the obvious pyloric sphincter of circular muscle fibers, there is a dilator mechanism provided by the manner of termination of the longitudinal muscle fibers.

First Part of the Duodenum

The distal gastric surgical unit includes only the first 2.5 cm (1 inch) of the duodenum. It retains the primitive ventral and dorsal mesenteries, the hepatoduodenal ligament (the distal part of the lesser omentum), and the distal part of the gastrocolic ligament (greater omentum). This segment can thus be mobilized together with the pylorus and antrum of the stomach. In the supine patient, the first 2.5 cm of the duodenum may lie between the levels of the first and fourth lumbar vertebrae.

Relations of the Distal Gastric Surgical Unit

The lesser curvature of the antrum, the pylorus, and the upper border of the duodenum are attached to the hepatogastric and hepatoduodenal ligaments (lesser omentum). The greater curvature is attached to the gastrocolic ligament (greater omentum).


Posterior relations:


– Floor of lesser sac

– Transverse mesocolon

– Head and neck of pancreas

– Aorta and celiac trunk and its branches

– Celiac ganglion and plexus

– Hepatic triad

– Gastroduodenal artery

Anterior relations:


– Anterior abdominal wall

– Medial segment of left lobe and anterior segment of right lobe of liver

– Transverse mesocolon

– Neck of gallbladder (if stomach is empty)

Gastric Wall

The gastric wall consists of the serosa, the muscular layer, the submucosal layer, and the mucosal layer.


The serosa is nothing more than the peritoneum, a thin layer of loose connective tissue underlying a layer of simple squamous mesothelium.

Muscular Layer

The structure and arrangement of the muscular layer will differ, depending on the type of specialization of the person who speaks of it; each will be correct. The anatomist describes the three well-known layers of musculature: an outer longitudinal layer, middle circular layer, and inner oblique layer. The surgeon considers the three layers as one in the operating room. An investigator or a researcher sees that the muscular layer and the formation of gastric and duodenal ulcers might be related (a concept postulated by Oi50-54 and discussed in this chapter under “Oi Concept”).

The outer longitudinal layer of muscle is present along the lesser and greater curvatures. Its fibers are arranged into two groups. The first set is continuous with the outer smooth muscle layer of the esophagus. The fibers are best developed along the curvatures, ending proximal to the pyloric region. The second set begins in the body and passes to the right, with the fibers becoming thicker as they near the pylorus. Some of the more superficial longitudinal muscle fibers are continuous with those of the duodenum. As noted previously, other longitudinal fibers turn inward at the sphincter, becoming interlaced with its circularly arranged muscle layer.

The middle circular layer rather uniformly encircles the entire stomach and “is the principal part of the muscular coat,” according to Woodburne.55 This layer becomes thicker at the pylorus, most distally forming the pyloric sphincter. The circular muscle bundle is continuous with the circular muscle layer of the esophagus, but is separated from that layer of the duodenum by a septum of connective tissue.56

The inner oblique muscle layer, internal to the circular layer, is limited to the body of the stomach and is most developed near the cardiac orifice. These fibers sweep away from the cardiac orifice (many are nearly parallel with the lesser curvature) forming a rather distinct edge to this layer to the left of the lesser curvature. Other fibers, passing to the left, blend with the circular fiber layer as they approach the greater curvature.

The muscle layers of the stomach are responsible for gastric motility.

We quote from Sun et al.:57

[I]n adults who have had pyloromyotomy for infantile hypertrophic pyloric stenosis in infancy, patterns of pyloric motility are abnormal; pyloric tone is higher, whereas the number and amplitude of phasic pyloric pressure waves are less…the stomach has the capacity to compensate for changes in pyloric motility to minimize effects on gastric emptying.

Submucosal Layer

The submucosa is composed of loose, areolar connective tissue which connects the mucosa to the external musculature. It can be called the “vascular layer” (both arterial and venous) of the gastric wall. It houses plexuses that have rich anastomoses through extensive ramifications that form excellent collateral flow. All the parts of the gastric mucosa receive blood from these submucosal plexuses except the lesser curvature, which receives blood from the right and left gastric arteries. Griffith25 describes the submucosal plexus very well in his chapter on anatomy of the stomach and duodenum.

Mucosal Layer

Griffith’s organization of the layers of the wall of the stomach is very practical and very “surgical.”25,26 Following here, interspersed with modifications, is a summary of that organization which includes the distal esophagus, cardia, fundus and body, antrum, and proximal duodenum.

The distal esophagus is lined by stratified squamous epithelium, with mucous cells in the abdominal esophagus. Some esophageal mucosal glands (which are called cardiac glands because of their resemblance to others in the cardiac portion of the stomach) are located in the lamina propria. These secrete too little mucus to be protective against acid reflux from the stomach, however.58

A study by Fass and Sampliner59 found that squamous cell extension into the proximal stomach is a newly recognized mucosal abnormality with presently unknown clinical significance. This mucosal abnormality may represent an esophageal mucosal response to proximal gastric injury.

The cardia contains simple columnar cells. There is some minimal mucus production at the cardia. The small mucus-secreting, simple and branched tubular glands of the cardia contain a variable number of the acid-producing parietal cells.

The fundus and body contain two types of cells: parietal (oxyntic) acid-secreting cells and chief pepsin-secreting cells.

The lamina propria of the fundus and body is filled with branched, tubular gastric glands. The neck and base of these glands contain parietal, chief (zymogenic), and enteroendocrine cells.

The cytoplasm of the parietal cells is very rich in mitochondria. According to Debas,29 this is indicative of the high energy requirement of acid secretion. The parietal cells may also produce gastric intrinsic factor, which is extremely important in the absorption of vitamin B12, significant in preventing pernicious anemia. Parietal cells also produce histamine, which is a strong stimulus for acid secretion.

The chief cells contain the inactive enzyme pepsinogen and also produce the enzymes pepsin and lipase. When pepsinogen is released into the acid environment of the stomach, it converts into the highly active proteolytic enzyme pepsin.60

The enteroendocrine cells produce serotonin and endorphin, which act not only as gastrointestinal hormones, but also as neurotransmitters in the central nervous system when released into the bloodstream. Thus, in conjunction with intrinsic nerve plexuses of the stomach, such products may play very important roles in the control of gastrointestinal motility and local autonomic reflex activity.58

One of the principal secretory products of the fundus is serotonin (5-hydroxytryptamine). The large quantity of serotonin produced by carcinoid tumors of the cells here is responsible for the clinical symptoms of the disease.

The antrum contains G cells that secrete gastrin, and mucus-secreting cells. The cells of the pyloric glands produce mucus and appreciable amounts of lysozyme. Certain gastrin cells (C cells) produce gastrin, which stimulates the secretion of acid by the parietal cells. Other enteroendocrine (D) cells here release somatostatin, which inhibits the release of other hormones, including gastrin.60

Endocrine cells contained within the mucosa of the stomach are shown in Table 15-7.

Table 15-7. The Exocrine and Endocrine Cells of the Stomach and Their Secretory Products

  Cells Secretory Products
Exocrine Mucous Mucus
Oxyntic Acid
Chief Pepsin
Endocrine G Gastrin
D Stomatostatin
A* Glucagon 
EC Serotonin plus various peptides
ECL Unknown
P Unknown
X Unknown

*In fetus or newborn; only exceptionally found in adults.

Source: Debas HT. Physiology of gastric secretion and emptying. In: Miller TA (ed). Physiologic Basis of Modern Surgical Care. St. Louis: CV Mosby, 1988, pp. 280-291; with permission.

The proximal duodenum contains Brunner’s glands (Fig. 15-18C), which secrete alkaline mucus. The highly alkaline (pH 8.1-9.3) secretion acts to protect the duodenal mucosa against the effects of gastric acid and to bring the pH of intestinal contents to a level which is optimal for pancreatic enzyme action.

Histology and Physiology of Gastric Mucosa

The trilaminar gastric mucosa is formed by surface epithelium, lamina propria, and muscularis. It has several physiologic destinies.

The single layer surface epithelium is composed of various cells, according to location in the different anatomic areas of the stomach. In summary, there are mucus-secreting cells, chief cells, parietal cells, and endocrine cells. According to Davenport,61 the mucosa functions as a barrier, preventing hydrogen ions and bacteria from permeating and injuring the mucosa itself. In this, it is joined by the three barriers of the surface epithelium: apical cell membrane, cytosol (hyaloplasm), and basement membrane.

Tepperman and Jacobson62 stated that microcirculation plays an important role in prevention of injury to the mucosal layer. Ritchie and Mercer63 reported that gastric mucosal injury can be prevented by administration of prostacyclin, which may increase blood flow. According to Green and Dockray,64 microcirculation may also be augmented by the influence of sensory neurons, which are afferent and contain vasoactive neuropeptides.

Mercer65 commented on the role of the gastric mucosal barrier:

In summary, maintenance of the gastric mucosal barrier depends on the integrity of the surface epithelial cell plasma membrane and synthesis of epithelial secretory products. These products include, but are not limited to, bicarbonate, mucus, eicosanoids, nitric oxide and neuropeptides. Repair is a complex process but involves both restitution and surface epithelial cell proliferation and is influenced by a variety of trophic hormones. Gastric mucosal blood flow seems to be extremely important in the repair process and is regulated by a host of mediators.

Brown et al.66 reported the gastric mechanisms responsible for emptying separately solids and liquids as follows:


1. Liquids are “decanted” into the duodenum early.

2. By closure of the pylorus, large particles are not permitted to enter the duodenum.

3. The gastric antrum in late emptying grinds the large particles.

“Acid Antrum” and the Parietal Cell Mass Extension

In 1961 Lowicki and Littlefield67 experimentally defined the dimensions of the gastric antrum. In 1965 Moe et al.68 demonstrated the functional anatomy of the canine gastric antrum.

Griffith25 presented an excellent chapter about the anatomy of the stomach and duodenum. One of the original investigators of cell mass extension, he used Congo red pH to investigate cell mass extension distally. Discussing anatomic variations of the distal extension of the antrum toward the gastroduodenal junction, he stated that in some patients the parietal cell mass extended to within 2 cm of the pylorus.

Griffith also presented findings of several authors such as McCrea,69 Skandalakis et al.,70-72 and Nielsen et al.73 describing the variable length of the anterior nerve of Latarjet. The nerve may be short (8.5 cm from the pylorus) or may travel down to the first part of the duodenum. Griffith agrees with Poppen et al.74 that there is no connection between the distal extent of the parietal cell mass extension and the distal extent of the anterior nerve of Latarjet.

Popiela and Turczynowski75 state that the corpus-antrum border is located by the last branch of the anterior nerve of Latarjet. Groups of parietal cells can be located even in the pyloric area, which in this case is called the acid antrum. Donahue and Nyhus76 disagree with this view. They postulate that occasionally the posterior nerve of Latarjet is longer than the anterior in the same patient, and also that acid secretion is mediated by the nerves entering the greater curvature of the stomach.

When we use the term acid antrum, we mean a downward distal extension of the parietal cell mass in the pyloric area. Investigators disagree about the location of the corpus-antrum line. According to different sources, it may vary from 1 to 8.5 cm from the pylorus. Naik et al.77 (the Leeds group) found an “acid antrum” in 16% of cases. They reported that the border of the parietal cells exceeded the anatomic border by more than 1 cm.

Proximal gastric vagotomy is a good choice in treating duodenal ulcer because it:


minimizes operating room mortality

minimizes operating room anatomic complications

minimizes late complications

reduces gastric acidity to an acceptable level

has acceptable recurrence rates

Perhaps in the near future, if Donahue et al.78 and Rosati79 are right, the acid level and recurrence rates will be reduced enough that extended proximal gastric vagotomy will be considered the ideal procedure.

At the present time, duodenal ulcer is treated conservatively at first by eradication of H. pylori and by suppression or reduction of gastric acid. Only if conservative measures fail is surgery considered. We need more physiologic studies and more randomized surgical studies so the student of the vagus nerve and its vagaries will start evaluating the current procedures with greater insight.

An algorithm by Donovan et al.80 for selective treatment of perforated duodenal ulcer is shown in Figure 15-23. Kuzin and Alimov81 advocate selective proximal vagotomy and duodenoplasty as the procedure of choice for the treatment of duodenal stenosis in patients with duodenal ulcer.

Fig. 15-23.

An algorithm of suggested selective treatment of a duodenal ulcer that has perforated. Asterisk indicates future consideration of elective ulcer-definitive surgery. (Modified from Donovan AJ, Berne TV, Donovan JA. Perforated duodenal ulcer: An alternative therapeutic plan. Arch Surg 1998; 133:1166-71; with permission.)

Chang et al.82 reported that patients who underwent partial duodenectomy with highly selective vagotomy for obstructing duodenal ulcer gained weight. Further, the surgery was successful in restoring gastric emptying almost to normal.

Uravic et al.83 advocated that the treatment of choice for duodenal obstruction secondary to chronic pancreatitis is vagotomy and gastroenterostomy.

Peptic Ulcer Disease

Any discussion of peptic ulcer disease encompasses issues of anatomy, histology, and physiology. There are also questions of terminology and nomenclature for the condition itself. Is the term “peptic ulcer” correct? Maybe not. The terms “gastric” and “duodenal” are more exact.

What anatomic entity or physiologic action is responsible for the genesis of gastric and duodenal ulcers? Is the muscular network, the mucosal layer, or both, responsible for the production of peptic ulcer disease?

It is known that the ulcer (Fig. 15-24) always occurs at the alkaline side of the gastroesophageal junction. It is known that excess formation of hydrochloric acid and pepsin is responsible for the genesis of peptic ulcer disease, together with a lack of protection of gastric and duodenal mucosa.

Fig. 15-24.

Ulcers always form on alkaline side of junction in peptic ulcer disease.

Lu and Schulze-Delrieu84 stated that the most common site for peptic lesions in the pyloric segment was the protuberance of the lesser curvature called the pylorus torus. Many torus lesions extended into the destroyed distal pyloric muscle loop, causing widening of the gastric outlet and increased duodenogastric reflux.

Sometimes we are also confused by the so-called postbulbar ulcer. Is that ulcer located distal to the junction of the first and second parts of the duodenum, or is it located just distal to that part of the duodenum that is covered completely by the peritoneum? Remember, the first part of the duodenum is covered with peritoneum; the second part is located retroperitoneally. In other words, is the diagnostic “fleck” of the radiologist bulbar, postbulbar, or just duodenal? Is the duodenal bulb or cap the first part of the duodenum in toto, and, if so, where is the normal or abnormal “fleck” located? Does that “fleck” appear to represent the presence of an ulcer when it actually is just a “normal” collection of barium between rugae at the apex of the bulb? We don’t know. We are confused, and we are afraid that our radiologists may be uncertain also.

In this region, the anatomy of the duodenal mucosa must be interpreted by the sixth sense of the radiologist. Where exactly does the duodenal mucosa change from the “antral-like mucosa” to the parallel plicae mucosa of the small bowel? Does this happen at the middle of the bulb or at the end of the bulb? Radiologists, at least in conference, do not agree.

According to Portis and Jaffe,85 the postbulbar ulcer is located 5 cm distal to the pyloric ring. If we accept the number that represents the accepted length of the first part of the duodenum, then when we say postbulbar, we mean the area between the first part and the second part of the duodenum.

Duodenal Ulcer: Contributing Factors/Pathophysiology


High HCl

High pepsin

Low or abnormal mucus

Low alkaline pancreatic juice

Low bile

Hereditary factors

High stimulation of the vagus nerves

In 10% of cases,86 there is no duct of Santorini. Maybe this is an additional cause for the genesis of duodenal ulcer.

Gastric Ulcer: Contributing Factors/Pathophysiology


Normal, or low HCl

Low resistance of the gastric mucosa


– Drug induced

– Bile gastritis


For all practical purposes, the gastric mucosa is covered by mucus secreted by the mucus-secreting cells. The parietal cells, forming the tubular gastric or oxyntic glands, secrete hydrochloric acid, pepsinogen, intrinsic factor, and mucus. The pyloric glands secrete mucus, gastrin, and pepsinogen.

Of the four types of mucosa in the stomach, according to Griffith,26 the esophageal squamous epithelium is least resistant to acid-pepsin ulceration. Following the esophageal squamous epithelium, in order of increasing resistance, are the duodenal mucosa, the antral mucosa, and the parietal mucosa. Because parietal mucosa is the most resistant to acid-peptic ulceration, gastric ulcers are rare there.

Hurwitz et al.87 reported that gastric hypoacidity is not directly associated with aging. Their study of elderly white persons indicated that sequelae of achlorhydria, including bacterial overgrowth or malabsorption of drugs, should not be expected.

Sadchikov et al.88 believe that gastric ulcers should be regarded as a precancerous condition. Data by LaVecchia et al.89 confirm that the risk of gastric cancer increases in the presence of gastric ulcer.

Oi Concept

In 1959, Oi presented his classic work about locations of gastric, duodenal, and esophageal ulcers. He wrote about the association of esophageal, gastric, and duodenal ulcers,50 the location of gastric ulcer,51 and the location of the duodenal ulcer.52 Dual control of peptic ulcers by both the gastric mucosa and musculature was described in 1966,53 and a possible dual control mechanism in the origin of peptic ulcer was described in 1969.54 Wastell stated,90 “Oi’s theory is interesting, and it probably largely explains the localization but not the cause of some duodenal ulcers.”

Wastell contends that the distinction between gastric and duodenal ulcers “is becoming somewhat blurred” due to evidence that H. pylori is an etiological factor for gastric ulcer and gastritis, as well as duodenal ulcer.91

Indications and corresponding operative procedures for acute gastritis, gastric ulcer, duodenal ulcer, and Zollinger-Ellison syndrome are shown in Table 15-8.

Table 15-8. Choice of Operative Procedure Based on Indication

Indication Procedure
Acute gastritis Vagotomy and pyloroplasty with oversewing of erosions or near-total gastrectomy
Gastric ulcer Subtotal gastrectomy with ulcer excision
Duodenal ulcer  
  Intractable pain Parietal cell vagotomy
  Perforation Simple closure or closure and parietal cell vagotomy
  Bleeding Vagotomy and antrectomy with suture ligation of bleeding vessel or
Vagotomy and pyloroplasty with suture ligation
  Obstruction Vagotomy and antrectomy
Zollinger-Ellison syndrome Tumor resection or parietal cell vagotomy or total gastrectomy

Source: Ashley SW, Cheung LY. Gastritis and peptic ulceration. In: Miller TA (ed). Physiologic Basis of Modern Surgical Care. St. Louis: CV Mosby, 1988, pp. 292-309; with permission.

Helicobacter pylori

The finding in the latter part of the twentieth century that H. pylori causes ulcer disease is a very significant breakthrough in the field of gastroenterology. In 1997 Breuer et al.92 wrote that at least half of the world’s population is infected with H. pylori. Blecker93 stated that colonization with H. pylori is determined by childhood factors. Most people infected with H. pylori do not develop peptic ulcers, according to Wallace,94 so it is reasonable to assume that other factors in addition to the bacterium are necessary for the pathogenesis of peptic ulcer disease.

Blum’s95 historical overview of attempts to understand helicobacter-associated disorders records the following conflicting findings. In 1928 Konjetzny reported an association between gastritis and gastric ulcer. But Büchner, in 1927 and 1931, and Schindler, in 1946, reported no association between them. The controversy raged for many years.

It is now well accepted that H. pylori is associated with peptic ulcer disease and gastritis. Epidemiologic data strongly support an association between the bacterium and gastric cancer and lymphoma92 (Fig. 15-25). In 1983 Isaacson and Wright96 concluded that the development of gastric lymphoma originates from mucosa-associated lymphoid tissue (MALT).

Fig. 15-25.

Acquisition of H. pylori infection. Association of bacterium, gastric cancer, and lymphoma. (Modified from Ernst PB, Michetti P, Smith PD (eds). The Immunobiology of H. pylori: From Pathogenesis to Prevention. Philadelphia: Lippincott-Raven, 1997; with permission.)

Stephens and Smith97 reported that there is strong evidence that primary gastric lymphoma is associated with H. pylori, and they advised that eradication of H. pylori is the first step in the treatment of lowgrade MALT lymphoma.

Yamamoto et al.98 stated that there is a relation between residual gastritis and H. pylori infection. They also believe that H. pylori is not the sole cause of residual gastritis after partial gastrectomy.

According to Ota and Genta,99H. pylori colonizes the antral mucosa, oxyntic mucosa, and mucosa of the gastric cardia with equal frequency. A severe and active inflammation of the entire stomach is called pangastritis. According to Breuer et al.,92 duodenal ulcer is associated with an antral-predominant form of gastritis that leaves acid secretion from the corpus intact. In contrast, progressive multifocal gastritis with atrophy and intestinal metaplasia is an outcome associated with gastric cancer (Fig. 15-26). Xia et al.100 reported that with H. pylori infection, atrophic gastritis and intestinal metaplasia occurs primarily at the gastric antrum and incisura. Antralization of the gastric incisura in infected patients is associated with a greater risk of these two conditions.

Fig. 15-26.

Hypothetical natural history of H. pylori gastritis. DU, Duodenal ulcer. GU, Gastric ulcer. MALT, Mucosa-associated lymphoid tissue. (Modified from Ernst PB, Michetti P, Smith PD (eds). The Immunobiology of H. pylori: From Pathogenesis to Prevention. Philadelphia: Lippincott-Raven, 1997; with permission.)

In the preface to The Immunobiology of H. pylori: From Pathogenesis to Prevention,101 the authors write, “Some of the results to date in animal models have been so impressive that a vaccine seems to be tantalizingly close to reality.”

Sontag102 advises biopsy and histologic examination of the ulcer. He thinks that a gastric ulcer may be cured with a regimen of antibiotics, ranitidine bismuth citrate, bismuth, and proton pump inhibitors. Sontag prophesies that gastric ulcer disease will soon be “history.”

Feldman et al.103 found that serologic testing indicating undetectable antibody levels is a reliable indicator of cure of H. pylori infection in patients treated with antimicrobial drugs for more than one year.

Pernicious Anemia

A study by Hsing et al.104 confirmed that patients with pernicious anemia are at risk of developing gastric cancer and several other types of cancer. El-Newihi et al.105 presented a case of gastric adenocarcinoma associated with pernicious anemia. Sculco and Bilgrami106 collected data about 136 patients with pernicious anemia associated with gastric carcinoid tumors. Becker and Gabriel107 reported that gastric carcinoids may develop in patients with chronic atrophic gastritis with and without pernicious anemia.

Vascular Supply of the Stomach

It is well known that the stomach is among the best vascularized of organs (Fig. 15-27). Not only is it served by many arteries, but its wall also contains a rich anastomotic network of vessels extrinsically and intrinsically. General surgeons are well aware of this “friendly enemy” that may work for them, or against them. They know that the stomach can survive after ligation of all but one of its primary arteries, and that extragastric ligation will not control bleeding from a gastric ulcer.

Fig. 15-27.

Arterial supply to stomach. L Inf Ph, Left inferior phrenic artery. SG, Short gastric artery. LGE, Left gastroepiploic artery. RGE, Right gastroepiploic artery. S, Splenic artery. GP, Great pancreatic artery. Inf P, Inferior pancreatic artery. PD, Pancreaticoduodenal artery. DP, Dorsal pancreatic artery. GD, Gastroduodenal artery. RG, Right gastric artery. H, Hepatic artery. CT, Celiac trunk. LG, Left gastric artery. Post G, Posterior gastric artery. (Modified from Skandalakis JE, Gray SW, Rowe JS Jr. Anatomical Complications in General Surgery. New York: McGraw-Hill, 1983; with permission.)

Following is a listing, or summation, of all the arteries that supply the stomach. Each of the principal arteries of supply originates from the celiac trunk.

Left gastric


ascending branch (gives rise to esophageal)

descending branch (gives rise to gastric)



right gastric



– anterior superior pancreaticoduodenal

– retroduodenal

– posterior superior pancreaticoduodenal

– supraduodenal

– right gastroepiploic (major branches)



posterior gastric

short gastrics

left gastroepiploic (major branches)


Left Gastric Artery

In approximately 90% of individuals, the left gastric artery (Fig. 15-27) is a branch of the celiac axis. However, it may arise as an independent unit from the common hepatic artery, splenic artery, aorta, or superior mesenteric artery. In 4% it arises from a gastrosplenic trunk; in 3% it has a direct aortic origin; in 2% the left gastric is a branch of a hepatogastric trunk.108 Yildirim et al.109 reported a case of left gastric artery originating directly from the aorta.

The left gastric artery travels upward and to the left retroperitoneally to the proximal one-third of the lesser curvature of the stomach. Here, in about 95% of cases, it provides esophageal branches. According to Swigart et al.,110 it provides origin to 1-3 ascending esophageal branches in 78%, and to a cardioesophageal branch which supplies the esophagus, cardia, and fundus.

The left gastric artery commonly divides into an anterior and a posterior branch before attaining the lesser curvature111 and, in such cases, the esophageal and cardioesophageal arteries may arise, variably, from either of those vessels; more commonly, however, the cardioesophageal artery arises from the anterior gastric branch.

After its cardioesophageal branch the left gastric artery usually curves downward to the right and thereafter descends along the lesser curvature. As it descends, it bifurcates into an anterior branch which sends branches to the anterior gastric wall, and a posterior branch which, similarly, supplies the posterior gastric wall. In 25 of 60 specimens, El-Eishi et al.108 found both anterior and posterior branches. A smaller continuation of the left gastric artery continued along the lesser curvature.

The esophageal branches tend toward rather distinct anterior and posterior distributions toward the dextral side of the esophagus, which are not associated with the locations of the vagal trunks.108 An aberrant left hepatic branch from the left gastric artery occurs in about 30% of cases.111 The esophageal and cardioesophageal arteries often arise from it.108,110,111

Although the esophageal arterial supply is segmental in nature between its cervical and abdominal portions,110 anastomoses are rich between esophageal and gastric branches.111 The presence of such anastomoses between intercostal esophageal, left inferior phrenic (Fig. 15-28), and left gastric branches can result in troublesome retrograde bleeding from these branches on the posterior side of the esophagus.

Fig. 15-28.

Arteries of distal esophagus and proximal stomach. A, Inferior phrenic artery supplies margin of hiatus. Esophageal branch of left gastric artery supplies distal esophagus and passes through hiatus to anastomose with thoracic esophageal arteries. B, Branches of inferior phrenic and left gastric arteries supply distal esophagus with no thoracic anastomoses. C, Branches of inferior phrenic artery supply distal esophagus with thoracic anastomoses (rare). (Modified from Skandalakis JE, Gray SW, Rowe JS Jr. Anatomical Complications in General Surgery. New York: McGraw-Hill, 1983; with permission.)

From its origin, the anterior branch of the left gastric artery angles rather obliquely across the body of the stomach toward the greater curvature. It ends in numerous small ramifications and forms a vascular “crow’s foot” (of Payne) similar to the anterior gastric nerve of Latarjet. The anterior branch terminates about 4-6 cm proximal to the pylorus, about 1 cm from the lesser curvature, by piercing the muscular coat.108

In most cases, the posterior branch follows the lesser curvature a centimeter or two from its edge until it anastomoses with the right gastric artery. When the left gastric artery continues along the lesser curvature after providing origin to the anterior and posterior branches, it typically ends as a branch at the angular notch, and usually anastomoses with the right gastric artery. The first ramus of the posterior branch is relatively large and extensive in distribution, providing supply to the cardia and proximal posterior part of the corpus of the stomach.

The anterior and posterior gastric branches may possess direct interconnections with one another or with the continuing segment of the parent left gastric artery.108

Right Gastric Artery

The right gastric artery (Fig. 15-27) is a small branch which arises most commonly from the proper hepatic artery (50-68%), left hepatic artery (28.8-40.5%), common hepatic (3.2%), or other, less frequent sources, as noted by Eckmann and Krahn,112 and summarized by Van Damme and Bonte111 and Lippert and Pabst.113 Often, the novice student cannot find this thin, tiny artery, which avulses and disappears during dissection.

Some of the discrepancies in description of the usual origin of the right gastric artery are attributable to confusion in the nomenclature of the common hepatic and proper hepatic artery. Authors have often used the term “hepatic,” without designating whether they are speaking of the common hepatic artery or its proper hepatic branch.

The right gastric artery gives origin to one or more suprapyloric branches. Anterior and posterior branches from these anastomose with infrapyloric vessels and with the supraduodenal artery, providing for the distal gastric unit (antrum, pyloric canal, first inch (2.54 cm) of the first part of the duodenum). The right gastric passes along the lesser curvature for about 4 to 6 cm, about 0.5 cm from the lesser curvature, before anastomosing with the left gastric.108 In about 13% of individuals, the right gastric artery provides origin for the supraduodenal artery.111

Several authors, including Wilkie,114 Reeves,115 Barlow et al.,116 Nakayama,117,118 and Womack119 have studied the blood supply of the lesser curvature, distal stomach, and the first part of the duodenum. The reader can refer to their various works for discussions of questions pertaining to adequacy of blood supply, anastomoses, end arteries, and the relations of duodenal ulcer to arterial circulation.

Gastroduodenal Artery

The gastroduodenal artery (Fig. 15-27) arises as one of the two terminal branches of the common hepatic artery branch of the celiac trunk. Shortly after it arises from the common hepatic artery branch, the gastroduodenal artery gives origin to the supraduodenal, retroduodenal, and posterior superior pancreaticoduodenal arteries. The supraduodenal and retroduodenal arteries arise, variably, as branches of the posterior superior pancreaticoduodenal artery. The gastroduodenal artery ends by dividing into the right gastroepiploic and anterior superior pancreaticoduodenal arteries.

Splanchnic artery aneurysms of the gastroduodenal artery are very rare, and are often missed preoperatively. Konstantakos et al.120 reported obstructive jaundice in a patient with gastroduodenal artery aneurysm.

Right Gastroepiploic Artery

The right gastroepiploic artery (Fig. 15-27) is a branch of the gastroduodenal artery (or its continuation) in most cases. It occasionally arises from the superior mesenteric artery or from the anterior superior pancreaticoduodenal artery. After giving origin to an infrapyloric branch, the artery passes along the greater curvature of the distal gastric surgical unit within the gastrocolic ligament. It gives origin to 8-18 singular or paired anterior and posterior branches for the gastric wall.

The gastric branches of the right gastroepiploic artery pass mostly undivided in a submucosal position about one-fifth of the distance from the greater curvature. They anastomose extensively with branches from the left gastric artery.108 If the first infrapyloric branch is particularly large, the first following gastric branch tends to take its origin further around the greater curvature. In about 75% of individuals, the right gastroepiploic artery clearly anastomoses with the left gastroepiploic,111 although not so profusely as is seen in the anastomoses with the left gastric artery.

We quote from Sakamoto et al.,121 who reported an anomalous right gastroepiploic artery arising from the superior mesenteric artery:

The gastroduodenal artery ran in front of the common bile duct and descended along the posterior surface of the head of the pancreas (posterior superior pancreaticoduodenal artery). The enlarged pancreatic branch arising from the superior mesenteric artery mainly supplied the anterior surface of the head of the pancreas and then continued to become the right gastroepiploic artery.

For omental viability during distal gastrectomy, the right gastroepiploic artery should be preserved by one of the following methods: ligating its gastric branches; ligating the anterior epiploic; or ligating the right gastroepiploic distal and close to the origin of the right epiploic. One of the authors of this chapter (JES) prefers to ligate the gastric branches and to imbricate the greater curvature with continuous absorbable suture at its “nude,” not resected part.

Left Gastroepiploic Artery

According to Michels,15 the left gastroepiploic artery (Fig. 15-27) is a highly variable artery. It arises in most cases (72%) from the distal splenic, inferior splenic terminal, middle part of the splenic trunk, or superior splenic terminal. The observations of Van Damme and Bonte111 differ rather widely from those of Michels. They state that the left gastroepiploic arises from the splenic directly in only 26% of individuals. In 66% it takes origin from the splenic artery by a common stem with the inferior polar splenic branch.

The left gastroepiploic artery is the largest branch of the splenic artery.56 It reaches the stomach at a point about halfway along the greater curvature, after it has already delivered several rather long branches to the more proximal part of the greater curvature. These branches reach the stomach by way of the gastrosplenic ligament.

The left gastroepiploic artery gives off the left epiploic and the anterior epiploics. Together with similar branches of the right gastroepiploic, they form the arc of Barkow (Fig. 15-17), which is reinforced also by the posterior epiploic branches of the inferior pancreatic artery (see “Blood Supply of the Greater Omentum” in this chapter).

In splenectomy, ligation of the splenic artery and vein should be located distal to the origin of the left gastroepiploic vessels. Better still, the surgeon should ligate the splenic vessels very close to their entrance into the spleen. This avoids the postoperative bleeding which follows ligation proximal to the origin of the left gastroepiploic artery.

Short Gastric Arteries

Approximately five to seven short gastric arteries (Fig. 15-27) arise from the terminal branches of the splenic artery or from the left gastroepiploic artery. They may appear to emerge from the substance of the spleen at its hilum, having arisen from terminal splenic branches within the organ. The short gastric vessels take care of the arterial blood supply of the fundus and upper part of the body of the stomach. Although their branches are distributed both to the anterior and posterior surfaces, they appear to supply somewhat more of the posterior surface than the anterior.

The first short gastric artery reaches the greater curvature about 2-6 cm to the left of the esophagogastric junction.108 Some short gastric arteries are quite short; some are long. The shorter vessels are the troublemakers in splenectomy, and careful ligation is a must. Imbrication of the greater curvature is a safety measure, not only to avoid bleeding, but also to avoid possible necrosis with perforation. The short gastric arteries alone are inadequate to supply the distal part of the proximal gastric surgical unit.

 Read an Editorial Comment

Posterior Gastric Artery

An additional artery may arise variably from the proximal, middle, or distal segment of the splenic artery. It travels upward, posterior to the stomach, to vascularize the distal esophagus, cardia and fundus. This is the posterior (or dorsal) gastric artery (Fig. 15-27).

According to DiDio et al.,122 the posterior gastric artery was discovered by Walther in 1729123 and named by Haller in 1748. Its presence was overlooked by anatomists and surgeons alike for many years. This artery has received greater attention in recent years, particularly because of its potential clinical significance. Inadvertent ligation or division of the artery can cause complications including focal necroses and postoperative hemorrhage.

The posterior gastric artery has also been referred to as the accessory left gastric artery or ascending posterior esophagogastric artery.15 The artery is of significant size in one-half to two-thirds of patients.124 The posterior gastric artery alone is incapable of supplying the distal part of the proximal gastric unit.

Estimates of the frequency of appearance of the posterior gastric artery vary widely (Table 15-9), from about 12.7% to 99% of cases, most commonly ranging from 48% to 68%.124 Differences in the data are attributable in some measure to whether the vessel is identified with angiography or by dissection. The criteria for identification of the vessel differ also. In some studies, an artery was determined to be the posterior gastric only if it provided no supply to the spleen.111 In others, even if a branch to the superior pole left the vessel before it reached the fundic region, it might be considered to be a posterior gastric artery if it arose from the splenic artery prior to its distal terminal major splenic branches.125

Table 15-9. the Posterior Gastric Artery: Frequency of Incidence by Radiography or Dissection

Anatomic Dissections
Year Name of Author %
1910 Piquand 99
1983 Wald and Polk 88
1962 Aboltin 77
1963 Tanigawa (fetal studies) 67.8
1959 Chiarugi 66
1932 Versari 66
1904 Rossi and Cova 65.8
1967 Delteil et al. 64.3
1978 Suzuki et al. 62.3
1968 Levasseur and Couinaud 50
1912 Rio-Branco 50
1957 Weisz and Bianco 48
1991 Berens et al. 48
1980 DiDio et al. 46
1985 Trubel et al. 37.5
1967 Kupic et al. 36.8
1963 Tanigawa (adult studies) 36
1988, 1990 Van Damme and Bonte 36
1988 Trubel et al. 27.7
1928 Adachi 21.6
1915 Helm 16
1907 Leriche and Villemin 12.7
1972 Laude et al. 4.0

Source: Data from Berens AS, Aluisio FV, Colborn GL, Gray SW, Skandalakis JE. The incidence and significance of the posterior gastric artery in human anatomy. J Med Assoc Ga 80(8):425-428, 1991; with permission.

The posterior gastric artery arose from the proximal part of the splenic artery and was independent of splenic supply in 46% of cases in radiographic studies by DiDio et al.124 In dissection studies by Berens et al.,126 the vessel was found in 36 of 75 specimens (48%) (Table 15-10). It arose from the first three centimeters of the splenic artery. In studies by Van Damme and Bonte111 using combined techniques of radiography, corrosion, and dissection, the posterior gastric artery was found in only 36% of cases. This excluded any cases in which any branch passed to the spleen.

Table 15-10. Reported Incidence of Posterior Gastric Artery

Year Authors Incidence Breakdown
1729 Walther N/A N/A
1745 Haller N/A Origin-Midportion
1796 Sommerring “Sometimes” N/A
1873 Hyrtl “Inconsistent” N/A
1901 Haberer “In most cases” N/A
1904 Rossi and Cova 65.8% 2.5 cm from celiac
1907 Leriche and Villemin 12.7% Origin-Distal
1910 Piquand 99.0% Origin-Distal
1912 Rio-Branco 50.0% Origin-Proximal
1915 Helm 16.0% Origin-Proximal
1928 Adachi 21.6% 3-5 cm from celiac
1931 Testut and Latarjet N/A N/A
1932 Versari 66.0% N/A
1952 Franchi and Stuart N/A N/A
1955 Michels N/A N/A
1957 Weisz and Bianco 48.0% N/A
1959 Chiarugi 66.0% 2.5 cm from celiac
1962 Aboltin 77.1% N/A
1963 Tanigawa Adults 36.0% 2.2-13.1 cm from celiac axis
Fetuses 67.8%
1963 Couinaud N/A N/A
1967 Delteil et al. 64.3% N/A
1967 Kupic et al. 36.8% N/A
1968 Levasseur & Couinaud 50.0% N/A
1972 Laude et al. 4.0% N/A
1977 Ruzicka and Rankin N/A N/A
1978 Suzuki et al. 62.3% 18.4% Proximal third
47.8% Middle third
34.2% Distal third
1980 DiDio et al. 46.0% N/A
1983 Wald and Polk 88.0% N/A
1985 Trubel et al. 37.5% 33% had splenic branch
1986 Van Damme and Bonte 36% N/A
1988 Trubel et al. 72.9% PGA only 27.7%
GSA 45.2%
1990 Yu et al. 84.0% 13% Proximal third
78% Middle third
9% Distal third
1990 Kaneko Fetuses–16.0% N/A
1991 Berens 48% Proximal third (3 cm)

PGA, Posterior gastric artery; GSA, Gastrosplenic artery.

Source: Berens AS, Aluisio FV, Colborn GL, Gray SW, Skandalakis JE. The incidence and significance of the posterior gastric artery in human anatomy. J Med Assoc Ga 80(8):425-428, 1991; with permission.

Based on personal observations of the authors of this chapter, and on those of Trubel et al.,127 we present the following generalization: the further distally along the splenic artery one finds an artery arising which passes vertically toward the fundic region, the more likely it is to provide a superior splenic polar branch. Most distally, at the terminal branching of the splenic artery, a small fundic branch to the stomach may arise from the superior polar branch.

Trubel et al.125 suggest the use of alternate names for the branching vessel, and provide incidence percentages for various patterns:


Posterior gastric artery (stomach supply only), 27.7%

Gastrosplenic artery (supplying stomach and superior polar branch), 8.7%

Superior polar artery (with small gastric branch), 2.9%

Some aberrations of data on the frequency of appearance of the posterior gastric artery can be eliminated by discounting all arterial branches which reach and then pass along either the lesser curvature (the accessory left gastric) or the greater curvature (short gastric arteries), along the line of omental attachment. This is in keeping with comments by Van Damme and Bonte.111


The basic pattern of naming vessels relates to the target attained and supplied by an artery, not to the origin of the artery. Only by adhering to this guideline as closely as possible, as often as possible, can the greatest level of agreement of observations be attained.

Celiac Axis and Median Arcuate Ligament

The esophageal hiatus is separated posteriorly from the aortic hiatus by a fusion of right and left diaphragmatic crura. Crural fibers form a cord over the anterior surface of the aorta just above the origin of the celiac trunk. If the tendinous portions of the crura are fused, the median arcuate ligament is formed as a round, fibrous cord 1-3 mm in width, behind the esophagus and in front of the aorta. This structure holds sutures well and is extremely useful in vertical posterior approximation of the crura128 (Fig. 15-29). If the fusion is of muscular fibers only, the ligament may be ill-defined, or even absent.129 In five of ten cadavers with hiatal hernia in a study by Androulakis et al.,23 a definite median arcuate ligament was present. In the others, it was poorly developed.

Fig. 15-29.

Median arcuate ligament. (Modified from Skandalakis JE, Gray SW, Rowe JS Jr. Anatomical Complications in General Surgery. New York: McGraw-Hill, 1983; with permission.)

The median arcuate ligament passes in front of the aorta at the level of the first lumbar vertebra, usually just above the origin of the celiac trunk. The left and right celiac ganglia lie on either side of the base of the celiac trunk, concealing its origin from the aorta.

The celiac trunk appears in the fifth week of embryonic development at the level of the seventh cervical segment. It descends rapidly by caudal replacement of its roots, and by aortic elongation during the sixth and seventh weeks.130 The limit of this descent is variable. George131 found that the celiac trunk usually reached its final position at the lower third of the first lumbar vertebra. An artery covered by the median arcuate ligament is usually the result of an unusually low ligament, rather than an unusually high artery.129

Among 75 cadavers dissected by Lindner and Kemprud,129 the artery originated cranial to the median arcuate ligament in 26% of specimens. Such a covered celiac artery can be compressed by the ligament. In 1955, Michels15 described a cadaver in which the celiac trunk was narrowed by the fused crura so that it was smaller in diameter at its origin from the aorta than it was distally. Collateral arterial flow occurred in the form of an enlarged dorsal pancreatic artery.

Celiac artery compression syndrome has an enigmatic etiology. Since both the median arcuate ligament and the celiac ganglion are very closely related to the celiac artery, it is not known which anatomic entity is responsible for this rare clinical picture. The median arcuate ligament is most likely involved in celiac artery compression syndrome.

The diagnosis of celiac artery compression syndrome is difficult.132-134 Taylor et al.135 found compression of the celiac artery by the arcuate ligament to be reversible with relaxation of the diaphragmatic crura at deep inspiration. They documented this phenomenon with ultrasound and arteriography.

Compression of a celiacomesenteric artery has been reported.136 Celiacomesenteric trunk is a rare (1%-2.7%) embryologic anomaly.137 Because many patients with celiac arterial stenoses are asymptomatic, collateral circulation is probably adequate. Radiographic evidence of celiac compression has been assigned often to atherosclerotic plaques, until operation revealed the extrinsic cause of obstruction.138 Permanent fibrotic changes in the intima of the compressed artery have been described in some patients. In these, sectioning the median arcuate ligament will not relieve the stenosis.139

Hill140 advocated the use of the median arcuate ligament to fix the gastroesophageal junction, where possible. He believed that the primary attachment of the esophagus is by a strong, fibrous, posterior phrenoesophageal ligament that inserts into the preaortic fascia and the median arcuate ligament.141,142

Adams and Lobb143 described esophagoaortic hiatal hernias in which there was no arcuate ligament. The esophageal and aortic openings were confluent. Such confluence of the openings was denied by Hayward.144 Regardless of the merits of this controversy, Gray et al.145 found the ligament sufficiently well-developed to be used in hiatal hernia repair in only 14 (56%) of 25 cadavers without hiatal hernia that they dissected.

Takach et al.146 advised tailoring the operation for celiac compression according to findings in the operating room. They recommended decompression by division of the median arcuate ligament and the sympathetic neural fibers, and, when necessary, revascularization.

Arterial Circulation of the Gastric Walls

As previously stated, the abundant arterial blood supply of the stomach arises from diverse sources. The stomach can survive ischemia or necrosis after ligation of all but one of its major branches. It is well known that the stomach may have adequate circulation after ligation of three major branches (left gastric, hepatic, and splenic) with preservation of the right gastric or right gastroepiploic artery and leaving in situ the other ligated branches. As each of the smaller branches of these arteries perforates the muscularis of the stomach, it supplies the submucosal plexus.

The main arteries of the plexus are about 200 Ìm in diameter. These supply arterioles that pass inward to the plexus. The arterioles are characterized by arteriovenous anastomoses no larger than 140 Ìm in diameter.116 The volume of blood reaching the mucosa is controlled by these anastomoses.147 It is this submucosal plexus that interconnects the extragastric arteries and allows the stomach to tolerate extensive ligation of its extrinsic blood supply.

It should be noted that the mucosa of the lesser curvature is supplied by small extrinsic branches of the left and right gastric arteries, rather than by vessels from the submucosal plexus.116

With the richness of the vascular supply of the stomach, why do necrosis and perforation of the lesser curvature occur with proximal selective vagotomy in some cases? Similarly, what accounts for the necrosis and perforation of the greater curvature after ligation of the short gastric arteries during splenectomy? Could this be due to failure of collateral circulation, as Griffith26 suggests, or something else? Perhaps it might relate to vascular variations.

Ischemic necrosis of the gastric remnant was reported by Rutter,148 Spencer,149 and Fell et al.150 The submucosal plexus, which contains many arteriovenous shunts, is the collecting arterial blood reservoir. There is no question that this plexus is the central point of collateral circulation. As a matter of fact, dye can reach all intramural branches of all four major arteries after ligating all extramural vessels and injecting any one of the four major arteries.151 Peters and Womack152 demonstrated that the maximum supply of arterial blood to gastric mucosa for the production of gastric fluids depends on the closing mechanism of the arteriovenous shunts and, therefore, of the stasis of arterial blood in the gastric mucosa.

Because of these findings, Griffith25 stated the following, with which we agree:


Ligation of extrinsic arteries will not control bleeding from gastric ulcer.

The stomach remains viable after ligation of all arteries except the right gastroepiploic artery and perhaps the right gastric artery.

With subtotal gastrectomy, the gastric remnant remains viable with circulation from the left gastroepiploic artery and the short gastrics, because the left gastric is ligated. The posterior gastric artery, if present, participates in the survival of the remnant.

With 90% to 95% gastrectomy, the remnant stays alive and well because of the descending esophageal arteries. As Griffith stated, however, fatal necrosis can occur.

Womack119 pointed out the potential danger that accompanies the frequent presence of gastric arteries of anomalous and perhaps unsuspected origin. He advised consideration of this problem.

The blood supply of the proximal gastric pouch depends on three sources:


Ascending branch of the left gastric artery

Short gastric arteries

Posterior gastric artery (if present)

If the spleen is removed, the short gastric arteries must be sacrificed. Every effort should be made to avoid the ascending branches of the left gastric arteries. Adhesions of the posterior gastric wall must not be cut, as such adhesions may contain the posterior gastric artery.

Inferior Phrenic Arteries

A few words are added here about the inferior phrenic vessels which, for all practical purposes, are not strictly gastric vessels, but phrenoesophageal. Swigart et al.110 noted that the left inferior phrenic artery (Figs. 15-27, 15-28) provided vascular supply to the abdominal esophagus in about 56% of specimens, the right inferior phrenic artery in 3.3%.

The left inferior phrenic artery crosses the left crus and passes behind the esophagus, then forward along the left side of the esophagus. It gives off superior suprarenal branches to the left adrenal gland. This artery arises from the celiac trunk in 52.2% of cases, and from the aorta in 44%.153 In the remainder of instances, the left inferior phrenic arises from the left gastric, left renal, or aberrant left hepatic arteries. In 56% of Swigart’s110 specimens, the left inferior phrenic artery also provided an esophageal branch; in about 4%, such a vessel arose from the right inferior phrenic artery.

Vascular Supply of the First Part of the Duodenum

The blood supply of the duodenum is confusing due to the highly variable nature of the origin and distribution of the vessels supplying it. This is especially true of the blood supply of the first portion of the duodenum. In his excellent presentation about the stomach and duodenum, Griffith25 is very cautious in discussing the blood supply.

Akkinis153 stated that there is no collateral circulation beyond the terminal arcades of the small bowel. Do we have the same phenomenon in the first portion of the duodenum? What about the anemic spot of Mayo which corresponds to the distribution of the supraduodenal artery? Does it exist? Do the variations of the arteries represent, as Griffith25 speculated, an underlying factor in necrosis and leakage? We do not want to take a position. Our only advice is good surgical technique and a restrained approach for benign disease.

The first part of the duodenum is supplied by the supraduodenal artery and the posterior superior pancreaticoduodenal branch of the gastroduodenal artery (retroduodenal artery of Edwards, Michels, and Wilkie). According to Michels,15 the supraduodenal artery is not an end artery. The gastroduodenal artery is a branch of the common hepatic artery. In many individuals, the upper part of the first centimeter is also supplied by branches of the right gastric artery.

After giving origin to posterior superior pancreaticoduodenal and supraduodenal branches, the gastroduodenal artery descends between the first part of the duodenum and the head of the pancreas. It terminates by dividing into the right gastroepiploic and anterior superior pancreaticoduodenal arteries. Both supply twigs to this part of the duodenum.

Surgical Applications


The submucosal arterial network supplies the anterior and posterior gastric walls except at the area of the lesser curvature. This area is supplied by minute branches springing directly from the left and right gastric arteries (which perhaps are end arteries). If the surgeon decides to excise the bleeding ulcer of the lesser curvature for biopsy, it is essential after closing the defect to imbricate in order to avoid necrotic perforation. Similar reperitonealization and invagination should take place at the greater curvature after ligation of the short gastric vessels.

The vessel responsible for bleeding duodenal ulcer is, most likely, the gastroduodenal artery itself or its posterior superior pancreaticoduodenal branch if the origin is high and if the branch is fixed to the posterior wall of the first part of the duodenum. Our late friend Charles Griffith26 writes of “the futility of ligating the gastroduodenal or superior pancreaticoduodenal arteries to control bleeding from duodenal ulcer” because of the anastomoses of the inferior pancreaticoduodenal arteries.

Remember the topography of the duct of Santorini. It is located in the vicinity of the gastroduodenal artery or in the vicinity of the superior posterior pancreaticoduodenal artery when the latter originates high on the gastroduodenal. Do not dissect too deeply in this area. The gastroduodenal artery is an excellent landmark for avoiding injury to the duct of Santorini.

A typical duodenal ulcer will respond very well to proximal gastric vagotomy (recurrence rate 5-10 percent155). A postbulbar ulcer does not respond as well.

Neither an aberrant nor an accessory left hepatic artery from the left gastric should be ligated because these may be the only arterial blood vessels to the left lobe of the liver.

Be careful with three arteries – right gastroepiploic, left gastroepiploic, and middle colic – and three omenta – gastrocolic, greater omentum, and transverse mesocolon. The gastroepiploic vessels are within the bilaminar gastrocolic omentum (ligament). The middle colic artery is located within the transverse mesocolon traveling toward the transverse colon in front of the second and occasionally the proximal portion of the third part of the duodenum.

Retrocolic anastomosis is preferable to antecolic anastomosis for obese individuals undergoing a Billroth II gastrojejunostomy. By vision or palpation, locate the middle colic artery. A window may be created in the avascular area of the transverse mesocolon to the left of the middle colic vessels. Avoid the marginal artery of Drummond (see chapter on large intestine).

Diffuse hemorrhagic gastritis can sometimes be controlled by extragastric arterial ligation. Rittenhouse et al.156 ligated the right and left gastric arteries and the right and left gastroepiploic arteries with immediate cessation of bleeding. Among 29 patients of Richardson and Aust,157 there were no cases of necrosis of the stomach following ligation. Here the submucosal vascular plexus aids the surgeon in preventing ischemia while permitting reduction of overall blood pressure to the mucosa. Devascularization may be useful for avoiding gastrectomy in some patients.

Life-threatening upper intestinal tract bleeding may develop during chemotherapy or radiation therapy for gastric lymphoma. Kelessis et al.158 reported that devascularization of the involved part of the stomach is safe and effective.

From an anatomic standpoint, ischemic gastropathy is a very enigmatic disease. The vascular supply of the stomach is very rich. Its collateral circulation is excellent, as Brown and Derr151 demonstrated. Schlein,159 using cadaveric demonstration, was able to achieve complete filling of intramural gastric vessels with patency of only one major gastric artery. It is well known that an almost total gastric devascularization can be performed with impunity, as has been reported by Jacobsen,160 Richardson and Aust,152 and Isabella et al.161 Why then should we encounter ischemic gastropathy after subtotal or almost-total gastrectomy, superselective vagotomy, splenectomy, or other procedures? We do not know.

In a study of seven patients with ischemic gastropathy, Casey et al.162 pointed out that the condition is often masked as gastritis, gastric ulceration, or gastric atony. They emphasized that gastric ischemia should be considered in differential diagnosis, because early diagnosis and treatment may improve the survival rate.


Costantino163 and Witte et al.164 have reported upon ruptured gastric artery aneurysms.


It is well known that the most important site of collateral circulation of the gastric veins is located at the abdominal distal esophagus. This is where the left gastric vein (portal system) (Fig. 15-30) communicates with the azygos veins (caval system). When portal vein occlusion is present, esophageal varices form here.

Fig. 15-30.

General arrangement and drainage of veins of stomach. The short gastric and gastroepiploic veins drain into the splenic vein. (Modified from Hollinshead WH. Anatomy for Surgeons: Vol 2. New York: Hoeber-Harper, 1956; with permission.)

It is also well known that occlusion of the splenic vein (Fig. 15-31) will produce obvious extrinsic varicosities of the veins of the proximal gastric unit (short gastrics, left gastroepiploic). Also possible is occasional involvement of the right gastroepiploic vein, intrinsic gastric collaterals, and even duodenal varices. Salam and Warren165 stated that the endoscopist can mistake these varices for hemorrhagic gastritis.

Fig. 15-31.

Collateral circulation characteristic of isolated occlusion of splenic vein. (Modified from Salam AA, Warren WD. Anatomic basis of the surgical treatment of portal hypertension. Surg Clin North Am 1974;54:1247-1257; with permission.)


A great venous arch can develop between the left and right gastroepiploic veins during portal hypertension, forming a congested vascular bridge between the splenic and portal veins.

Salam and Warren165 effectively demonstrated the collateral circulation of the stomach with the occurrence of thrombosis of the splenic vein. They used their knowledge to develop the selective distal splenorenal shunt (Fig. 15-32), which, by the way, is not very popular today.

Fig. 15-32.

Selective distal splenorenal shunt. (Modified from Salam AA, Warren WD. Anatomic basis of the surgical treatment of portal hypertension. Surg Clin North Am 1974;54:1247-1257; with permission.)

Left Gastric (Coronary) Vein

The left and right gastric veins (Fig. 15-30) join along the lesser curvature of the stomach, forming there a venous “tiara,” or crown, the coronary vein. Either, or both of these veins end by draining into the splenic or portal veins. Although technically both the right and left gastric veins contribute to the coronary vein, the left gastric vein is very much the larger of the two vessels; thus, for the sake of simplicity and functional realism, the left gastric vein is often referred to as the coronary vein.

One can find the beginning of the left gastric vein at the middle of the lesser curvature. It runs close to the curvature to a point 2 to 3 cm from the inferior border of the esophageal hiatus where it receives one to three esophageal tributaries. From this point the left gastric vein turns downward and obliquely to the right to join the portal vein, or turns backward to enter the splenic vein. Bleeding from severed distal branches of the left gastric vein can be profuse because of the anastomoses between left gastric, esophageal, and hemiazygos veins.

In a study of 22 cadavers by Gray et al.,145 the left gastric vein entered the portal vein in 16 (73%) and entered the splenic vein in six (27%). Estimates in the literature for the termination of the left gastric vein in the portal vein range from 67%166 to 83%.167 In cases where the drainage is to the portal vein, the left gastric vein lies in the hepatogastric ligament, the proximal lesser omentum.

In the same study by Gray et al., a curved Kocher clamp applied very close to the lesser curvature at the hepatogastric omentum always included the following: left gastric artery, anterior and posterior nerves of the lesser curvature (nerves of Latarjet), hepatic division of the anterior vagal trunk, left aberrant hepatic artery (present in four cases), and left gastric vein (in all cases with portal vein drainage, and three cases with splenic vein drainage).

A regular, curved, or straight Kocher clamp applied to the hepatogastric omentum at a 45 degree angle in the same cadavers always included the left gastric vein regardless of its drainage, as well as all the structures listed above.

Right Gastric Vein

The small right gastric vein (Fig. 15-30) partially drains the distal gastric unit. It is a fellow traveler of the right gastric artery at the distal lesser curvature, passing from left to right and draining into the portal system. The prepyloric vein of Mayo is a tributary of the right gastric vein and is a landmark of the gastroduodenal junction.168

Statistics for the drainage of venous vessels in the area are not in total agreement, nor is there agreement on the exact sites of termination. Good examples of this are the sites of termination of the inferior mesenteric vein and the right and left gastric veins (Figs. 15-33, 15-34). To facilitate memorization, we rounded the relevant percentages (22-38%) to the number 30. Note that this is a memorization aid only, and not completely accurate.

Fig. 15-33.

Major variations in place of ending of right gastric vein (black). Diagonally shaded area represents neck of pancreas. (Modified from Douglass BE, Baggenstoss AH, Hollinshead WH. The anatomy of the portal vein and its tributaries. Surg Gynecol Obstet 1950;91:562; with permission.)

Fig. 15-34.

Major variations (indicated by black vessels) in the place of ending of left gastroepiploic vein (top row), left gastric vein (middle row), and short gastric veins (bottom row). Diagonally shaded area represents neck of pancreas. Not shown, the common ending of one of the several short gastrics in a splenic vein (about 50% of specimens), and the occasional (about 2%) ending of one in the left gastroepiploic. P, portal vein; S, splenic vein; SM, superior mesenteric vein. (Modified from Douglass BE, Baggenstoss AH, Hollinshead WH. The anatomy of the portal vein and its tributaries. Surg Gynecol Obstet 1950;91:562; with permission.)


The inferior mesenteric vein enters the splenic vein in 30%.

The inferior mesenteric vein enters the superior mesenteric vein in 30%.

The inferior mesenteric vein enters the splenomesenteric junction in 30%.

The right gastric vein enters the upper portal in 30%.

The right gastric vein enters the lower portal in 30%.

The right gastric vein enters the junction in 30%.

The left gastric vein enters the upper portal in 30%.

The left gastric vein enters the lower portal in 30%.

The left gastric vein enters the splenic vein in 30%.

Takayasu et al.169 noted that the right gastric vein emptied into the left portal vein in 1.5 percent of 200 patients, an anomaly that might be confused as a mass lesion radiologically. Such a communication might allow direct metastasis from the lesser curvature directly to the left lobe. The left gastric vein may drain in similar fashion.170

Right Gastroepiploic Vein

The right gastroepiploic vein drains the distal surgical unit and accompanies the right gastroepiploic artery. It receives the anterior superior pancreaticoduodenal vein and the middle colic vein in front of the pancreatic head. Its relationship to the uncinate process is well known to the pancreatic surgeon. The right gastroepiploic (Fig. 15-35) empties into the superior mesenteric vein in most cases.

Fig. 15-35.

Major variations in the place of ending of right gastroepiploic vein (black). P, Portal vein. S, Splenic vein. SM, Superior mesenteric vein. Diagonally shaded area represents neck of pancreas. (Modified from Douglass BE, Baggenstoss AH, Hollinshead WH. The anatomy of the portal vein and its tributaries. Surg Gynecol Obstet 1950;91:562; with permission.)

Voiglio et al.171 reported that the gastrocolic vein (Henle’s gastrocolic trunk) is present in 70% of the cases and it is formed by confluence of the gastroepiploic vein and right upper colic vein. The gastrocolic vein is short and is located beneath the root of the transverse mesocolon, traveling along the anterior surface of the head of the pancreas.

Left Gastroepiploic Vein

The left gastroepiploic vein drains into the splenic vein or into one of the terminal branches of the splenic vein, usually the lower terminal branches (Fig. 15-34). Because the ending of this vein is variable, ligation should be placed as close as possible to the hilum, as previously described in “Left Gastroepiploic Artery.”

Left Inferior Phrenic Vein

The left inferior phrenic vein (Fig. 15-36) may drain into the left superior suprarenal vein, the inferior vena cava, or both. In the first case, it does not approach the gastroesophageal junction. In subjects in which drainage is partly or wholly to the vena cava, the vessel passes in front of the esophageal hiatus and is vulnerable to possible injury from procedures in the hiatal area, especially anterior crural approximation.

Fig. 15-36.

The inferior surface of the diaphragm and the vascular structures close to the esophageal hiatus. (A), hepatic veins; (B), sternocostal triangles (of Morgagni); (C), inferior phrenic vein; (D), inferior phrenic artery; (E), renal artery and vein; (F), suprarenal vein; (G), superior mesenteric artery; (H), celiac artery; (I), median arcuate ligament; (J), esophageal hiatus. (Modified from Gray SW, Rowe JS Jr, Skandalakis JE. Surgical anatomy of the gastroesophageal junction. Am Surg 1979;45:575-587; with permission.)

Short Gastric Veins

The short gastric veins (Fig. 15-34) partially drain the proximal gastric unit. Within the gastrosplenic ligament, the great majority of these vessels drain into the splenic vein or one of its branches. According to Douglass et al.167 these veins empty directly into the upper part of the spleen.

The short gastric vessels should be ligated carefully. They retract easily and, together with the left gastroepiploic vein, can be associated with postoperative bleeding. Imbrication of this part of the great curvature is recommended.

Lymphatics of the Stomach

Classical View of Gastric Lymphatic Drainage

In 1941, Coller et al.172 outlined four drainage zones of the gastric lymphatics. His system has been followed by most writers.173,174 The four zones are illustrated in Figure 15-37A and described in Table 15-11. Most of the lymphatic drainage of the stomach finds its way to the celiac nodes.

Table 15-11. Lymphatic Drainage of the Stomach

Zone I (inferior gastric) Nodes around right gastroepiploic and gastroduodenal arteries to nodes around hepatic artery to celiac nodes
Zone II (splenic) Nodes around left gastroepiploic and short gastric arteries to pancreaticosplenic nodes to splenic artery nodes to celiac nodes
Zone III (superior gastric) Nodes around left gatric artery to celiac nodes
Zone IV (hepatic) Nodes around right gastric artery to celiac nodes

Source: Skandalakis JE, Gray SW, Rowe JS Jr. Anatomical Complications in General Surgery. New York: McGraw-Hill, 1983; with permission.

Fig. 15-37.

Lymphatic drainage of the stomach. A, The 4 zones as outlined by Coller.172 I, Inferior gastric; II, Splenic; III, Superior gastric; IV, Hepatic. B, Arrows indicate that most of the drainage finds its way to the celiac nodes. (B, Modified from Skandalakis JE, Gray SW, Rowe JS Jr. Anatomical Complications in General Surgery. New York: McGraw-Hill, 1983; with permission.)


Zone I (inferior gastric) drains into the subpyloric and omental nodes

Zone II (splenic) drains into the pancreaticosplenic nodes

Zone III (superior gastric) drains into the superior gastric nodes

Zone IV (hepatic) drains into the suprapyloric nodes

These tidy divisions (Figs. 15-37A & B) are misleading. Dye injected into the midanterior wall of the gastric body and into the pylorus spreads to lymph nodes on both curvatures.175 Because lymphatic drainage follows the arteries of the stomach or parallels gastric venous return,176 metastases are not confined to the nodes draining a single zone. Secondary drainage along the splenic and gastroduodenal arteries brings metastases from the greater curvature to the celiac nodes, “the collecting point for all four primary pathways draining the stomach.”177 Coller et al.172 concluded that all four zones must be resected to minimize metastatic spread.

Balfe et al.178 reported, after studying 200 CT scans, that the finding of rounded structures greater than 8 mm in the hepatogastric ligament is a reliable indicator of left gastric node involvement by carcinoma or lymphoma or of coronary venous dilatation.

Recent Lymphatic Studies

It is well known that an incidence of celiac node involvement of 10% occurs when cancer is located at the cervical and upper thoracic esophageal area.179 Guernsey and Knudsen180 found 44% involvement of celiac nodes with cancer of the middle one-third of the esophagus and up to the distal 10 cm.

Jaehne et al.181 reported that in 193 gastric resections, a total of 7,112 lymph nodes was found (approximately 37 per patient).

Smith et al.182 advocated radical lymph node dissection to improve prognosis. Radical dissection does not increase postoperative complications.

 Read an Editorial Comment

Iriyama et al.187 advised that extensive lymphadenectomy is not mandatory for patients with intramucosal carcinoma of the stomach of the protruded type (Borrmann’s type 1), since lymph nodes rarely become involved in this type of carcinoma of the stomach. However, Otsuji et al.188 reported that survival of patients with submucosal tumors was prolonged by extensive lymphadenectomy with gastrectomy.

There is a longstanding controversy about whether the lymphatics of the stomach anastomose with those of the duodenum. According to Hollinshead,168 Horton believed that there is not an anastomosis between the duodenal and gastric lymphatics, while Coller and colleagues believed the opposite.

Statistics vary regarding the percentage of duodenal lymphatic involvement with the stomach. However, from what we have read from a clinical standpoint, cancer of the stomach may involve the duodenum. This was expressed in percentages of total cases by Coller et al.,172 Zinninger and Collins,189 and Marvin190:

Coller et al. 26.4%
Zinninger and Collins 27%
Marvin 38%

An area in which care must be practiced is in making judgments based on the classification of lymph node metastasis from carcinoma of the stomach. The following was reported by de Manzoni et al.191:

Both anatomic location and number of node metastases are important predictors of survival in gastric cancer patients. Caution should be used when comparing series classified according to the new TNM [1997] with series coded according to the old TNM [1987], as the two classifications group patients in different ways. We believe that further studies enrolling a larger number of patients are necessary to determine if a combined classification based on the number of metastatic nodes in the different tiers could be useful.

Boku et al.192 studied 274 patients with primary cancer of the stomach. They reported the following:


In early gastric cancer (protruded type) of the lower one-third of the stomach, metastasis is to the lymph nodes near the lesion

If cancer with muscularis propria involvement occurs, distant lymph nodes were found to be involved with the cancer

Among the cases with lymph node metastasis, differentiated early gastric cancer had more lymph node involvement and wider extent of metastasis than undifferentiated cancers

The same authors advised complete removal of lymph nodes and fat in all types of cancer except carcinoma in situ, and with early differentiated carcinoma.

Shchepotin et al.193 reported that patients with locally recurrent gastric cancer benefit from re-excision. Patients receiving radiotherapy and chemotherapy tended toward improved survival.

A study by Jakl and colleagues194 found that the extent of gastric resection – total or proximal – did not influence the survival rate in patients with gastroesophageal cancer. However, the majority of investigators disagree.

Adachi et al.195 presented a simple classification of positive lymph nodes around the stomach.


Level I: Perigastric positive

Level II: Intermediate positive

Level III: Distant positive

The five year survival rate was as follows:


Level I: 67%

Level II: 35%

Level III: 26%

Namieno et al.196 reported that in 1137 cases of gastric carcinoma (single primary lesion) the overall incidence of metastasis was 9.5 percent. Mucosal lesions metastasized in 2.6 percent and submucosal in 16.5 percent with involvement of the perigastric nodes along the lesser and greater gastric curvatures.

Mak et al.197 reported that gastric mucosal lymphangiectasia is associated with gastric carcinoma; it signifies lymph node metastases and can spread via intramucosal lymphatics along the gastric mucosa.

Kodera et al.198 reported that extended lymphadenectomy for Borrmann type IV gastric carcinoma seems to improve outcome, but the survival rate remains unsatisfactory. In a later study, Kodera et al.199 stated that the number of metastatic regional nodes in carcinoma of the stomach has a strong prognostic significance.

According to Gabella,200 gastric lymphatics are continuous at the cardiac orifice with the esophageal vessels and at the pylorus with the duodenal vessels. On the whole, they follow blood vessels and form four groups (Fig. 15-38).


1. Left gastric lymph nodes: related to the pathway of left gastric artery and draining areas of both anterior and posterior gastric walls

2. Pancreaticosplenic nodes: draining gastric fundus and body and related to left gastroepiploic vessels and short gastric vessels

3. Right gastroepiploic nodes: related to the right gastroepiploic artery and drain the right half of the greater curvature, occasionally including the pylorus

4. Hepatic-pyloric-left gastric nodes: drain the pyloric part of the stomach

Fig. 15-38.

Lymphatic drainage of the lower esophagus, stomach, and duodenum.

It is extremely difficult to designate specific groups of lymph nodes to which lymph drains. The overall lymphatics of the human body are full of Delphian and Byzantine ambiguities.

From a surgicoanatomic standpoint, we present eight groups of lymph nodes of the stomach (Fig. 15-39):


Paracardial nodes

Left gastric nodes at the left gastric artery

Celiac nodes at the celiac artery

Suprapyloric nodes

Infrapyloric nodes

Right gastroepiploic nodes at the pathway of the right gastroepiploic artery

Pancreaticosplenic nodes at the pathway of the left gastroepiploic artery

Upper greater curvature nodes at the short gastric vessels

Fig. 15-39.

Eight groups of lymph nodes of the stomach.


All roads lead to Rome, and, in other words, to celiac axis lymph nodes.

Concept of Visalli and Grimes

A different view of gastric lymphatic drainage based on developmental patterns is presented by Visalli and Grimes.201,202 In this view, several groups of nodes, such as the gastroepiploic and pancreaticosplenic, are of secondary importance since their efferent vessels drain the celiac group, which Visalli and Grimes referred to as “the vortex of the metastatic whorl.”

The embryology of the stomach and related organs is such that the body and tail of the pancreas (derived from the dorsal pancreatic anlage), together with the spleen, lie in the dorsal mesogastrium. They share both a common blood supply (left gastric and splenic arteries) and a common lymphatic drainage with the proximal portion of the stomach.

The head of the pancreas (derived from the ventral pancreatic anlage) lies in the mesoduodenum. The pancreatic head shares its blood supply (pancreaticoduodenal and gastroduodenal arteries) and lymphatic drainage with the duodenum, the distal common bile duct, and the distal stomach.

Theoretically, cancer of the proximal stomach can be effectively treated by en bloc resection of the organs supplied by the left gastric and splenic arteries (Fig. 15-40A):


distal esophagus

proximal two-thirds of the stomach and greater omentum


body and tail of the pancreas

Fig. 15-40.

Plans for en bloc resections. A, Proximal stomach and related organs sharing a common blood supply and lymphatic drainage. B, Distal stomach and related organs. (Modified from Skandalakis JE, Gray SW, Rowe JS Jr. Anatomical Complications in General Surgery. New York: McGraw-Hill, 1983; with permission.)

Similarly, cancer of the distal stomach can be treated by en bloc resection of the organs supplied by the common hepatic artery (Fig. 15-40B), sparing, of course, the artery itself:


head of the pancreas

distal stomach and greater omentum


distal bile duct

Visalli and Grimes201 believed that the procedures just described will check metastatic spread more effectively than will extirpation of peripheral lymph nodes only. The reader is reminded of the high morbidity and mortality that may accompany such procedures.

However, Sakaguchi et al.203 stated that the poor prognosis of cancer of the proximal third of the stomach is due mainly to esophageal invasion. They advised early diagnosis, aggressive lymph node dissection, and possibly chemotherapy.

Ellis et al.204 reported that patients 70 years and older have comparable rates of palliation and survival compared to younger patients when aggressive surgery for the treatment of cancer of the esophagus and cardia is performed.

Practical Problems of Lymphatic Drainage

One must know the lymphatic drainage of any organ to plan a proper en bloc resection. However, a simplistic view based solely on lymphatic drainage cannot always be followed. In the stomach, as in other organs, the very presence of cancer can alter the normal lymphatic drainage. Obstructed vessels can divert the drainage so that metastases appear in unexpected nodes. Collateral lymphatics can form, producing a shift in the drainage pattern. Finally, the gastric submucosal lymphatic plexus is probably as rich as the vascular plexus, so that carcinoma can spread intramurally to all parts of the stomach and into the esophagus. Even though the submucosal lymphatics are not continuous across the pyloric sphincter, subserosal channels can carry metastases to the duodenum. The possibly multicentric origin of gastric cancer further complicates a simple approach to the proposed resection.

Lymphatics and Metastatic Cancer

The diversity of distribution of lymphatic vessels from the stomach wall was alluded to, somewhat unwittingly, by Virchow,205 who wrote that he “. . .repeatedly noted that in abdominal disease of unknown character, its cancerous nature is manifested in a remote but easily accessible place, namely the jugular lymph nodes.”

What Virchow stated 150 years ago is well known today. Cancer of the stomach may metastasize not only to the supraclavicular or scalene nodes206,207 but also to other areas such as the axillary region.208

In 1989, Kawaura209 advised en bloc resection of the lower esophagus, stomach, first portion of the duodenum, spleen, greater omentum, lesser omentum, and, if necessary, the distal pancreas and transverse colon in patients with gastric cancer involving the esophagus.

In 1995, Nishi210 reported 61.6 percent survival for curative resection.

The practicing surgeon today still faces the decision of whether to perform total or subtotal gastrectomy in patients with gastric carcinoma. What about carcinoma of the G-E junction? Is total gastrectomy and distal (10 cm) esophagectomy the answer? Most likely yes, from an embryologic and anatomic standpoint. How much lymphadenectomy? Are Roukos and Encke211 correct (in discussion of a paper about lymphoma by Tanaka et al.212) in stating that the extent of surgery is still controversial? The age of the patient, and problems with the heart, liver, pulmonary or renal systems are not necessarily the deciding factors for the given patient.

We must never forget what Lord Moynihan213 stated in 1908: “The surgery of the malignant disease is not the surgery of the organs; it is the anatomy of the lymphatic system.”

Jentschura et al.214 reported that in patients with early gastric cancer, there was no statistical significance in survival rates between those who had undergone subtotal gastric resection, total gastrectomy, and either a proximal or an atypical resection. Wanebo et al.215 stated that at the time of their research (1996) patients with carcinoma of the stomach did not show benefit from extragastric D2 node dissection.

In 1998, Namieno et al.216 advised limited surgery, such as wedge resection including all gastric layers, if the following criteria are present in patients with gastric carcinoma:


Carcinoma limited to mucosa

Macroscopically elevated tumor

Histologically well differentiated tumor

Tumor 10 mm

No ulcer or ulcer accompanying tumor

In another study, Namieno and colleagues217 found moderately differentiated carcinomas of the stomach unsuitable for surgery. These lesions would be classified histologically as diffuse.

Surgical Applications

We consider here surgical applications to the two surgical gastric units, proximal and distal.

Proximal Unit

Lymph from the proximal gastric surgical unit drains along the left gastric artery into the superior left gastric lymph nodes, which are located at the hepatogastric ligament. The proximal greater curvature drains to the pancreatosplenic nodes.

Kodera et al.218 commented on the definition of cancer of the gastric cardia and its bearing on surgical intervention:

A striking difference in the distribution of types of adenocarcinoma of the gastroesophageal junction was observed in Japan compared with previously reported western data. A subgroup of carcinoma of the proximal stomach identified as types II and III may not require proximal gastrectomy from the viewpoint of sufficient lymphadenectomy.

These authors referred to the 1996 classification of carcinoma of the cardia proposed by Siewert and Stein,219 Type I: lower esophagus (distal); Type II: cardia (“true carcinoma of the cardia”); Type III: upper stomach (subcardial) (most common). This classification was accepted in 1997 by the International Gastric Cancer Association Consensus Conference on Adenocarcinoma of the Esophagogastric Junction.

In a more recent publication, Siewert et al.220 elaborated on adenocarcinomas of the esophagogastric junction (AEG):

A complete removal of the primary tumor and its lymphatic drainage has to be the primary goal of any surgical approach to adenocarcinoma of the esophagogastric junction. In patients with a potentially resectable, true carcinoma of the cardia (AEG Type II), this can be achieved by a total gastrectomy with transhiatal resection of the distal esophagus and en bloc removal of the lymphatic drainage in the lower posterior mediastinum and along the celiac axis and superior border of the pancreas. This approach is associated with lower morbidity and provides equal long-term survival as compared with the more radical transmediastinal or abdominothoracic esophagogastrectomy.

Distal Unit

The distal gastric surgical unit drains to the suprapyloric nodes at the vicinity of the lateral part of the hepatogastric ligament, as well as to the hepatoduodenal ligament at the distal part of the lesser curvature.

We agree with Mulholland176 that all discrete anatomic lymphatic groupings are misleading because the gastric walls demonstrate rich intramural and extramural communication.

Specimen from Total Gastrectomy for Carcinoma of the Stomach

The ideal specimen from total gastrectomy for carcinoma of the stomach (Fig. 15-41) should include the following if the general condition of the patient permits:


Distal 8-10 cm of esophagus and periesophageal lymph nodes of lower mediastinum

Stomach (in toto)

All tissues and lymph nodes around G-E junction. Perform truncal vagotomy

Lesser omentum and its lymph nodes. Remove by


– Ligation of left gastric artery at its origin (protect celiac plexus by careful dissection)

– Ligation of coronary vein

– Skeletonization of hepatic artery and portal vein up to hepatic portas. Sweep all tissues (including left gastric and hepatic arterial lymph nodes) toward lesser curvature

Gastrocolic ligament and greater omentum

Two to three centimeters of first portion of duodenum including supra- and infrapyloric nodes

Distal pancreatectomy and splenectomy including splenopancreatic nodes

Fig. 15-41.

Highly diagrammatic representation of the ideal specimen of total gastrectomy. Some anatomic entities are not shown.

Of course total gastrectomy is a super-radical procedure with high morbidity and high mortality. Is it “ideal”? Will the patient be cured with this highly controversial procedure? The literature is full of controversies. The general surgeon’s own philosophy must serve for guidance.

Gastric Innervation

Two nerve plexuses, the submucosal or Meissner and the myenteric or Auerbach plexus, . . .represent “a brain within the gastrointestinal tract.”—Elder and Deakin221

Parasympathetic Innervation (Vagus Nerve)

Waisbren and Modlin222 wrote the following about Lester R. Dragstedt, the formulator of truncal vagotomy and one of the great surgical physiologists of our time.

The date October 22, 1993, marks the centenary of the birth of Lester R. Dragstedt. He emerged from humble roots of Swedish immigrant parents to become one of the pre-eminent surgical innovators of the twentieth century. Early in his scientific career, Dragstedt was profoundly influenced by another Swede, A.J. Carlson, who was initially employed as a Lutheran minister in Dragstedt’s hometown of Anaconda, Montana. Carlson left the ministry for graduate school and later became chairman of The Department of Physiology at the University of Chicago. When Dragstedt finished his schooling, Carlson convinced him to attend the University of Chicago. In addition to Carlson, Dragstedt’s research was influenced by many prominent physiologists and surgeons, including Pavlov and Latarjet. Their work, along with his own investigations, helped him both to formulate his hypotheses on the regulation of gastric acid secretion and to formalize the operation of truncal vagotomy. In 1943, Dragstedt initiated the clinical use of this procedure in North America. Although he studied his patients carefully and documented his results meticulously, the operation initially met with considerable resistance from both his medical and surgical colleagues. Over time, many other surgeons accepted vagotomy as a viable procedure and further modified his technique. The unique ability of Dragstedt to transfer his research studies to the development and implementation of rational surgical therapy remains an enduring example for the surgical profession.

Today we would say that vagotomy is totally irrational!

The left and right vagus nerves (Fig. 15-42) descend parallel with the esophagus and contribute to a rich external esophageal nerve plexus between the level of the tracheal bifurcation and the level of the diaphragm. From this plexus, two vagal trunks, anterior and posterior, form and pass through the esophageal hiatus of the diaphragm. Each trunk subsequently separates into two divisions.

Fig. 15-42.

The terminology of vagal structures of the thorax and abdomen. In this example two vagal trunks pass through the hiatus to enter the abdomen. (Modified from Skandalakis JE, Rowe JS, Gray SW, Androulakis JA. Identification of vagal structures at the esophageal hiatus. Surgery 1974;75:233-237; with permission.)

From the anterior vagal trunk, the hepatic division passes to the right in the lesser omentum, branching before it enters the liver. One branch turns downward to reach the pylorus and, sometimes, the first part of the duodenum. The second division, the anterior gastric, descends along the lesser curvature of the stomach, giving branches to the anterior gastric wall.

The celiac division and the posterior gastric division arise from the posterior trunk. The celiac division passes through the celiac plexus. The posterior gastric division supplies branches to the posterior gastric wall.

Debas29 reported that more than 80% of the fibers in the vagi are afferent neurons carrying information back to the brain.

Identification of Vagal Structures at the Hiatus

The pattern of the vagus nerves at the esophageal hiatus is important to the surgeon planning a vagotomy. The basic configuration and variations are well known.41,223,224 The thoracic pattern is not visible to the abdominal surgeon, who must therefore proceed on the basis of the structures that can be seen.

In a study of components of the esophageal hiatus in 100 cadavers, Skandalakis et al.71 found the following:


Two vagal structures only (Fig. 15-42): 88%. The usual structures at the esophageal hiatus are the anterior and posterior vagal trunks, which have not yet split to form the four typical divisions discussed above. Both trunks are usually to the right of the midline of the esophagus. The posterior trunk lies closer to the aorta than to the esophagus (Fig. 15-43).

Four vagal structures (Fig. 15-44A): 7%. The four divisions of the vagal trunks (hepatic, celiac, anterior gastric, and posterior gastric) appear when division has occurred above the diaphragm.

More than four structures (Fig. 15-44B, C): 5%. When there are more than four structures at the hiatus, these may be divisions and branches of divisions (Fig. 15-44B) (the anterior and posterior trunks lie entirely within the thorax) or elements of the esophageal vagal plexus (Fig. 15-44C) (the anterior and posterior trunks lie entirely within the abdomen).

Fig. 15-43.

Relation of anterior and posterior vagal trunks to aorta and esophagus in 88 specimens. Trunks are usually to right of midline (anterior, 80 [91%]; posterior, 76 [86%]). Anterior trunks are closer to esophagus than are posterior trunks. (Modified from Skandalakis JE, Rowe JS, Gray SW, Androulakis JA. Identification of vagal structures at the esophageal hiatus. Surgery 1974; 75: 233-237; with permission.)

Fig. 15-44.

Four or more vagal structures emerge through hiatus. A, Divisions that have separated just above diaphragm. B, Divisions and their branches that arise above diaphragm. C, Elements of esophageal plexus that extend below diaphragm. Vagal trunks entirely within abdomen. 1, hepatic division; 2, anterior gastric division; 3, celiac division; 4, posterior gastric division. (Modified from Skandalakis JE, Rowe JS Jr, Gray SW. Identification of vagal structures at the esophageal hiatus. Surgery 1974;75:233-237; with permission.)

Nerve structures at the hiatus in addition to the usual two have sometimes been called accessory.225 This term is incorrect. True accessory nerves would have to originate from the esophageal plexus and descend without making further connection with the normal vagal trunks. In our experience, all such parallel descending fibers rejoin the normal trunks below the diaphragm. This shows them to be part of the esophageal plexus.

Branches of the division, especially those of the posterior gastric division, are inconstant. Where they originate above the diaphragm, there may be several branches descending through the hiatus. Indeed, among Jackson’s41 50 specimens, there were only two in which the posterior gastric division originated as a single bundle. Such branches of divisions are not accessory.

Distribution of the Vagal Nerves to the Stomach

Anterior Gastric Division

Among 100 specimens dissected in a study by Skandalakis et al.,70 the separation of the anterior gastric and hepatic divisions occasionally occurred above the diaphragm, but usually lay on the abdominal esophagus or the cardia.

In 96 of those 100 specimens, a major branch of the anterior gastric division formed the principal anterior nerve of the lesser curvature (anterior nerve of Latarjet). It usually lay from 0.5 to 1.0 cm from the lesser curvature. In one of the specimens it lay beneath the serosa of the gastric wall. This nerve can be traced distally to about the level of the incisura in most subjects, but in many it reaches the pylorus and in a few it is visible as far as the first part of the duodenum (Fig. 15-45A and Table 15-12).

Table 15-12. Visible Distal Extent of the Anterior and Posterior Gastric Divisions of the Vagus

Termination Number of anterior divisions Number of posterior divisions
Above the incisura 6 21
At the incisura 17 4
At or above the incisura 62 71
Just below the incisura 6 1
At the pylorus 3 3
At the first part of duodenum 6 0
Total 100 100

Source: Skandalakis JE, Gray SW, Rowe JS Jr. Anatomical Complications in General Surgery. New York: McGraw-Hill, 1983; with permission.

Fig. 15-45.

Distribution of anterior gastric and hepatic divisions of vagus. A, “Typical” distribution. B, Duplication of anterior gastric division (nerve of Latarjet). C, Separation of anterior gastric division into superior and inferior portions (specimen with carcinoma). D, Antral innervation from pyloric branch of hepatic division. E, Multiple nerves of hepatic division each contribute to descending pyloric branch. (Modified from Skandalakis JE, Gray SW, Soria RE, Sorg JL, Rowe JS Jr. Distribution of the vagus nerve to the stomach. Am Surg 1980;46:130-139; with permission.)

Legros and Griffith226 demonstrated that vagal branches that are subserosal at the surface may penetrate the muscularis and continue downward to the antrum by the submucosal (Meissner’s) plexus. From two to twelve branches pass from the principal nerve to the stomach wall. The average in the subjects in the study by Skandalakis et al.70 was six. In two of those subjects, the anterior nerve of Latarjet was duplicated; each nerve supplied its own branches to the stomach wall (Fig. 15-45B). Loeweneck227 has called the longer of these nerves the “antral nerve.”

In some subjects, there is no true nerve of Latarjet; a fan of gastric branches arises from the anterior vagal trunk above the origin of the hepatic division, and one or more long branches below the origin descend to supply the antrum (Fig. 15-45C). Even where a definite nerve of Latarjet is present, there are usually some, and often many, branches to the gastric cardia and fundus that arise from the anterior trunk proximal to the origin of the hepatic division (Fig. 15-45D).

Occasionally the hepatic division is formed by multiple nerves, each participating in the formation of the pyloric branch (Fig. 15-45E).

While the nerve of Latarjet is often seen branching in the “crow’s foot” formation (Fig. 15-46), this pattern is far from constant, being equivocal in some cases and absent in many. The term “crow’s foot” was originally applied to the termination of the left gastric artery at the same location by Payne.228

Fig. 15-46.

Vagotomy (nerves to be preserved are in black). A, Truncal vagotomy. B, Selective vagotomy. C, Parietal cell or proximal gastric vagotomy. (Modified from Skandalakis LJ, Gray SW, Skandalakis JE. The history and surgical anatomy of the vagus nerve. Surg Gynecol Obstet 1986;162:75-85; with permission.)

Hepatic Division

The hepatic division of the anterior vagal trunk usually separates from the anterior gastric division at the level of the abdominal esophagus (Fig. 15-46). It lies between the leaflets of the avascular portion of the hepatogastric ligament. It is frequently found in multiple, usually closely parallel branches (Table 15-13).

Table 15-13. Variations in Number and Position of Elements of the Hepatic Division of the Vagus

Pattern Number
  In normal position 71
  From distal lesser curvature 1
  Contributions from both vagal trunks 5
  In normal position 3
  Both in normal position 2
  One from middle of lesser curvature 2
  All in normal position 3
  One from middle of lesser curvature 1
  All in normal position 12
Total 100

Source: Skandalakis JE, Gray SW, Rowe JS Jr. Anatomical Complications in General Surgery. New York: McGraw-Hill, 1983; with permission.

Posterior Gastric Division

In most subjects, the posterior gastric division (Fig. 15-47) forms the principal posterior nerve of the lesser curvature (posterior nerve of Latarjet). As a rule, the posterior nerve appears to terminate slightly higher on the lesser curvature and possesses fewer gastric branches than the anterior nerve. In no case has a posterior nerve been observed to reach the duodenum.

Fig. 15-47.

Distribution of posterior gastric and celiac divisions of vagus. A, Usual pattern. Nerve of Latarjet springs from celiac division. Most cranial branch from posterior trunk is “criminal nerve” of Grassi. B, Absence of posterior nerve of Latarjet. Pylorus well innervated posteriorly. (Modified from Skandalakis JE, Gray SW, Soria RE, Sorg JL, Rowe JS Jr. Distribution of the vagus nerve to the stomach. Am Surg 1980;46:130-139; with permission.)

In a number of specimens, the gastric branches of the posterior nerve of Latarjet fell into a superior and inferior group. The superior branches arose from the posterior vagal trunk just below or even above the diaphragm. The inferior branches arose from the descending posterior nerve of Latarjet and supplied only the lower body and the antrum. Between these two groups of branches, the lesser curvature had no grossly visible nerve supply.

In many specimens in the study by Skandalakis et al.,70 the most superior gastric branch, the “criminal nerve” of Grassi,229 arose at or above the origin of the celiac division (Fig. 15-47A). Usually no single branch seemed worthy of this dramatic term. Such early rising branches, which can easily escape casual detection, can arise singly or multiply from either of the vagal trunks.

In 18 of 100 specimens in the study by Skandalakis et al.,70 a true posterior nerve of Latarjet was absent (Fig. 15-47B); branches arising from the celiac division turned back toward the lesser curvature, and one or more long branches descended to innervate the antrum.

Celiac Division

The celiac division is the largest of the four vagal divisions. It lies in the gastropancreatic peritoneal fold. In all cases it is single and leads directly to the celiac plexus. The celiac division may follow the left gastric artery or the right crus of the diaphragm or take an intermediate position in the triangle bounded by the artery, the crus, and the right margin of the stomach.

Vagal parasympathetic fibers connect the stomach with the brainstem or terminate in the gastric mesenteric plexus. We emphasize again that according to Debas,29 over 80% of the fibers in the vagi are afferent neurons carrying information back to the brain.

The various techniques used by investigators in sectioning the vagus nerves from 1814 to 1979 are shown schematically in Figure 15-48.

Fig. 15-48.

Various techniques employed by different investigators in sectioning of vagus nerves from 1814 to 1979. (Modified from Waisbren SJ, Modlin IM. Lester R. Dragstedt and his role in the evolution of therapeutic vagotomy in the United States. Am J Surg 167:344-359, 1994; with permission.)

Vagaries of the Vagus

Do we know the anatomy of the parasympathetic and sympathetic innervation of the stomach? There is much yet to be revealed to us about the anatomy of this complex system and even more concerning its neurophysiology.

When Skandalakis et al.70-72 studied the exogastric anatomy of the vagus nerve many years ago, they never thought about other possible sites of efferent nerves of vagal origin entering the stomach. However, the possibility is there, and Donahue of the University of Illinois demonstrated this in several publications.230-233 Are there more vagaries of the vagus nerve? Most likely, yes.

Donahue believes that the possible sites of preganglionic efferents are as follows:


1. Esophageal plexus (periesophageal region)

2. Lesser curvature (gastric corpus anterior and posterior)

3. Heel of crow’s foot

4. Gastropancreatic fold

5. Short gastric arteries

6. Left gastroepiploic artery

7. Right gastroepiploic artery

These seven sites form the “checkerboard” of Donahue (Fig. 15-49). Preganglionic efferent vagus nerves reach the parietal cell mass in these areas. Areas 3, 4, 6, and 7 are divided routinely during extended highly selective vagotomy. Area 5 is preserved because the nerves at this site cannot be divided without sacrificing essential blood supply to the proximal part of the stomach.

Fig. 15-49.

The seven areas of vagotomy. 1, Esophageal plexus (periesophageal region). 2, Lesser curvature (gastric corpus anterior and posterior). 3, Crow’s foot area. 4, Gastropancreatic fold. 5, Short gastric arteries. 6, Left gastroepiploic pedicle. 7, Right gastroepiploic pedicle. (Modified from Donahue PE, Richter HM, Liu KJM, Anan K, Nyhus LM. Experimental basis and clinical application of extended highly selective vagotomy for duodenal ulcer. Surg Gynecol Obstet 1993;176:39-48; with permission.)

The accepted technique of proximal gastric vagotomy is by “parasympathetic denervation” of the proximal two-thirds of the stomach, preserving the antral and pyloric innervation as well as the hepatic and celiac divisions. This “denervation” is accomplished by sectioning gastric branches of the anterior and posterior nerves of Latarjet from the gastroesophageal junction (abdominal esophagus) to the crow’s foot.

Sympathetic Innervation

The sympathetic chains, the thoracic splanchnic nerves containing afferent and efferent fibers, and the celiac ganglia form the basic elements for the sympathetic innervation of the stomach and duodenum.


The thoracic splanchnic nerves are formed mostly of preganglionic fibers from the intermediolateral cell column of the spinal cord at levels T5-T10. These terminate within the celiac ganglia, where preganglionic fibers synapse upon the collateral ganglion cells. Postganglionic fibers from the celiac ganglia reach the stomach and duodenum via their blood supply.


Afferent fibers for the sense of pain from the organs supplied by the celiac artery (including the stomach) pass through the celiac plexus and thoracic splanchnic nerves to the sympathetic chains. They reach spinal nerves at thoracic levels T5-T10. The cell bodies of these fibers are found in dorsal root ganglia at those levels.

Therefore, the sympathetic nerve pathway is the connection of the stomach to the spinal cord via sympathetic neurons and dorsal root ganglia. Efferent fibers start in the celiac ganglia and end in the gastric wall as target cells. Afferent fibers whose peripheral processes originate in the gastric wall have cell bodies in the dorsal root ganglia, with central processes entering the spinal cord.

Basing his conclusion on experimentation, Bruckner234 stated that the sympathetic nervous system “exerts a depressing influence on gastric acid and pepsin secretion and also on motility.”

The vagus nerve inhibits the motility of the antrum. The submucosal plexus of Meissner together with the myenteric plexus of Auerbach participate in normal antral motility.

Surgery of the Stomach

He who combines the knowledge of physiology and surgery, in addition to the artistic side of his subject, reaches the highest ideal in medicine. —Billroth235

It is not within the scope of this part of the chapter to repeat the history of gastric surgery or to give step-by-step details of surgical technique for the many procedures of the stomach. However, we present a diagrammatic review of gastric operations which will be useful to students of anatomy and surgery, as well as to their teachers.

There is no originality here. Kudos belongs to Waugh and Hood, Wangensteen, Bünte, Laughan, Landor, Nyhus, Wastell, and so many others. We have only collected the colorful “pebbles on the shore of the boundless ocean. . .,”236 which were spread by the pioneers of gastric surgery to whom we owe so much. It has been said that any time Boerhaave mentioned the name of Sydenham he removed his hat. The surgeons of today must stand aright and with awe (to be somewhat biblical) anytime we talk about the great events and the great thoughts that produced the triumphant surgery of today.

The history of gastric surgery is beautiful and heroic. Here we present simple line drawings in chronological order with minimal, if any, explanation. Many procedures have a great number of modifications; we apologize that we cannot include them all.


One can probably argue successfully that the first case of gastrorrhaphy was either that of Nolleson le fils in 1767 or that of Croll in 1602 (reported in 1609) when both gastrotomy and gastrorrhaphy were performed for the removal of a foreign body (Fig. 15-50).

Fig. 15-50.


Gastrorrhaphy is performed for traumatic laceration of any part of the stomach. Debridement often is necessary. The gastric wall is sutured in two layers after all bleeding points are ligated.


Gastrotomy is performed for the removal of foreign bodies, mucosal biopsies, etc.

Gastrostomy, Duodenostomy, Duodenoduodenostomy

Gastrostomy (Fig. 15-51) is performed for pathological entities such as acute gastric dilatation, carcinoma of the esophagus or stomach for elemental diets, gastrointestinal decompression to avoid nasogastric tube, etc.

Fig. 15-51.

Gastrostomy (Witzel, Stamm, Kadar, Marwedel, Tavel, Beck-Carrell, Jianu, Janeway, Spivack, Patton). (First eight figures after Waugh JM, Hood RT Jr. Gastric operations: a historic review. Part 1. Q Rev Surg Obstet Gynecol 1953 Dec; 10(4):201-214, Fig. 1. Last illustration [Patton] after Nyhus LM, Wastell C. Surgery of the Stomach and Duodenum, 4th ed. Boston: Little, Brown, 1986, fig. 30-9D.)

From the time of Sédillot (mid 1800s) to that of more recent surgeons such as Sacks and Preshaw, several methods of gastrostomy have been presented in the literature. We present some as presented by Waugh and Hood in their classic publication,237 and we add a few more from the literature. Percutaneous gastrostomy has been supported by several recent authors.

Catheter duodenostomy (Fig. 15-52) is used to prevent “blowout” of the duodenal stump.

Fig. 15-52.

Catheter duodenostomy for doubtful or inadequate duodenal stump closure. A, Catheter should be at least No. 16F. If available, a large omental tag should be arranged to seal tract around catheter. B, Tube brought out of abdomen through stab wound separate from wound is used to drain stump closure. Note: Tube enters side of the duodenum separately from duodenal closure. (After Nyhus LM, Wastell C. Surgery of the Stomach and Duodenum, 4th ed. Boston: Little, Brown, 1986, Fig. 30-12.)

Duodenoduodenostomy is the procedure of choice for duodenal atresia and stenosis (Fig. 15-53).

Fig. 15-53.

Duodenoduodenostomy. (After Nyhus LM, Wastell C. Surgery of the Stomach and Duodenum, 4th ed. Boston: Little, Brown, 1986, Fig. 27-13.)


Leakage and peritonitis may be avoided by:


Pursestring suture around the catheter

Fixation of the gastric wall to the anterior abdominal wall

Use the same technique for duodenostomy as for gastrostomy

Use of a stab wound for the tube


Gastropexy is the fixation of the anterior wall of the stomach in the area of the lower curvature to the anterior abdominal wall. This procedure is used for the surgical treatment of gastric volvulus, although a laparoscopic approach now has adherents.238 Sometimes gastropexy is used in combination with a Stamm gastrostomy.

For all practical purposes, gastropexy as a gastric procedure for the treatment of gastroptosis is no longer performed. Perhaps gastropexy should be considered the fixation of the gastrostomy to the abdominal wall. One of the authors of this chapter (JES) has performed gastropexy in two cases of gastric volvulus.

Posterior gastropexy is the Hill procedure and its modification. In posterior gastropexy, the posterior phrenoesophageal bundle is anchored to the median arcuate ligament.


Several procedures have been developed for the treatment of cardiospasm or megaesophagus (Fig. 15-54).

Fig. 15-54.

Operations for cardiospasm. Marwedel and Wendel procedures use a longitudinal incision at the gastroesophageal junction, involving all layers, and with a transverse closure. The Heyrovsky variation includes separate incisions of the distal esophagus and the proximal stomach, and an esophagogastroanastomosis. Heller consists of a longitudinal incision of serosa and muscular layers, keeping the esophagogastric mucosa intact. It is closed transversely. The Backer-Gröndahl procedure is the same as the Heyrovsky, but the incisions are united. The Wangensteen variation is an esophagogastrectomy (distal esophagus-proximal stomach) and end-to-end anastomosis and Rammstedt pyloromyotomy (see Fig. 15-55). (After Waugh JM, Hood RT Jr. Gastric operations: a historic review. Part 2. Q Rev Surg Obstet Gynecol 1954 March; 11(1):1-18, Fig. 11.)

Fig. 15-55.

Pyloroplasty. The Heineke-Mikulicz procedures consist of incising the gastroduodenal wall longitudinally through all layers and closing the defect transversely. The Jaboulay variation has separate longitudinal incisions involving the gastric and duodenal walls and a side-to-side anastomosis. Finney consists of a horseshoelike gastroduodenal incision (distal part of stomach and first and second parts of duodenum) with a transverse closure. Defour and Fredet modified the Heineke-Mikulicz procedure for infants. It consisted of a longitudinal incision of the serosa and muscular layers, leaving the mucosa intact, and closing transversely. A procedure identical to that of Defour and Fredet was described two years later by Weber. The Rammstedt procedure is essentially a Defour-Fredet operation except that he did not close the gastroduodenal wall, leaving the mucosa uncovered. The Horsley variation is a Heineke-Mikulicz pyloroplasty with excision of small ulcers from the anterior duodenal wall. The defect was closed transversely. Judd and Judd-Nagel procedures consisted of excision of an elliptical segment of the duodenal wall including the ulcer, followed by transverse closure. (After Waugh JM, Hood RT Jr. Gastric operations: a historic review. Part 1. Q Rev Surg Obstet Gynecol 1953 Dec; 10(4):201-214, Fig. 2.)

Bleeding Peptic Ulcer

It is advisable to remove the gastric ulcer to rule out malignancy by frozen section. Close the defect in two layers.

Closure of Perforated Peptic Ulcer

For gastric ulcer, see “Bleeding Peptic Ulcer” above.

For peptic ulcer, closure is by direct suture or use of any pedunculated piece of fat. Sharma et al.239 favor the use of the free omental plug as a safe and expeditious procedure applicable in emergency situations.

Blomgren240 reported that with its low rate of serious recurrences (14%), simple closure of perforated peptic ulcer is the procedure of choice in the elderly. However, Wysocki et al.241 favor gastric resection to oversewing of the perforation because of high mortality in suture patients.


As a drainage procedure, pyloroplasty (Fig. 15-55) is performed for partial or complete pyloric obstruction secondary to peptic ulcer disease with or without vagotomy. It is also used to complement surgery of the proximal gastric unit (resection, hiatal hernia, etc.).

Pyloroplasty is used as a drainage procedure with or without vagotomy for the treatment of obstructed or unobstructed peptic ulcer disease.


Pyloromyotomy (Fig. 15-56) is the procedure of choice for infants with hypertrophic pyloric stenosis.

Fig. 15-56.

Anatomy of pyloromyotomy. (Modified from Skandalakis JE, Skandalakis PN, Skandalakis LJ. Surgical Anatomy and Technique: A Pocket Manual (2nd ed). New York: Springer-Verlag, 2000; with permission.)


Gastrojejunostomy (Fig. 15-57) is a drainage procedure. For all practical purposes, it has the same indications as pyloroplasty. Gastrojejunostomy can be performed with or without vagotomy. It can be posterior or anterior, depending on where the jejunal loop is anastomosed to the wall of the stomach. It can be antecolic or retrocolic, isoperistaltic or anisoperistaltic (antiperistaltic) (Figs. 15-58, 15-59, and 15-60).

Fig. 15-57.

Gastrojejunostomy (Wölfler, Courvoisier, von Hacker, Jaboulay, Braun, Doyen, Roux, Mayo, Lahey). (After Waugh JM, Hood RT Jr. Gastric operations: a historic review. Part 1. Q Rev Surg Obstet Gynecol 1953 Dec; 10(4):201-214, Fig. 3.)

Fig. 15-58.

Various drainage procedures. A, Heineke-Mikulicz pyloroplasty. B, Finney pyloroplasty. C, Exclusion pyloroplasty. D, Posterior gastroenterostomy. E, Anterior juxtapyloric gastroenterostomy. F, Pyloric dilatation by gastrotomy. In each, note the associated vagotomy. (After Nyhus LM, Wastell C. Surgery of the Stomach and Duodenum, 4th ed. Boston: Little, Brown, 1986, Fig. 1-1.)

Fig. 15-59.

Antecolic gastrojejunostomy. A, Antiperistaltic anastomosis. B, Isoperistaltic anastomosis. (After Nyhus LM, Wastell C. Surgery of the Stomach and Duodenum, 4th ed. Boston: Little, Brown, 1986, Fig. 20-25.)

Fig. 15-60.

Posterior gastrojejunostomy. A, Antiperistaltic anastomosis. B, Isoperistaltic anastomosis. (After Nyhus LM, Wastell C. Surgery of the Stomach and Duodenum, 4th ed. Boston: Little, Brown, 1986, Fig. 20-26.)


Surgical Anatomy of Gastric Mobilization (Including Abdominal Esophagus and

Proximal Duodenum)

We present here the anatomy of the gastroesophageal area, stomach, and gastroduodenal area.

Gastroesophageal Area

(Fig. 15-61) The periesophageal space of the abdominal esophagus and the cardia of the stomach was first described in 1981 by Donahue and Nyhus242 of the University of Illinois. The periesophageal space has the following boundaries:


Anterior: Peritoneal reflection

Posterior: Aorta

Right: Hepatogastric ligament (envelops segment of abdominal esophagus)

Left: Proximal part of gastrosplenic ligament

Fig. 15-61.

Exposure of periesophageal space. A, Incision. B, Clean out fat. C, Finger is used to identify abdominal esophagus. D, The “stout” or dorsal mesoesophagus is disrupted by digital pressure and final perforation. If the mesoesophagus is “very stout,” divide between clamps with direct vision, avoiding the posterior right esophageal wall. E, The esophagus is retracted. At least 5 cm of the distal esophagus should be mobilized. (Modified from Donahue PE, Nyhus LM. Exposure of the periesophageal space. Surg Gynecol Obstet 1981;152: 218-220; with permission.)

Very slowly and with extreme care incise the left triangular and left coronary ligaments toward the gastroesophageal junction. Stop at the halfway point. Using finger dissection, continue the separation of the lateral segment of the left lobe of the liver from the diaphragm. It is important to continue in this manner because the left hepatic vein (which is very close to the falciform ligament) is vulnerable, and any tear may incorporate the IVC. Such an injury will result in several units of blood loss within a few seconds.

Slowly and carefully, partially separate the medial segment of the left lobe from the diaphragm. The left lobe of the liver may be displaced downward, inward, and to the right. Finger perforation of the avascular peritoneal reflection permits the finger to approach the abdominal esophagus; the esophagus can be tractioned down with a Penrose drain.

Further mobilization of the cardia and fundus is accomplished by careful dissection of the proximal lesser and greater curvatures. Ligate branches of the left gastric and short gastric vessels.

Remember the anatomic entities present in this area: right and left vagus trunk; mediastinal pleurae; left hepatic vein; small arteries, veins, and bile ducts associated with the left coronary ligament; and the accessory or replaced left hepatic artery.

The term “accessory” indicates that there is an additional source of vascular supply other than that which is most common and most typical. A “replaced” hepatic artery, according to Healey and Schwartz,243 is one that arises from a source other than the celiac axis and apparently supplies the entire liver or an entire lobe of the liver. Hemingway and Gibson244 offer this definition: “an artery which takes the place of the normal artery is called a replaced artery.”


The form of mobilization of the stomach depends on whether the procedure is tailored for benign or malignant lesions.

Benign Disease

To mobilize the stomach for benign disease, the surgeon will most likely perform a proximal or distal gastrectomy or a sleeve resection. With benign disease, we prefer to ligate the gastric branches inside the gastroepiploic arcade corresponding to the part of the stomach that will be resected.

We do not have any objection to ligating the vessels outside the gastroepiploic arcade by dividing the greater omentum, leaving 3-5 cm of omentum attached to the greater curvature and to the arcade,245 but at the same time ligating and dividing the left gastroesophageal artery. Griffith246 advised ligation of the anterior epiploic arteries by transection of the gastrocolic ligament. Be kind and gentle to the spleen. Avoid omental, gastric, or splenic traction. Avoid splenectomy by carefully mobilizing the splenic flexure (avoid traction). Carefully ligate the small peritoneal attachment to the spleen. This protects the splenic capsule.

Ligate the right gastroepiploic vessels now by dissection of the distal greater curvatures, or ligate later after duodenal mobilization. Enter the lesser sac. With the hand between the transverse mesocolon and the posterior wall of the stomach, palpate the transparent hepatogastric omentum and ligate the left gastric vessels. First ligate the coronary vein located above the artery and then the left gastric artery close to the celiac axis. Then ligate the right gastric vessels by dissection of the distal lesser curvatures, or ligate later after duodenal mobilization.

Malignant Disease

When mobilizing the stomach for malignant disease, remove the omentum by dividing and ligating the omentum from the transverse colon. Be sure to enter the gastrocolic ligament carefully to avoid injury to the middle colic artery of the transverse mesocolon, which is sometimes fixed with the gastrocolic omentum.

To avoid injury to the hepatic triad or replaced hepatic vessels, ligate the lesser omentum away from the lesser curvature. Remove fat, lymphatics, and lymph nodes in continuity with the specimen, if possible. Perform both gastrectomy (proximal, distal, or total) and lymphadenectomy.

Ohgami et al.247 reported that laparoscopic surgery for early gastric cancer in carefully selected patients is curative.

Hayes et al.248 reported that total gastrectomy with extended lymphadenectomy for “curable” gastric cancer carries high operative morbidity and increased mortality and therefore novel approaches should be explored.

Eguchi et al.249 stated that early diagnosis has improved the prognosis of young patients with gastric cancer.

Rodriguez-Sanjuán et al.250 reported that most patients with primary lymphomas of the stomach host Helicobacter pylori and that serosal involvement and high-grade tumors adversely influence survival but that gastrectomy type, resection margin invasion, and chemotherapy did not appear to have any influence.

Gastroduodenal Area

The first (ascending) part of the duodenum is 5 cm long. The proximal half is mobile; the distal half is fixed. The initial 2.5 cm is freely movable. It is covered by the same two layers of peritoneum that invest the stomach. The hepatoduodenal portion of the lesser omentum attaches to the superior border of the duodenum; the greater omentum attaches to its inferior border.

The distal 2.5 cm of the ascending duodenum is covered with peritoneum only on the anterior surface of the organ. The posterior surface is in intimate contact with the bile duct, portal vein, and gastroduodenal artery. The duodenum is separated from the inferior vena cava by a small amount of connective tissue.

The first part of the duodenum is supplied by the supraduodenal artery and the posterior superior pancreaticoduodenal branch of the gastroduodenal artery (retroduodenal artery of Edwards, Michels, and Wilkie), which is a branch of the common hepatic artery. In many individuals, the upper part of the first 1 cm is also supplied by branches of the right gastric artery. After giving origin to posterior superior pancreaticoduodenal and supraduodenal branches, the gastroduodenal artery descends between the first part of the duodenum and the head of the pancreas. It terminates by dividing into the right gastroepiploic and anterior superior pancreaticoduodenal arteries, both supplying twigs to this part of the duodenum.

The authors are uncertain about the most common pattern of blood supply to the first part of the duodenum. Good technique, conservative skeletonization, and good anatomic knowledge will produce positive results with surgical procedures in this area.

More information on the gastroduodenal area is found in the section on the duodenum in the chapter on the small intestine.

Surgical Anatomy of Gastric Resection

The surgical anatomy of gastric resection is presented in Table 15-14. We are grateful to Prof. Volker Schumpelick251 for advising us on the history of surgery.

Table 15-14. Surgical Anatomy of Gastric Resection

Location of Carcinoma Procedure Reason
Upper ⅓ Total gastrectomy: En bloc specimen should include 1) distal esophagus up to tracheal bifurcation, 2) 1-2 cm of duodenum, 3) distal pancreas, 4) spleen, 5) ligation of left gastric artery at its origin, 6) skeletonization of hepatic artery, 7) removal of lesser omentum with lymph nodes, 8) removal of greater omentum, 9) Roux-en-Y anastomosis, 10) resection of other organs such as left lobe of liver and transverse colon if required, or proximal partial gastrectomy and pyloroplasty Tumor metastasizes to all regional lymph nodes. Prognosis is poor. Any treatment at present is considered palliative.
Middle ⅓ 80%-90% gastrectomy, and steps 2-8 and 10 as above as needed with any Billroth II reconstruction or total gastrectomy Tumor metastasizes to certain group of regional lymph nodes. Prognosis better than for upper ⅓. Rates of morbidity and mortality are better than for upper ⅓.
Lower ⅓ Distal partial gastrectomy and steps 2-8 and 10 as above as needed with any Billroth II reconstruction Tumor also has a “regional mentality.” Best prognosis in comparison to the lesions located in the middle and upper one-thirds.

Source: Data from Nyhus LM, Wastell C (eds). Surgery of the Stomach and Duodenum (4th ed). Boston: Little, Brown, 1986.

Figure 15-62 and Table 15-15 in combination show lymph node groups possibly involved in carcinoma of the upper, middle, and lower thirds of the stomach. Remember the possible gastrectomies: the removal of all N1 nodes, or the more extreme gastrectomy to remove all the N2 lymph nodes.

Table 15-15. Lymph Node Groups for Gastric Cancer

Site of Primary Lesion N1 Nodes N2 Nodes
Upper third (Fig. 15-62)  Left cardiac Supra- and infrapyloric
  Right cardiac Left gastric
  Lesser curvature Common hepatic
  Greater curvature Splenic artery
    Celiac axis
Middle third (Fig. 15-62)  Right cardiac Splenic artery
  Lesser curvature Left cardiac
  Greater curvature Left gastric
  Supra- and infrapyloric Common hepatic
    Celiac axis
Lower third (Fig. 15-62)  Lesser curvature Right cardiac
  Greater curvature Left gastric
  Supra- and infrapyloric Common hepatic
    Celiac axis
N3 nodes: hepatoduodenal ligament, posterior aspect of pancreas, root of mesentery, diaphragmatic, paraesophageal
N4 nodes: middle colic, para-aortic nodes

Note: Nodes not common to site groups listed for N1 and N2 nodes have been designated N3 or N4 nodes depending on site.

Source: Modified from Fielding JWL, Hallissey MT. Adenocarcinoma of the stomach. In: Wastell C, Nyhus LM, Donahue PE (eds), 5th ed. Boston: Little, Brown & Co., 1995; with permission.

Fig. 15-62.

N1 and N2 lymph node groups related to carcinomas of upper, middle, and lower third of stomach. (After Nyhus LM, Wastell C (eds). Surgery of the Stomach and Duodenum (4th ed). Boston: Little, Brown, 1986.)

Billroth I

Variations of the Billroth I procedure are shown in Fig. 15-63.

Fig. 15-63.

Technical variations of Billroth I gastrectomy (classic Billroth I, standard Billroth I, and others). (After Siewert JR, Hölscher AH. Billroth I gastrectomy. In: Nyhus LM, Wastell C. Surgery of the Stomach and Duodenum, 4th Ed. Boston: Little, Brown and Co, 1986.)

Billroth II

Variations of the Billroth II procedure are shown in Figs. 15-64, 15-65, 15-66, and 15-67.

Fig. 15-64.

Partial gastrectomy: Billroth II gastric resection and variations.

Fig. 15-65.

Gastric resection with gastrojejunal anastomosis; several variations of Billroth II.

Fig. 15-66.

Gastric resection with Roux-en-Y anastomosis; a variation of Billroth II.

Fig. 15-67.

Billroth II resection with Roux-en-Y anastomosis.

An additional Billroth II modification is shown in Fig. 15-68, as well as another modification of Billroth I and a Roux-en-Y gastrojejunostomy.

Fig. 15-68.

Reconstructive procedures for partial gastrectomy. A, Modified Billroth I. B, Billroth II. C, Roux-en-Y gastrojejunostomy. (After Nyhus LM, Wastell C. Surgery of the Stomach and Duodenum, 4th ed. Boston: Little, Brown, 1986, Fig. 25-5.)

The miscellaneous gastric operations in Fig. 15-69 are not typical Billroth I or II procedures, but they have been used by several authors.

Fig. 15-69.

Miscellaneous gastric operations. (After Waugh JM, Hood RT Jr. Gastric operations: a historic review. Part 2. Q Rev Surg Obstet Gynecol 1954 March; 11(1):1-18, Fig. 7.)

Total Gastrectomy

(Figs. 15-70, 15-71) There are two indications for total gastrectomy: 1) Zollinger-Ellison syndrome, in which extreme gastric hyperacidity cannot be controlled by antisecretory agents; 2) gastric malignancy (epithelial and nonepithelial).

Fig. 15-70.

Total gastrectomy. (After Waugh JM, Hood RT Jr. Gastric operations: a historical review – II. Q Rev Surg Obstet Gynecol 11(1):1-18, 1954.)

Fig. 15-71.

Reconstructive procedures for total gastrectomy. A, Roux-en-Y. B, Henley jejunal interposition. C, Omega gastrojejunostomy. (After Nyhus LM, Wastell C. Surgery of the Stomach and Duodenum, 4th ed. Boston: Little, Brown, 1986, Fig. 25-6.)

Ovaska et al.252 reported that despite the increasing number of curative operations and more radical surgery, no progress in the 5-year survival rate was noted in patients with curative surgery.

Kwon and members of the Korean Gastric Cancer Study Group253 found no beneficial effect or increase in survival rate from splenectomy after curative total gastrectomy for gastric cancer.



Frozen section of margins is essential.

There are surgical options for nonepithelial tumors. For lymphomas, chemotherapy after surgery with or without radiation treatment is the procedure of choice.

For all patients with a postoperative life expectancy of at least 6 months, Schwarz et al.254 recommend the Ulm pouch (Fig. 15-72) with preservation of the duodenal passage. According to these authors, the quality of life and regulation of gastrointestinal hormones (blood glucose, insulin, cholecystokinin, motilin, secretin, and pancreatic polypeptide) for patients who have had this procedure was better in comparison to patients with different pouch formations.

Heimbucher et al.255 reported altered motility in asymptomatic patients after Hunt-Lawrence-Rodino pouch formation.

Fig. 15-72.

Ulm pouch: small and large pouch versions with 10 cm and 20 cm pouch lengths. PL, pouch length. (Modified from Schwarz A, Büchler M, Usinger K, Rieger H, Glasbrenner B, Friess H, Kunz R, Beger HG. Importance of the duodenal passage and pouch volume after total gastrectomy and reconstruction with the Ulm pouch: prospective randomized clinical study. World J Surg 20:60-67, 1996; with permission.)


Variations of esophagogastrectomy are shown in Fig. 15-73.

Fig. 15-73.

Esophagogastrectomy. (After Waugh JM, Hood RT Jr. Gastric operations: a historic review. Part 2. Q Rev Surg Obstet Gynecol 1954 March; 11(1):1-18, Fig. 10.)


Vagotomies may be truncal, highly selective, or extended highly selective.

Schuricht et al.256 stated that thoracoscopic vagotomy is a safe procedure to be used in patients with previous subdiaphragmatic vagotomy.

We quote from Donahue257 about highly selective vagotomy (HSV):

If operation is required for patients with an ulcer, vagotomy of some type will be part of the treatment; HSV provides the best overall operative treatment, and can be safely performed in most patients, excluding only those who are unstable or who have severe concomitant medical conditions which preclude the extra time (approximately 30-45 minutes) required for its performance.

While the role of open operations for ulcer complications is well established, there is little doubt that laparoscopic approaches will increase in popularity for appropriate candidates. Because practicing surgeons prefer to perform operations that they have performed for years and because there is a diminishing incidence and prevalence of ulcer in most areas of the world, surgeons who are interested in practicing HSV must make an effort to preserve it. If we continue to consider the merits of this operation compared with alternatives, HSV will always have a well-deserved place among surgical options.

Sapala et al.258 report cholelithiasis after truncal vagotomy and subtotal gastrectomy. Similar findings were noted by Rehnberg and Haglund,259 Lorusso et al.,260 Cipollini et al.,261 Ise et al.,262 and Inoue et al.263

Gastric Surgery for Colossal and Morbid Obesity

As reported by Deitel,264 Kremen et al. experimentally evaluated the nutritional importance of the proximal and distal small bowel in 1954. This was the genesis of surgical treatment (bariatric surgery) of clinically severe, morbid, and colossal obesity. Today, jejunoileal bypass has been abandoned and new techniques are in use. We advise the interested student to read the many excellent articles on the topic in World Journal of Surgery, Volume 22, 1998.

Surgical procedures for the treatment of morbid obesity include several types of gastric bypass (Figs. 15-74, 15-75, and 15-76), the vertical banded gastroplasty (Fig. 15-77), and pancreatobiliojejunal bypass (Fig. 15-78).

Fig. 15-74.

Gastric bypass for obesity. The gastric bypass has a measured, upper pouch volume of 25 ml or less with a 12-mm diameter stoma. (After Nyhus LM, Wastell C. Surgery of the Stomach and Duodenum, 4th ed. Boston: Little, Brown, 1986, Fig. 29-1.)

Fig. 15-75.

Gastric bypass with Roux-en-Y gastroenterostomy. (After Nyhus LM, Wastell C. Surgery of the Stomach and Duodenum, 4th ed. Boston: Little, Brown, 1986, Fig. 29-3.)

Fig. 15-76.

Gastric bypass operation involves creation of small stapled proximal gastric pouch connected by 10-mm anastomosis to 40-cm Roux-en-Y jejunal limb. (After Benotti PN, Forse RA. The role of gastric surgery in the multidisciplinary management of severe obesity. Am J Surg 169:361-367, 1995; Fig. 2.)

Fig. 15-77.

Vertical banded gastroplasty operation involves creation of small (15 mL to 20 mL) gastric reservoir along lesser curvature of stomach. Reservoir empties through narrow banded channel into remaining stomach. (After Benotti PN, Forse RA. The role of gastric surgery in the multidisciplinary management of severe obesity. Am J Surg 169:361-367, 1995; Fig. 1.)

Fig. 15-78.

Pancreatobiliojejunal bypass, including 60% gastrectomy and cholecystectomy. (After Nyhus LM, Wastell C. Surgery of the Stomach and Duodenum, 4th ed. Boston: Little, Brown, 1986, Fig. 29-4.)

Benotti et al.265 found an expanding body of evidence confirming significant weight control following gastric operations to manage severe obesity.

Curry et al.266 reported that resectional gastric bypass is an alternative procedure for morbid obesity (Fig. 15-79).

Fig. 15-79.

Resectional gastric bypass. Following resection of distal stomach, 30 cc to 50 cc gastric remnant is anastomosed to 50 cm retrocolic Roux-en-Y limb of jejunum. (After Curry TK, Carter PL, Porter CA, Watts DM. Resectional gastric bypass is a new alternative in morbid obesity. Am J Surg 175:367-370, 1998.)

Laparoscopically placed adjustable banding has seen used in Europe with no mortalities, reasonable complication rate, and successful weight reduction.267 A laparoscopic Roux-en-Y surgery similar to the open procedure in configuration and formation of the gastric pouch was reported by Higa et al.268

Gastric Transposition

Spiro et al.269 found that the stomach can be used for restoration of GI continuity after cervical esophagectomy or circumferential pharyngectomy. Mortality is high (11%) and complications are frequent (55%). Most of the complications are relatively minor, except for anastomotic leaks (13%) and partial gastric necrosis (3%). However, Ruangtrakool and Spitz270 found that the low major life-threatening morbidity and low mortality of gastric transposition makes it the preferred procedure for pediatric esophageal replacement.

Gastric Cancer: Some Observations on Etiologic Factors

Gastric carcinoma is the second most common cause of cancer-related death.271 Controversy regarding the genesis of gastric cancer continues. Adenomatous polyps, intestinal metaplasia, gastric ulcer, pernicious anemia, and other etiologic factors are reported as possibly playing a great role.

Adenomatous Polyps

Kimura et al.272 speculated that gastric cancer may develop in adenomatous gastric polyps.

Intestinal Metaplasia

Evrensel et al.273 speculated that intestinal metaplasia is a risk factor for cancer of the stomach. In a different publication, Evrensel and colleagues274 advanced the theory that Helicobacter pylori (H. pylori) is responsible for the formation of chronic gastritis and intestinal metaplasia which may be responsible for the genesis of gastric carcinoma. Watanabe275 also stated that intestinal metaplasia and gastric cancer are related.

McCloy et al.276 found that atrophy and intestinal metaplasia are associated with precancerous lesions and gastric carcinoma. Weston et al.277 wrote that intestinal metaplasia of the cardia of the stomach is infrequently found in patients with Barrett’s esophagus and most likely developed independently in relation to other parts of the stomach. Morales et al.278 believe that more work must be done to prove that gastric adenocarcinoma develops from intestinal metaplasia.

The Surgeon, Peptic Ulcer Disease, and Gastric Malignancy

The Surgeon and Peptic Ulcer Disease

In most cases of benign peptic ulcer disease today, conservative treatment is successful. Simeone et al.279 reported that 84% of giant peptic ulcers (>2 cm) may be treated without surgical intervention. It is only in those cases in which the battery of conservative approaches fails that the surgeon has to decide which procedure to perform. What follows here is a sampling of the current medical thought.

Hansson280 provides an excellent overview of the issues:

At present, there is no convincing evidence that pharmacologic inhibition of acid secretion increases the risk of stomach cancer. However, some studies indicate that prolonged treatment with PPIs may accelerate the development of atrophic gastritis, a risk factor of stomach cancer, in individuals infected with H. pylori. Thus long-term use of acid-inhibiting therapy should be viewed with caution. A link between previous gastric resection and stomach cancer has been found in a number of studies and appears to be real. The risk increases with the passage of time, and a twofold risk can be expected more than 15 years after the initial operation.

Donahue et al.233 reported that highly selective gastric vagotomy is an ideal procedure for the majority of patients with duodenal ulcer. We quote his further thought on laparoscopic extended highly selective vagotomy (EHSV)281:

Our bias that laparoscopic EHSV will emerge as the laparoscopic procedure of choice is a hopeful statement that modern surgeons will adopt this procedure. In this light, perhaps the greatest challenge is to the teaching centers, which have a responsibility to show resident surgeons the best of the old as well as the new.

Jordan and Thornby282 advocate parietal cell vagotomy for duodenal ulcer in preference to selective vagotomy-antrectomy because occasionally that procedure leaves a patient disabled.

Martin et al.283 believe that H. pylori infection is not influenced by highly selective vagotomy, and that ulcer recurrence depends on the completeness of vagotomy rather than on H. pylori status. Johnson284 cautions that future antibiotic-resistant strains of H. pylori may make elective proximal gastric vagotomy a more standard surgery.

Huang and Hunt285 advise combinations of antisecretory drugs with antimicrobial drugs to accelerate the healing of peptic ulcer.

Khulusi et al.286 believe that duodenal gastric metaplasia is partially due to both acid and H. pylori.

The Surgeon and Gastric Malignancy

Much controversy surrounds the specifics of surgically treating the stomach for malignant disease of epithelial type (adenocarcinoma), nonepithelial type (sarcoma), or stromal tumors. This leaves surgeons in the unenviable position of debating which procedure to use and how much of the stomach to resect. Should surgeons perform partial or extended lymphadenectomy? Distal pancreatectomy and splenectomy? Omentectomy? How much esophageal resection should be done for carcinoma of the cardia? What is the stage of the tumor? Has it metastasized to other organs? To the lymph nodes? Has it penetrated the layers of the gastric wall?

All these questions and considerations, as well as the age and condition of the patient, play a great role in the surgeon’s choice of procedure. The current thinking of several authors follows in an attempt to shed some light on the questions.

Kikuchi et al.287 found that extended lymphadenectomy in gastric cancer yields no positive impact on survival upon non-curative resection either in patients with gastric cancer and simultaneous metastases to the adjacent peritoneum or to the distant peritoneum. Karpeh and colleagues288 concluded that the number of positive nodes, rather than their location, is the best predictor of metastatic status in gastric cancer. Survival estimates are more truly represented when at least 15 nodes are examined.

Nurnberger et al.289 advised splenectomy only with advanced proximal tumors, especially those of the greater curvature. For early tumors they advised splenic lymphadenectomy of the hilum of the spleen to increase the completeness of radical gastrectomy.

We quote Pierie et al.290 on the treatment of gastrointestinal stromal tumors:

Complete gross surgical resection is presently the only means of cure for gastrointestinal stromal tumors. Tumors with more than 1 mitosis and a size larger than 5 cm have an especially poor prognosis, with decreased survival, and increased local and/or distal recurrence.

Kasakura et al.291 advocate splenic resection when positive node metastasis from advanced gastric cancer is found around the splenic hilum and artery. Pancreaticosplenectomy should be performed when the cancer lesion invades the pancreas.

Wanebo et al.292 recommended splenectomy in patients with gastric carcinoma only when absolutely necessary in order to completely remove the tumor. This usually occurs in stage IV cancer or in patients with metaplastic extension to the spleen and pancreas or macroscopic nodal metastases to the splenic hilum.

Bosing et al.293 concluded that since systematic lymphadenectomy is able to improve long-term survival for some tumor stages, it should be recommended as a standard procedure in all gastric cancer patients resected with curative intent.

Tsujitani et al.294 advised limited operation for cancer of the stomach in elderly patients. Ishigami et al.295 concluded that node dissection should be limited to perigastric nodes in elderly patients with early gastric cancer.

Hundahl et al.296 compared the outcome of stomach cancer surgery in Japanese patients in Tokyo (according to Japanese techniques) with surgery of Japanese in Honolulu (according to Western techniques). The authors observed higher survival rates in Tokyo. The study design eliminated race or the “different-disease” hypothesis as variables. The researchers hypothesized that lymphadenectomy-related stage migration and/or differences in therapeutic efficacy can explain the discrepancy in outcome between the two groups.

A study by Wanebo et al.215 did not support the view based on Japanese data that curative extragastric lymphadenectomy (D2 node dissection) increases the survival rate of patients undergoing gastrectomy. They recommended further studies. Roukos et al.297 reported that extended (D2) lymphadenectomy in patients with carcinoma of the stomach is a safe procedure. It is controversial, but has possible beneficial effects.

We quote from Meyer and Jähne:298

Complete tumor removal with margins of clearance at the resection lines must be the aim of today’s surgical treatment of gastric cancer, and this must be applied even in lymph node dissection. But, over the last few decades, the extent and impact of lymphadenectomy remains controversial. Whereas Japanese centers advocate extensive lymph node dissection as the base of their excellent results, many Western surgeons, supported by actual randomized trials, believe that the potential benefit of such procedures cannot outweigh the risk of increased postoperative morbidity and mortality. . . systematic lymph node dissection should be an integral part of the curative resection sought. Limited or no lymphadenectomy might be indicated in noncurative surgery or in special types of mucosal early gastric cancer, respectively.

Maehara et al.299 concluded that prophylactic lymph node dissection in patients with carcinoma of the stomach will prolong survival time. Wu et al.300 suggested that systematic dissection of nodes may be beneficial to patients with cancer of the distal one-third of the stomach.

Since a positive esophageal margin is an independent poor prognostic factor for long-term survival in gastric cancer, Chan et al.301 urge that all efforts be made to clear the esophageal margin in total and proximal gastrectomies.

Rüdiger Siewert and colleagues302 presented a topographicoanatomic classification of tumors of the esophagogastric junction:

Type I Adenocarcinoma of the distal esophagus
Type II True carcinoma of the cardia
Type III Subcardial gastric cancer infiltrating the distal esophagus

They add that this classification is based on marked differences between the tumor types and provides guidance for the selection of surgical approach. They found that for patients with Type II tumors, esophagectomy offered no advantage over extended gastrectomy if a complete tumor resection was achievable.

Kitamura et al.303 recommended the following for the surgical treatment of early gastric carcinoma:


Local gastric resection without lymphadenectomy for mucosal cancers of <2 cm in diameter and for elevated submucosal cancers of <1 cm in diameter

Gastrectomy with dissection of the perigastric nodes, the nodes along the left gastric artery, and the common hepatic artery, for the treatment of other early cancers of the stomach

Shimuzu et al.304 prefer laparoscopically assisted distal gastrectomy to the open procedure for the treatment of early gastric cancer, citing faster postoperative recovery, shorter hospital stays, and improved cosmetic outcome.

Imada et al.305 recommended pylorus-preserving gastrectomy rather than conventional distal gastrectomy for early gastric cancer in carefully selected patients because it resulted in better gallbladder function, improved condition of the stomach remnant, and better gastric emptying.

We quote from Nakane et al.306:

Pylorus-preserving gastrectomy has advantages over distal gastrectomy in terms of the avoidance of dumping syndrome and protection against duodeno-gastric reflux. However, more time was necessary for improved gastric fullness or food intake. Pylorus-preserving gastrectomy should be applied in younger patients with early gastric cancer expecting long survival.

Seto et al.307 concluded that patients with early-diagnosed gastric cancer, even with a single positive node, are at high risk for recurrence.

Otsuji et al.308 reported that total gastrectomy with distal pancreatectomy and splenectomy did not affect the survival of patients with gastric cancer. These authors emphasized severe complications of these procedures.

Kyzer et al.309 reported that in cases of gastric malignancy the Billroth I procedure is accompanied by significantly lower postoperative complication and mortality rates than the Billroth II procedure. The number of lymph nodes removed was practically the same in both procedures. The authors indicated that perhaps the Billroth I may have a higher recurrence rate.

Shibao et al.310 concluded that the gastric remnant with Billroth II anastomosis is susceptible to development of carcinoma.

Schepotin et al.311 strongly advised that for patients with T4 gastric cancer the treatment of choice should be aggressive en bloc surgical resection.

Graham et al.312 are very pessimistic about the outcome of adenocarcinoma of the gastric cardia. These authors used transhiatal esophagogastrectomy (78%), transthoracic esophagogastrectomy (21%), and transabdominal esophagogastrectomy (1%). The overall 5-year survival rate was approximately 16%.

Kodera et al.313 advised that in cancer of the proximal one-third of the stomach, pancreatosplenectomy should be performed only if the cancer extends to the pancreas.

Newman et al.314 reported that carcinoma that develops in the gastric remnant of patients who had previous gastrectomy for gastric cancer is practically the same cancer as the primary. The authors advise that the primary resection should be performed as a curative procedure.

Lo et al.315 stated that there is no difference in clinical behavior between resected gastric remnant cancer and other types of carcinoma of the stomach. Lo et al.316 reported the genesis of gastric cancer in the gastric remnant after partial gastrectomy for benign disease. The above authors stated that intestinal type of carcinoma was more common in the distal stomach (73%) than in the proximal stomach (50%).

Carcinoma of the gastric remnant is not as rare as formerly thought. According to Herrington and Sawyers,317 5% of gastrectomized patients are at risk 10 or more years after surgery. Farrands et al.318 reported endoscopic review of 71 patients more than 15 years after a gastric operation which revealed 11 with epithelial dysplasia. One of the lesions was carcinoma; another patient developed carcinoma 18 months later.

Nagamachi319 reported that after total gastrectomy in patients with gastric cancer, segmental pedicle interposition of the transverse colon is a feasible and useful procedure.

Tsujitani et al.320 reported that a limited gastrectomy and lymph node dissection is perhaps possible depending on the gross appearance and size of the tumor.

In patients for whom it has been at least 20 years since undergoing partial gastrectomy for benign disease, Greene321 advises a yearly endoscopy to rule out gastric remnant carcinoma.

Sano et al.322 advised total gastrectomy for lymphoma of the stomach followed by chemotherapy. Vaillant et al.323 reported that complete resection was better than primary resection in producing prolonged complete remission for both high- and low-grade localized gastric lymphomas.

Because of the possibility of invasion, Mulholland324 advised that attempts at curative resection of gastric stromal, carcinoid, or neuroendocrine tumors (leiomyomas, leiomyosarcomas, etc) are indicated in almost all cases.

Basing their study on a histologic classification of component mucosal endocrine cells, Schindl et al.325 raised issues of overtreatment and undertreatment of gastric neuroendocrine tumors:

It is obvious that surgical treatment of gastric neuroendocrine tumors influences the individual prognosis and the patient’s quality of life. Overtreatment of type 1 tumors by extended gastric resection will impair personal well-being unnecessarily without any advantage for an otherwise benign disease. Simple polypectomy of an invasive type 3 tumor implies a considerable risk of recurrence and progression of disease with lethal prognosis in certain cases. It was shown that gastric neuroendocrine tumors behave significantly divergently with respect to local invasion and distant spread, with the highest risk in type 3 tumors. Individual, type-adapted treatment is based on this knowledge, and any physician treating patients with gastric neuroendocrine tumors is responsible for its correct application.

Laparoscopic treatment of posterior gastric stromal tumors was advocated by Hepworth et al.,326 who stressed that delivery of the tumor through the gastrostomy is essential for success.

We quote Basso et al.:327

[L]aparoscopic resection of gastric stromal tumors should be considered as the treatment of choice. Wedge resection of anterior wall lesions is generally performed. The treatment of posterior wall lesions is still controversial. In our opinion the direct approach should be reserved for lesions located on the posterior wall of the body, which can be easily reached through the greater omentum, while the transgastric approach should be preferred for lesions located on the fundus and antrum.

In summary, proximal or distal subtotal or total gastrectomy is the treatment of choice for malignant tumors of the stomach; extended lymphadenectomy may be essential, but it is still controversial. With proximal gastrectomy, a complementary pyloroplasty is necessary for good gastric emptying.

The question of whether total or subtotal gastrectomy should be performed for carcinoma of the stomach continues to be debated. Harrison et al.328 stated that total and proximal gastrectomy have similar survival rates as well as time to recurrence and rates of recurrence. They concluded that total gastrectomy is not necessary for proximal gastric cancer (5 year survival was 43% for proximal vs 41% for total).

The following statement about node dissection and gastric carcinoma made by Dr. Blake Cady at the 39th Annual (2000) Meeting of the Georgia Surgical Society provides further evidence that there is confusion about how radical surgery should be:

The last decade’s enthusiasm by Japanese surgeons for super radical lymphatic resection for gastric cancer has now been put in perspective by the two recent multi-institutional trials from Holland and the United Kingdom, which show no difference in survival. In addition, as one might expect, the more radical surgery has had significant increases in complications and operative mortality. Even accounting for the increased operative mortality, there is no increase in survival in gastric cancer by the performance of a more radical lymph node dissection.

A Billroth II anastomosis, modified to the individual’s requirements and the surgeon’s preference, is recommended for most reconstruction procedures.

Appropriately placed clips will help the roentgenologist locate the area to be irradiated for unresectable lesions. Palliative bypass procedures may be used for huge unresectable tumors of the duodenum but not for those of the stomach.

Anatomic Complications of Gastric Surgery

The complications of gastric surgery (Table 15-16) are a combination of complications with multiple sequelae of surgery of the duodenum, pancreas, abdominal esophagus, spleen, and small or large bowel.

Table 15-16. Anatomic Complications of Some Gastric, Duodenal, and Pancreatic Procedures*

Procedure Vascular Injury Organ Injury Inadequate Procedure
Gastrectomy Hemorrhage, ischemia, or necrosis Spleen, liver, mediastinal pleurae, pericardium, cisterna chyli, esophagus, gastric remnant necrosis, omental infarction, common bile duct, pancreas, colon Incomplete vagotomy, inadequate gastric resection, small stoma, antral remnant, anastomotic leakage, duodenal blowout secondary to obstruction
  a. total   a. distal esophagus
  b. proximal   b. gastric remnant
  c. distal with or without vagotomy   c. duodenal cuff
Pyloroplasty Hemorrhage, leakage of suture line Pancreas, common bile duct Small stoma
Pyloromyotomy Hemorrhage Duodenum Small stoma
Vascular compression of duodenum Aorta, superior mesenteric artery and vein, inferior mesenteric artery and vein, middle colic artery, small vessels in suspensory muscle Colon, duodenum, jejunum Failure to sever ligament of Treitz
Exploration of proximal duodenum Right gastric artery, gastroduodenal artery, pancreaticoduodenal arcades, superior mesenteric artery and vein, inferior vena cava, aorta Common bile duct, pancreas, colon, right kidney Failure of good mobilization with multiple sequelae
Exploration of distal duodenum As above and inferior mesenteric vein As above and jejunum As above
Paraduodenal herniae Superior mesenteric artery and vein, inferior mesenteric vein Pancreas, colon, jejunum Inadequate closure of hernia ring, overenthusiastic closure of hernia ring with mesenteric vessel involvement
Pancreaticoduodenostomy and pancreaticojejunostomy As above As above As above

*The anatomic complications of gastroduodenal surgery are a combination of complications with multiple sequelae of surgery of the stomach, pancreas, and duodenum. The table is not complete. The philosophy behind it is to alert the surgeon to these complications with one glimpse and to enable him or her to exercise good surgical judgement and technique in the operating room.

Source: Skandalakis LJ, Pemberton LB, Gray SW, Colborn GL, Skandalakis JE. The duodenum. Part 4: Surgery. Am Surg 55(8):492-494, 1989; with permission.

Gastric resection is probably a rare operation today, given the modern treatment of peptic ulcer. The contemporary surgeon, whose experience may be more centered around operations to treat benign or malignant gastric neoplasms, might be said to be at a disadvantage in dealing with complications of gastric resection, one of yesterday’s most common procedures. For those occasions when gastric resection is the best choice, we present and analyze its anatomic complications.

The anatomic complications of gastric operations will be presented according to the procedure performed.

Pickleman et al.329 reported that foregut procedures were accompanied by increased rate of anastomotic leak, and they stated pessimistically that

. . .intestinal anastomotic dehiscence remains a major unsolved problem in gastrointestinal surgery. With few exceptions, there are few predictive patient or technical variables, and therefore little can be accomplished to reduce its incidence. Overall, more than 80% of all mortality stems from other factors and therefore, even if leaks were eliminated, little impact on surgical mortality rates would be noted. Leaks occur later postoperatively than is commonly assumed, and therefore early hospital discharge may prove harmful to some patients.

We do believe that good knowledge of embryology and anatomy and surgical technique is the answer.


Intraluminal or extraluminal bleeding may take place at the suture line due to a lack of ligation of the submucosal vascular plexus.

In the event of breakage of the suture line or peritonitis, reoperation is mandatory for ligation of bleeding and closure of the perforation.


Complications and reoperation are the same as for gastrorrhaphy.

Gastrostomy, Duodenostomy, Duodenoduodenostomy

Acutely ill patients are at high risk for morbidity and mortality following insertion of a percutaneous endoscopic gastrostomy tube. Abuksis et al.330 advise avoiding the procedure for these patients.

Cosentini et al.331 stated that surgical, percutaneous endoscopic, and percutaneous radiologic gastrostomies all have similar complications but that tube function after radiologic gastrostomy tends to be inferior.

Complications include bleeding and leakage around the tube, with abscess formation or peritonitis. Reoperation is essential for ligating the bleeding vessels of the stomach or the abdominal wall, and for fixation of the gastric wall stoma to the peritoneum of the abdominal wall.

Peritonitis is an iatrogenic complication, due to malpositioning332 or technical error (non-fixation of the gastric wall around the stoma to the peritoneum of the anterior abdominal wall).

According to Kobak et al.,333 percutaneous endoscopic gastrostomy tube removal later than 11 months after insertion requires surgery rather than traction to avoid persistent gastrocutaneous fistulous leakage.


Complications include bleeding from injury of the left gastric artery, inferior phrenic artery, spleen, or left hepatic vein. Lymphatic leakage may occur from injury of the cisterna chyli. Lower esophageal stenosis is also possible.

Reoperation is mandatory. If complications are noted in the operating room, thoracoabdominal incision and repair or ligation of the left hepatic vein can be performed. The cisterna chyli can be ligated with impunity. Fortunately, dysphagia secondary to esophageal stenosis is transient. Gastroesophageal reflux is rare, and can be treated conservatively.


The preferred procedure for cardioplasty is esophagomyotomy. Its main complications are:


Injury to the vagal trunk, resulting in pylorospasm and reflux esophagitis

Incomplete division of the muscular ring

Iatrogenic diaphragmatic hernia

Esophageal obstruction

Esophageal perforation

Aspiration pneumonitis

Bleeding Peptic Ulcer

Bleeding and suture line leakage are rare. If they occur, reoperate and repair. With duodenal ulcers, stop bleeding by under-running the base of the ulcer where the bleeding gastroduodenal artery (or rarely the posterior superior pancreatoduodenal artery) is located. Use non-absorbable sutures.


In 10% of patients86 there is only one pancreatic duct, the duct of Santorini. It is located under, or deep to, the gastroduodenal artery. A deep suture would ligate the only duct of the pancreas.

Closure of Perforated Peptic Ulcer

Gastroduodenal stenosis may take place. Employ ulcer-definitive surgery such as pyloroplasty with or without vagotomy.


Complications of the Heineke-Mikulicz pyloroplasty (1886-1887) (see Fig. 15-55), Finney pyloroplasty (1902), and Jaboulay pyloroplasty include bleeding, suture line break, and small stoma due to inadequate procedure. Rarer complications are pancreatitis due to ligation of the duct of Santorini, and jaundice caused by ligation of the common bile duct.

Reoperation and repair are in order after evidence that the jaundice is not transient.


Complications include bleeding, duodenal leakage, and small stoma due to inadequate procedure.

Reoperation is needed to stop bleeding. Close the perforation if it is leaking. Perform a pyloroplasty to repair the small stoma, or try to perform a new blunt separation of the muscular coat of the area. If perforation is recognized in the operating room, close with a single 4-0 purse-string absorbable suture.

Hulka et al.334 reported complications after pyloromyotomy in infants with hypertrophic pyloric stenosis. They stated that duodenal perforation is easily recognized and treated with minimal morbidity. Postoperative vomiting lasting more than 5 days should be evaluated and treated. Incomplete pyloromyotomy requires a second myotomy.


Complications include bleeding from the submucosal gastric vascular plexus, obstruction secondary to small stoma, and anastomotic leakage. Reoperation is imperative.

Hernia Following Retrocolic Gastrojejunostomy

The surgical anatomy of the ring can be appreciated in Figure 15-80A. The following are the boundaries of the ring:


Anterior: The gastrojejunostomy and the efferent or afferent jejunal loop, depending on whether the afferent loop is attached to the lesser or the greater curvature of the stomach

Posterior: Posterior parietal peritoneum

Superior: Transverse mesocolon and posterior wall of gastric remnant

Inferior: Ligament of Treitz and duodenojejunal peritoneal fold

Fig. 15-80.

Retroanastomotic hernias. A, Hernia following retrocolic gastrojejunostomy. B, Hernia following antecolic gastrojejunostomy. (Modified from Skandalakis JE, Gray SW, Akin JT Jr. The surgical anatomy of hernial rings. Surg Clin North Am 1974;54:1227-1246; with permission.)

Do not make an incision. Perform an enterostomy to facilitate reduction of the loop. Closure of the ring will prevent recurrence.

Hernia Following Antecolic Gastrojejunostomy

(Fig. 15-80B) The boundaries of the ring with the afferent loop attached to the greater curvature of the stomach are as follows:


Anterior: The gastrojejunostomy and the afferent jejunal loop

Posterior: Omentum and mesocolon

Superior: Transverse colon and mesocolon

Inferior: Ligament of Treitz and duodenojejunal peritoneal fold

The boundaries of the ring with the afferent loop attached to the lesser curvature of the stomach are as follows:


Anterior: Afferent jejunal loop with its mesentery

Posterior: Omentum, transverse colon, and mesocolon

Superior: The gastrojejunostomy, ligament of Treitz, and duodenojejunal peritoneal fold

Inferior: Jejunum with its mesentery

No incision of the ring is necessary.

Jejunogastric Intussusception

The cause of jejunogastric intussusception is unknown. Its occurrence in the absence of previous gastric surgery has been reported.335 We agree with Fromm336 that it does not make any difference whether an anastomosis is performed in an antecolic or retrocolic position, or whether the jejunal loop is placed in the so-called isoperistaltic or antiperistaltic position. Fromm stated that it is very rare to find a definite anatomic or pathologic cause for jejunogastric intussusception. Because of the slight chance, we mention this as a possible anatomic complication.


The complications of gastrectomy are as follows:


Bleeding from gastric wall (intraluminal or extraluminal)

Bleeding from spleen due to injury of the capsule

Bleeding from omentum

Duodenal stump blowout with bile peritonitis due to inadequate procedure

Common bile duct injury and bile peritonitis

Pancreatitis due to ligation of single pancreatic duct (duct of Santorini) occurring in 10% of patients86

Gastroduodenal stenosis in Billroth I due to inadequate procedure

Gastrojejunostomy dehiscence

Anastomotic leak

Gastrojejunostomy stenosis


Internal herniation

Afferent and efferent loop syndrome

Injury of the vaterian system

Inadequate procedure for total removal of distal antrum

Intussusception of distal loop

Inadequate procedure due to submucosal spread and infiltration of stomach cancer

Omental infarction

Gastric remnant necrosis

Zittel et al.337 reported a high prevalence of bone disorder after gastrectomy. This problem was, most likely, secondary to a calcium deficit process.

To present all the anatomic complications in detail is beyond the scope of this chapter. We feel, however, that for some complications such as afferent (proximal) and efferent (distal) acute or chronic loop syndrome, more detailed description is essential.

Afferent Loop Syndrome

Afferent loop syndrome (Fig. 15-81, Fig. 15-82, Fig. 15-83) is secondary to obstruction of the proximal stoma or extrinsic obstruction of a short limb by adhesions, by torsion or volvulus of a long limb, or by retroanastomotic hernia. The internal hernia complication may take place with antecolic or retrocolic gastrojejunostomy. Complications include distention of the proximal loop, blowout of the duodenal stump, bleeding, necrosis of the wall of the loop, and perforation.

Fig. 15-81.

Proximal loop (afferent loop) syndrome may result in blowout of duodenal stump. High-grade partial obstruction may result in distention of afferent loop. (Modified from Hardy JD. Complications of gastric surgery. In: Hardy JD (ed). Complications in Surgery and Their Management (4th ed). Philadelphia: WB Saunders, 1981; with permission.)

Fig. 15-82.

Some causes of afferent loop obstruction. A, Volvulus. B, Kink at site of anastomosis. C, Adhesions. D, Retroanastomotic hernia. (Modified from Fromm D. Complications of Gastric Surgery. New York: John Wiley & Sons, 1977; with permission.)

Fig. 15-83.

The retroanastomotic space which occurs after retrocolic and antecolic gastrojejunostomy. Most hernias, as indicated by the arrow, occur from right to left. (Modified from Fromm D. Complications of Gastric Surgery. New York: John Wiley & Sons, 1977; with permission.)

The method used most commonly to manage the afferent loop syndrome is enteroenteroanastomosis.

Efferent Loop Syndrome

Obstruction of the distal stoma by kinking or extrinsic obstruction of the loop by adhesion (Fig. 15-84) may produce gastric distention and vomiting. Neurogenic origin (functional obstruction) of this syndrome without obstruction or adhesions (gastric atony) was reported by Golden.338

Fig. 15-84.

Fibrinous adhesions may partially or totally obstruct distal loop. (After Hardy JD. Complications in Surgery and Their Management. Philadelphia: WB Saunders, 1981, Fig. 19-9.)

Management is conservative if gastric atony is suspected. If gastric atony is ruled out, reoperation is essential.

Reoperation is mandatory for immediate postoperative complications.


Gastroileostomy is an iatrogenic complication caused by anastomosis of the stomach to the ileum. The site of the stoma is from 6 cm proximal to the ileocecal junction to 183 cm distal to the ligament of Treitz.336


Complications of vagotomy include:


Unilateral or bilateral pneumothorax secondary to perforation of mediastinal pleurae

Esophageal perforation

Partial esophageal stenosis

Bleeding from spleen, liver, or omentum

Injury of pancreas, common bile duct, or colon

Postoperative hiatal hernia

Preexisting dysphagia appears to play an integral role in persistent dysphagia following laparoscopic vagotomy. Shiino339 advises the construction of a loose fundoplication in patients with dysphagia.

Anatomic Complications of Gastric Surgery for Colossal and Morbid Obesity

Complications of surgery for obesity include:


Postoperative leakage and peritonitis

Distal stomach necrosis secondary to gastric distention due to afferent limb syndrome

Marginal ulcer (in about 10 percent of gastric bypass surgeries)340

Weiss et al.341 reported that laparoscopic adjustable gastric banding resulted in an impairment of esophageal body function, leading to weak esophageal peristalsis.

Gastric Transposition

Complications depend on the extent of the procedure. Spiro et al.269 present the anatomic complications of gastric transposition in Table 15-17.

Table 15-17. Complications Directly Related to Gastric Transposition

  No. of Patients %
Tracheal rent, necrosis 21 18
Wound infection, necrosis, dehiscence 19 16
Anastomotic fistula 15 13
Bleeding 8 7
Splenectomy 8 7
Pleural effusion, empyema, chylothorax 4 3
Stomach necrosis (minor) 3 3
Central line injury 1 1

Source: Spiro RH, Bains MS, Shah JP, Strong EW. Gastric transposition for head and neck cancer: a critical update. Am J Surg 162:348-352, 1991; with permission.

Surgical Applications


Some complications, such as duodenal stump blowout, common bile duct injury, partial stomal obstruction, etc., are iatrogenic.

When the first part of the duodenum is highly distorted, it should be treated by duodenostomy.

The anastomotic stoma should be checked for adequate lumen.

We always (if the word always is permissible) advise Billroth II gastrojejunostomy for carcinoma of the stomach. This possibly avoids recurrence of carcinoma at the anastomosis. Chareton et al.324 reported that Billroth I gastrectomy is associated with an increased risk of fistula and of recurrence of carcinoma at the hepatic pedicle.

Wu et al.300 suggest that survival rate relates to the extent of lymph node metastasis. Systematic lymph node dissection may have a beneficial effect.

Markowitz,343 the authority on incarcerated retro-anastomotic hernias, stated: “Construction of each gastrojejunostomy, either as a procedure by itself or as a step in the performance of a partial gastrectomy, results in the formation of an internal hernia ring which may be the cause of severe difficulties in the immediate or distant postoperative period.”

Billroth II anastomosis may be converted to Billroth I for the treatment of dumping syndrome or Billroth II may be revised by narrowing the stoma of gastrojejunostomy (Fig. 15-85). In 1952 Henley344,345 proposed his technique which consists of interpositioning (Fig. 15-86) an isoperistaltic jejunal segment of 10 to 20 cm between the gastric remnant and duodenum. Later, antiperistaltic colonic or jejunal segment replaced the isoperistaltic jejunal segment. And modifications continue. These include the triply plicated isoperistaltic jejunal pouch between the gastric remnant or esophagus and duodenum (technique of Hays346) (Fig. 15-87) and the doubly plicated jejunal pouch that had an isoperistaltic inlet but an antiperistaltic outlet (technique of Poth). Jejunogastric intussusception takes place only on Billroth II anastomoses. According to Shackman,347 intussusception can occur with the afferent loop into the stomach, with the efferent loop into the stomach, or with both afferent and efferent loops into the stomach. Efferent loop intussusception is the most common, occurring in over 4% of cases.

Gastrojejunocolic fistula (Fig. 15-88) is a complication of retrocolic gastrojejunostomy.

Reflux gastritis (bile gastritis) (Fig. 15-89) is a complication of truncal vagotomy and partial proximal or distal gastrectomy in a number of patients, 10-20% according to Mulholland.155 These patients develop a poorly understood syndrome with the following very unpleasant clinical triad:


– Postprandial pain with occasional nausea and vomiting

– Bile reflux into the gastric remnant or into the stomach after vagotomy and some draining procedures

– Histologically proved gastritis

Fig. 15-85.

To treat dumping syndrome, conversion of Billroth II anastomosis to Billroth I may be done, or Billroth II may be revised by narrowing the stoma of the gastrojejunostomy. (Modified from Fromm D. Complications of Gastric Surgery. New York: John Wiley & Sons, 1977; with permission.)

Fig. 15-86.

Interposition operations. (Modified from Fromm D. Complications of Gastric Surgery. New York: John Wiley & Sons, 1977; with permission.)

Fig. 15-87.

Pouches used by Hays and Poth in treatment of dumping symptoms. (Modified from Fromm D. Complications of Gastric Surgery. New York: John Wiley & Sons, 1977; with permission.)

Fig. 15-88.

A gastrojejunocolic fistula associated with a retrocolic anastomosis. A proximal transverse colostomy has been performed. The inset shows the fistula in cross section. (Modified from Fromm D. Complications of Gastric Surgery. New York: John Wiley & Sons, 1977; with permission.)

Fig. 15-89.

Some of the operative procedures used to treat reflux gastritis. Roux-en-Y drainage of the stomach is the most effective procedure but approximates the Mann-Williamson ulcer model. Vagotomy combined with antrectomy is best to avoid recurrent ulceration. The Tanner 19 procedure is a method for perfoming Roux-en-Y drainage that may be technically simpler to do under certain anatomic conditions. However, this procedure carries the potential for the bacterial overgrowth syndrome because of the recirculating loop adjacent to the gastroenterostomy. A gastroenterostomy without vagotomy is rarely done today in patients with benign disease. A less effective procedure for preventing reflux is interposition of an isoperistaltic jejunal, or Henley, loop between the gastric remnant and duodenum. The Soupault maneuver is a method for converting the efferent limb of a Billroth II gastrectomy into a Henley loop. Effective acid antisecretory procedures must also accompany the Henley loop to prevent recurrent ulceration. (Modified from Fromm D. Complications of Gastric Surgery. New York: John Wiley & Sons, 1977; with permission.)

If medical treatment is not successful (antacids, H2-receptor antagonists, etc.), the treatment is operative diversion to avoid intestinal contents coming in contact with the gastric mucosa. A long Roux-en-Y gastrojejunostomy will divert the bile from the gastric remnant and perhaps help the patient.


The second most difficult decision to make in surgery is when to operate.The most difficult decision is when to re-operate.—R.K. Tompkins348

It is not within the scope of this book to present technique for reoperative surgery. The normal anatomy and the virginity of the area have been tampered with prior to reoperation. The surgeon must read the operating room report of the previous surgery very carefully. Perhaps he or she should consult with other surgeons. By all means, the surgeon should read, reread, and reread again “Some Principles of Reoperative Surgery,” a chapter by the late Dr. Robert M. Zollinger in the excellent book Reoperative Surgery,348 edited by Ronald K. Tompkins.

Reoperation after gastric surgery may be performed for several reasons. These include, but are not limited to:


Recurrent peptic ulcer disease

Failure of antireflux procedures

Alkaline reflux gastritis

Failure of gastric bypass for obesity

Postgastrectomy syndromes

The interested student will find valuable information about reoperation in the books of White et al.349 and McQuarrie and Humphrey.350

When trying to determine the advisability of reoperation due to prolonged gastric stasis or delayed gastric emptying after gastric surgery, keep in mind that the need for reoperation is rare. We agree with the findings of Bar-Natan et al.351 that gastric motility returns within 3 to 6 weeks.


1. Allgöwer, M. Foreword. In Nyhus LM, Wastell C. Surgery of the Stomach and Duodenum, 4th Ed. Boston: Little, Brown and Co, 1986.

2. O’Rahilly R, Müller F. Human Embryology & Teratology, 2nd Ed. New York: Wiley-Liss, 1996, p. 227.

3. Skandalakis JE, Gray SW (eds). Embryology for Surgeons: The Embryological Basis for the Treatment of Congenital Anomalies, 2nd Ed. Baltimore: Williams & Wilkins, 1994.

4. Velasco AL, Wolcomb GW III, Templeton JM Jr, Ziegler MM. Management of congenital microgastria. J Pediatr Surg 25:192-197, 1990. [PubMed: 2303988]

5. Gerbeaux J. Absence congenital d’estomac. Ann Pediatr 1971;18:349. [PubMed: 5580642]

6. Blank E, Chisholm AJ. Congenital microgastria: a case report with a 26-year follow-up. Pediatrics 51:1037, 1973. [PubMed: 4710436]

7. McMullin CJ Jr. Results and morbidity of 83 pyloromyotomies for infantile hypertrophic pyloric stenosis in a rural general surgery teaching program. Contemp Surg 2001;57:57-62.

8. Cook RCM. Gastric outlet obstructions. In: Lister J, Irving IM (eds). Neonatal Surgery, 3rd Ed. London: Butterworths, 1990, p. 415.

9. Spector JM, Chappell J. Gastric volvulus associated with wandering spleen in a child. J Pediatr Surg 2000;35:641-642. [PubMed: 10770406]

10. Gray SW, Johnson HC, Skandalakis JE. Gastric teratoma in an adult: with a review of the literature. South Med J 57(11):1346-1351, 1964.

11. Lingawi SS, Filipenko JD. Brunner’s gland hamartoma causing gastric outlet obstructive symptoms. South Med J 91(10):964-965, 1998.

12. Gahagan T. The function of the musculature of the esophagus and stomach in the esophagogastric sphincter mechanism. Surg Gynecol Obstet 114:293, 1962. [PubMed: 13896201]

13. Maingot R. Abdominal Operations, 6th Ed. New York: Appleton-Century-Crofts, 1974.

14. Paramanandhan TL. The duodenal spread of gastric carcinoma. Br J Surg 54:169-174, 1967. [PubMed: 6020984]

15. Michels NA. Blood Supply and Anatomy of the Upper Abdominal Organs. Philadelphia: Lippincott, 1955.

16. Van Damme JP. Behavioral anatomy of the abdominal arteries. Surg Clin North Am 73:699-725, 1993.

17. Weiglein AH. Variations and topography of the arteries in the lesser omentum in humans. Clin Anat 9:143-150, 1996. [PubMed: 8740472]

18. Hayward J. The lower end of the oesophagus. Thorax 16:36, 1961. [PubMed: 13712529]

19. Bombeck CT, Dillard DH, Nyhus LM. Muscular anatomy of the gastroesophageal junction and role of phrenoesophageal ligament. Autopsy study of sphincter mechanism. Ann Surg 164:643, 1966. [PubMed: 5924786]

20. Botha GSM. Mucosal folds at the cardia as a component of the gastroesophageal closing mechanism. Br J Surg 45:569, 1958. [PubMed: 13560757]

21. Dillard DH. Esophageal sphincter and reflux. Surg Clin North Am 44:1201, 1964.

22. Friedland GW, Meicher DH, Berridge FR, Gresham GA. Debatable points in the anatomy of the lower oesophagus. Thorax 21:487, 1966. [PubMed: 5972516]

23. Androulakis JA, Skandalakis JE, Gray SW. Contributions to the pathological anatomy of hiatal hernia. J Med Assoc Ga 55:295, 1966. [PubMed: 5918519]

24. Fukatsu K, Saito H, Han I, Yasuhara H, Lin MT, Inoue T, Furukawa S, Inaba T, Hashiguchi Y, Matsuda T, Muto T. The greater omentum is the primary site of neutrophil exudation in peritonitis. J Am Coll Surg 183:450-456, 1996. [PubMed: 8912613]

25. Griffith CA. Anatomy of the stomach and duodenum. In: Wastell C, Nyhus LM, Donahue PE (eds). Surgery of the Esophagus, Stomach, and Small Intestine, 5th Ed. Boston: Little, Brown and Co, 1995, pp. 388-417.

26. Griffith CA. Anatomy. In: Nyhus LM, Wastell C (eds). Surgery of the Stomach and Duodenum, 4th ed. Boston: Little, Brown & Co, 1986, pp. 47-48, 75.

27. Chang FY, Lu CL, Chen CY, Lee SD, Doong ML, Wang PS. Distal stomach appears essential in the regulation of gastrointestinal transit. J Gastroenterol 2000;35:424-428. [PubMed: 10864348]

28. Berger EH. The distribution of parietal cells in the stomach: a histotopographic study. Am J Anat 54:87, 1934.

29. Debas HT. Physiology of gastric secretion and emptying. In: Miller TA (ed). Physiologic Basis of Modern Surgical Care. St. Louis: CV Mosby Co, 1988, p. 280.

30. Harrington SW. Esophageal hiatal diaphragmatic hernia. Surg Gynecol Obstet 100:277, 1955. [PubMed: 14359078]

31. Terracol J, Sweet RH. Diseases of the Esophagus. Philadelphia: Saunders, 1958.

32. Allison PR. Peptic ulcer of oesophagus. Thorax 3:20, 1948.

33. Oglesby JE. The E-G junction. Two sphincters. Int Surg 60:135, 1975. [PubMed: 1123264]

34. Rector LE, Connerley ML. Aberrant mucosa in the esophagus in infants and in children. Arch Pathol 31:285, 1941.

35. Palmer ED. An attempt to localize the normal esophagogastric junction. Radiology 60:825, 1953. [PubMed: 13064387]

36. Robbins SL, Cotran RS. Pathologic Basis of Disease, 2nd Ed. Philadelphia: WB Saunders, 1979, p. 923.

37. Chandrasoma PT, Der R, Ma Y, Dalton P, Taira M. Histology of the gastroesophageal junction: an autopsy study. Am J Surg Pathol 2000;24:402-409. [PubMed: 10716154]

38. Lerche W. The Esophagus and Pharynx in Action: A Study of Structure in Relation to Function. Springfield IL: CC Thomas, 1950.

39. Pisko-Dubienski ZA. Anatomy of sliding hiatal hernia. In: Nyhus LM, Harkins HN (eds). Hernia. Philadelphia: Lippincott, 1964.

40. Mann CV, Greenwood RK, Ellis FH Jr. The esophagogastric junction. Surg Gynecol Obstet 118:853, 1964. [PubMed: 14141337]

41. Jackson RG. Anatomic study of the vagus nerves. With a technique of transabdominal selective gastric vagus resection. Arch Surg 57:333, 1948.

42. Winans CS. The pharyngoesophageal closure mechanism: a manometric study. Gastroenterology 63:768-777, 1972. [PubMed: 5079487]

43. Winans CS, Harris LD. Quantitation of lower esophageal sphincter competence. Gastroenterology 52:73-78, 1967.

44. Fisher RS. Lower esophageal sphincter or a barrier to gastroesophageal reflux before and after antireflux surgery. South Med J 71:22, 1978. [PubMed: 628870]

45. Lewis FT. The form of the stomach in human embryos with notes upon the nomenclature of the stomach. Am J Anat 13:477-503, 1912.

46. Salessiotis N. Measurement of the diameter of pylorus in man, Part I: Experimental project for clinical application. Am J Surg 124: 331, 1972. [PubMed: 5056893]

47. Salessiotis N. Gastroenteric anastomosis in Billroth II gastrectomy with maintenance of the physiologic diameter of the normal pylorus to prevent the dumping syndrome, Part II: results in 1300 cases. Am J Surg 129:656, 1975.

48. Browne D. The surgical anatomy of Ramstedt’s operation. Arch Dis Child 6:129, 1931.

49. Torgerson J. Muscular build and movements of stomach and duodenal bulb: especially with regard to problem of segmental divisions of stomach in light of comparative anatomy and embryology. Act Radiol [Suppl] 45:1, 1942.

50. Oi M, Oshida K. The association of esophageal, gastric, and duodenal ulcers; case report. Gastroenterology 1959;36:57-59. [PubMed: 13620017]

51. Oi M, Oshida K, Sugimura S. The location of gastric ulcer. Gastroenterology 1959;36:45-56. [PubMed: 13620016]

52. Oi M, Sakuri Y. The location of duodenal ulcer. Gastroenterology 1959;36:60-64. [PubMed: 13620018]

53. Oi M, Tanaka Y, Yoshida K, Yoshikawa K. Dual control of peptic ulcers by both the gastric mucosa and musculature. Rev Surg 1966;23:373-374. [PubMed: 5919611]

54. Oi M, Ito Y, Kumagi F, Yoshida K, Tanaka Y, Yoshikawa K, Miho O, Kijima A. A possible dual control mechanism in the origin of peptic ulcer. A study on ulcer location as affected by mucosa and musculature. Gastroenterology 1969;57:280-293. [PubMed: 5808487]

55. Woodburne RT, Burkel WE. Essentials of Human Anatomy, 8th Ed. New York: Oxford University Press, 1988, p. 440.

56. Williams PL, Warwick R, Dyson M, Bannister LH (eds). Gray’s Anatomy (37th ed). New York: Churchill Livingstone, 1989.

57. Sun WM, Doran SM, Jones KL, Davidson G, Dent J, Horowitz M. Long-term effects of pyloromyotomy on pyloric motility and gastric emptying in humans. Am J Gastroenterol 2000;95:92-100. [PubMed: 10638565]

58. Cormack DH. Ham’s Histology, 9th Ed. Philadelphia: JB Lippincott, 1987, p. 494.

59. Fass R, Sampliner RE. Extension of squamous epithelium into the proximal stomach: a newly recognized mucosal abnormality. Endoscopy 32(1):27-32, 2000.

60. Junqueira LC, Carneiro J, Kelley RO. Basic Histology (6th ed). Norwalk: Appleton and Lange, 1989.

61. Davenport HW. The gastric mucosal barrier. Digestion 5:162, 1972. [PubMed: 4111079]

62. Tepperman BL, Jacobson ED. Circulatory factors in gastric mucosal defense and repair. In: Johnson LR (ed). Physiology of the Gastrointestinal Tract (3rd ed). New York: Raven Press, 1994, pp. 1331-1352.

63. Ritchie WP Jr, Mercer DW. Mediators of the bile acid-induced alterations in gastric mucosal blood flow. Am J Surg 161:126, 1991. [PubMed: 1987846]

64. Green T, Dockray GJ. Characterization of the peptidergic afferent innervation of the stomach in the rat, mouse and guinea-pig. Neuroscience 25:181, 1988. [PubMed: 2455875]

65. Mercer DW. Special comment—the gastric mucosal barrier. In: Wastell C, Nyhus LM, Donahue PE (eds). Surgery of the Esophagus, Stomach, and Small Intestine, 5th Ed. Boston: Little, Brown and Co, 1995, p. 430.

66. Brown BP, Schulze-Delrieu K, Schrier JE, Abu-Yousef MM. The configuration of the human gastroduodenal junction in the separate emptying of liquids and solids. Gastroenterology 1993;105:433-440. [PubMed: 8335199]

67. Lowicki EM, Littlefield JB. An experimental method of precisely defining the dimensions of the gastric antrum. Surg Forum 12:308-309, 1961. [PubMed: 14466962]

68. Moe RE, Klopper PJ, Nyhus LM. Demonstration of the functional anatomy of the common gastric antrum. Am J Surg 111:80, 1966. [PubMed: 4159040]

69. McCrea E D’A. The abdominal distribution of the vagus. J Anat 59:18, 1924.

70. Skandalakis JE, Gray SW, Soria RE, Sorg JL, Rowe JS Jr. Distribution of the vagus nerve to the stomach. Am Surg 46(3):130-139, 1980.

71. Skandalakis JE, Rowe JS, Gray SW, Androulakis JA. Identification of vagal structures at the esophageal hiatus. Surgery 75(2):233-237, 1974.

72. Skandalakis LJ, Gray SW, Skandalakis JE. The history and surgical anatomy of the vagus nerve. Surg Gynecol Obstet 162:75-85, 1986. [PubMed: 3940410]

73. Nielsen HO, Muñoz JD, Kronborg O. The antrum in duodenal ulcer patients: relationship between antrum size, nerve of Latarjet, gastrin cell quantity, and gastric acid secretion. Scand J Gastroenterol 16:491, 1981. [PubMed: 7323687]

74. Poppen B, Delin A, Sandstedt B. Parietal cell vagotomy: localization of the microscopical antral-fundic boundary in relation to the macroscopical. Acta Chir Scand 142:251, 1976. [PubMed: 936955]

75. Popiela T, Turczynowski W. Letter to editor. Am J Surg 156:227, 1988. [PubMed: 3421430]

76. Donahue PE, Nyhus LM. Letter to the editor. Am J Surg 156: 227-228, 1988.

77. Naik KS, Lagopoulos M, Primrose JN. Distribution of antral G cells in relation to the parietal cells of the stomach and anatomical boundaries. Clin Anat 3:17-24, 1990.

78. Donahue PE, Bombeck T, Condon RE, Nyhus LM. Proximal gastric vagotomy versus selective vagotomy with antrectomy. Surgery 96:585-590, 1984. [PubMed: 6385314]

79. Rosati I, Serantoni C, Ciani PA. Extended selective proximal vagotomy. Observations on a variant in technique. Chir Gastroenterol 10:33-37, 1976.

80. Donovan AJ, Berne TV, Donovan JA. Perforated duodenal ulcer: An alternative therapeutic plan. Arch Surg 1998;133:1166-1171. [PubMed: 9820345]

81. Kuzin NM, Alimov AN. [Selective proximal vagotomy with duodenoplasty in the treatment of duodenal stenosis in patients with duodenal ulcer]. [Russian] Khirurgiia 1997;4:38-43. [PubMed: 9297001]

82. Chang TM, Chen TH, Shih CM, Gueng MK, Tsou SS. Partial or complete circular duodenectomy with highly selective vagotomy for severe obstructing duodenal ulcer disease: an initial experience. Arch Surg 1998;133:998-1001. [PubMed: 9749855]

83. Uravic M, Stimac D, Rubinic M, Kovac D, Zilli M, Petrosic N, Prica M. Duodenal obstruction from chronic pancreatitis. Minerva Chir 1997;52:885-89. [PubMed: 9354077]

84. Lu CC, Schulze-Delrieu K. Pyloric deformation from peptic disease. Radiographic evidence for incompetence rather than obstruction. Dig Dis Sci 1990;35:1459-1467. [PubMed: 2253530]

85. Portis S, Jaffe R. Study of peptic ulcer based on necropsy records. JAMA 110:6, 1938.

86. Silen W. Surgical anatomy of the pancreas. Surg Clin North Am 44:1253, 1964.

87. Hurwitz A, Brady DA, Schaal E, Samloff IM, Dedon J, Ruhl CE. Gastric acidity in older adults. JAMA 278(8):659-662, 1997.

88. Sadchikov VD, Galakhin KA. Cancer from an ulcer or ulcerative stomach cancer? (a review of the literature and the authors’ own observations) [Russian]. Lik Sprava (2):137-141, 1997.

89. LaVecchia C. Braga C, Negri E, Franceschi S. Risk of stomach cancer in patients with gastric or duodenal ulcer. Eur J Cancer Prev 6(1):20-23, 1997.

90. Wastell C. Chronic duodenal ulcer. In: Nyhus LM, Wastell C (eds). Surgery of the Stomach and Duodenum, 4th Ed. Boston: Little, Brown and Co, 1986, p. 302.

91. Wastell C, Baron JH. Chronic duodenal ulcer and gastric secretion tests. In: Wastell C, Nyhus LM, Donahue PE (eds). Surgery of the Esophagus, Stomach, and Small Intestine, 5th Ed, Boston: Little, Brown and Co, 1995, p. 456.

92. Breuer T, Malaty HM, Graham DY. The epidemiology of H. pylori-associated gastroduodenal diseases. In: Ernst PB, Michetti P, Smith PD, eds. The Immunobiology of H. pylori: From Pathogenesis to Prevention. Philadelphia: Lippincott-Raven, 1997.

93. Blecker U. Helicobacter pylori-associated gastroduodenal disease in childhood. South Med J 90(6):570-576, 1997.

94. Wallace JL. The role of nerve-immune interactions in the pathogenesis of peptic ulcer. In: Ernst PB, Michetti P, Smith PD, eds. The Immunobiology of H. pylori: From Pathogenesis to Prevention. Philadelphia: Lippincott-Raven, 1997.

95. Blum AL. An historical overview of helicobacter-associated disorders. In Ernst PB, Michetti P, Smith PD, eds. The Immunobiology of H. pylori: From Pathogenesis to Prevention. Philadelphia: Lippincott-Raven, 1997.

96. Isaacson PG, Wright DH. Malignant lymphoma of mucosa-associated lymphatic tissue: a distinctive type of B cell lymphoma. Cancer 1983;52:1410-1416. [PubMed: 6193858]

97. Stephens J, Smith J. Treatment of primary gastric lymphoma and gastric mucosa-associated lymphoid tissue lymphoma. J Am Coll Surg 187(3):312-320, 1998.

98. Yamamoto S, Yamasaki Y, Kuwata K, Yamasaki H, Nishida Y, Kobayashi Y. Role of Helicobacter pylori in residual gastritis after distal partial gastrectomy. World J Surg 22:28-34, 1998. [PubMed: 9465758]

99. Ota H, Genta RM. Morphological characterization of the gastric mucosa during infection with H. pylori. In: Ernst PB, Michetti P, Smith PD, eds. The Immunobiology of H. pylori: From Pathogenesis to Prevention. Philadelphia: Lippincott-Raven, 1997.

100. Xia HH, Kalantar JS, Talley NJ, Wyatte JM, Adams S, Chueng K, Mitchell HM. Antral-type mucosa in the gastric incisura, body, and fundus (antralization): a link between Helicobacter pylori infection and intestinal metaplasia? Am J Gastroenterol 2000;95:114-121. [PubMed: 10638568]

101. Ernst PB, Michetti P, Smith PD (eds). The Immunobiology of H. pylori: From Pathogenesis to Prevention. Philadelphia: Lippincott-Raven, 1997.

102. Sontag SJ. Guilty as charged: bugs and drugs in gastric ulcer. Am J Gastroenterol 92(8):1255-1261, 1997.

103. Feldman M, Cryer B, Lee E, Peterson WL. Role of seroconversion in confirming cure of Helicobacter pylori infection. JAMA 280:363-365, 1998. [PubMed: 9686554]

104. Hsing AW, Hansson LE, McLaughlin JK, Nyren O, Blot WJ, Ekbom A, Fraumeni JF Jr. Pernicious anemia and subsequent cancer. A population-based cohort study. Cancer 71(3):745-750, 1993.

105. el-Newihi HM, Dellinger GW, Mihas AA, Achord JL. Gastric cancer and pernicious anemia appearing as pseudoachalasia. South Med J 89(9):906-910, 1996.

106. Sculco D, Bilgrami S. Pernicious anemia and gastric carcinoid tumor: case report and review. Am J Gastroenterol 92(8):1378-1380, 1997.

107. Becker HD, Gabriel A. Therapy of carcinoids of the stomach. [German]. Langenbecks Arch Chir 381(1):18-22, 1996.

108. El-Eishi HI, Ayoub SF, Adb-El-Khalek M. The arterial supply of the human stomach. Acta Anat 86:565-80, 1973. [PubMed: 4785693]

109. Yildirim M, Ozan H, Kutoglu T. Left gastric artery originating directly from the aorta. Surg Radiol Anat 20:303-305, 1998. [PubMed: 9787400]

110. Swigart LVL, Siekert RG, Hambley WC, Anson BJ. The esophageal arteries. An anatomic study of 150 specimens. Surg Gynecol Obstet 90:234-243, 1950. [PubMed: 15401971]

111. Van Damme JPJ, Bonte J. The blood supply of the stomach. Acta Anat 131:89-96, 1988.

112. Eckmann VI, Krahn V. Untersuchung zur Haufigkeit verschiedener Ursprungsstellen der Arteria gastrica dextra. Anat Anz [Jena] 155:65-70, 1984. [PubMed: 6721206]

113. Lippert H, Pabst R. Arterial Variations in Man. Munich: JF Bergmann Velag, 1985.

114. Wilkie DPD. The blood supply of the duodenum, with special reference to the supraduodenal artery. Surg Gynecol Obstet 13: 399, 1911.

115. Reeves TB. A study of the arteries supplying the stomach and duodenum and their relation to ulcer. Surg Gynecol Obstet 30: 374, 1920.

116. Barlow TE, Bentley FH, Walder DN. Arteries, veins, and arteriovenous anastomoses in the human stomach. Surg Gynecol Obstet 93:657, 1951. [PubMed: 14893072]

117. Nakayama K. Simplification of the Billroth I gastric resection. Surgery 35(6):837-843, 1954.

118. Nakayama K. Radical operations for carcinoma of the esophagus and cardiac end of the stomach. J Int Coll Surg 21(1):51-66, 1954.

119. Womack NA. Special comment: gastric arteriovenous communications and their significance. In: Harkins HN, Nyhus LM. Surgery of the Stomach and Duodenum, 2nd ed. Boston: Little, Brown and Co, 1969, pp. 45-48.

120. Konstantakos AK, Coogan SM, Husni EA, Raaf JH. Aneurysm of the gastroduodenal artery: an unusual cause of obstructive jaundice. Am Surg 2000;66:695-698. [PubMed: 10917486]

121. Sakamoto H, Akita K, Sato T. An anomalous right gastroepiploic artery arising from the superior mesenteric artery. Surg Radiol Anat 1999;21:283-286. [PubMed: 10549087]

122. Walther A. De vena porta. Lipsiae, 1740. Cited by DiDio LJ, Christoforidis AJ, Chandnani PC. Posterior gastric artery and its significance as seen in angiograms. Am J Surg 139:333, 1980.

123. Walther A. De arteriae coelicae tabula. Lipisiae titius, 1729.

124. DiDio LJ, Christoforidis AJ, Chandnani PC. Posterior gastric artery and its significance as seen in angiograms. Am J Surg 139:333, 1980. [PubMed: 7362002]

125. Trubel W, Rokitansky A, Turkof E, Firbas W. Correlations between posterior gastric artery and superior polar artery in human anatomy. Anat Anz [Jena] 167:219-223, 1988. [PubMed: 3202326]

126. Berens AS, Aluisio FV, Colborn GL, Gray SW, Skandalakis JE. The incidence and significance of the posterior gastric artery in human anatomy. J Med Assoc Ga 80:425-428, 1991. [PubMed: 1940694]

127. Trubel W, Rokitansky A, Turkof E, Firbas W. Incidence, anatomy and territories supplied by the posterior gastric artery. Acta Anat 124:26-30, 1985. [PubMed: 4072606]

128. Kaminski DL, Codd JE, Sigmund MD. Evaluation of the use of the median arcuate ligament in fundoplication for reflux esophagitis. Am J Surg 134:724, 1977. [PubMed: 596535]

129. Lindner HH, Kemprud E. A clinicoanatomical study of the arcuate ligament of the diaphragm. Arch Surg 103:600, 1971. [PubMed: 5117015]

130. Evans HM. The development of the vascular system. In: Kibel F, Mall FP (eds). Human Embryology. Philadelphia: Lippincott, 1912.

131. George R. Topography of the unpaired visceral branches of the abdominal aorta. J Anat 69:196, 1934.

132. Carey JP, Stemmer EA, Connolly JE. Median arcuate ligament syndrome. Arch Surg 1969;99:441. [PubMed: 5808272]

133. Lawson JD, Ochsner JL. Median arcuate ligament syndrome with severe two-vessel involvement. Arch Surg 1984;119:226. [PubMed: 6696614]

134. Reilly LM, Ammar AD, Stoney RJ, Ehrenfeld WK. Late results following operative repair for celiac artery compression syndrome. J Vasc Surg 1985;2:79-91.

135. Taylor DC, Moneta GL, Cramer MM. Extrinsic compression of the celiac artery by the median arcuate ligament of the diaphragm. J Vasc Technol 1987;11:236.

136. Gautier R, Barrie J, Sarrazin R. Les angors abdominaux non atheromateux. A propos de deux cas. Lyon Chir 61:893, 1965. [PubMed: 5859386]

137. Cavdar S, Sehirili U, Pekin B. Celiacomesenteric trunk. Clin Anat 1997;10:231-234. [PubMed: 9213038]

138. Marable SA, Kaplan MF, Bemau FM, Molnar W. Celiac compression syndrome. Am J Surg 115:97, 1968. [PubMed: 5634682]

139. Lord RSA, Stoney RJ, Wylie EJ. Coeliac-axis compression. Lancet 2:795, 1968. [PubMed: 4175602]

140. Hill LD. An effective operation for hiatal hernia: an eight year appraisal . Ann Surg 166:681, 1967. [PubMed: 6061546]

141. Menguy R. Esophageal hiatal hernia. Am J Surg 113:91, 1967.

142. Hill LD, Tobias J, Morgan EH. Newer concepts of the pathophysiology of hiatal hernia and esophagitis. Am J Surg 111:70, 1966. [PubMed: 5901379]

143. Adams HD, Lobb AW. Esophagoaortal hiatus hernia. N Engl J Med 250:143, 1954. [PubMed: 13119863]

144. Hayward J. Sliding esophageal hiatus hernia. In: Nyhus LM, Harkins HN (eds). Hernia. Philadelphia: Lippincott, 1964.

145. Gray SW, Rowe JS Jr, Skandalakis JE. Surgical anatomy of the gastroesophageal junction. Am Surg 45(9):575-587, 1979.

146. Takach TJ, Livesay JJ, Reul GJ, Cooley DA. Celiac compression syndrome: tailored therapy based on intraoperative findings. J Am Coll Surg 183:606-610, 1996. [PubMed: 8957463]

147. Walder DN. Arteriovenous anastomoses of human stomach. Clin Sci 11:59, 1952. [PubMed: 14896518]

148. Rutter AG. Ischemic necrosis of the stomach following subtotal gastrectomy. Lancet 2:1021, 1953.

149. Spencer FC. Ischemic necrosis of remaining stomach following subtotal gastrectomy. Arch Surg 73:844, 1956. [PubMed: 13361748]

150. Fell SC, Seidenberg B, Hurwitt ES. Ischemic necrosis of the gastric remnant: an uncommon complication of radical subtotal gastrectomy. Surgery 43:490, 1958. [PubMed: 13647176]

151. Brown JR, Derr JW. Arterial blood supply of human stomach. Arch Surg 64:616-621, 1952. [PubMed: 14923082]

152. Peters RM, Womack NA. Hemodynamics of gastric secretion. Ann Surg 148:537, 1958. [PubMed: 13583928]

153. Greig HW, Anson BJ, Coleman SS. The inferior phrenic artery: types of origin in 850 body halves and diaphragmatic relationship. Quart Bull Northwestern Univ M Sch 25:345, 1951. Cited by Hollinshead WH. Anatomy for Surgeons: Volume 2. The Thorax, Abdomen and Pelvis (2nd ed). New York: Harper and Row, 1971.

154. Akkinis AJ. J Anat Paris 64:200. Cited in McGregor AL, Du Plessis DJ. A Synopsis of Surgical Anatomy (10th ed). Baltimore: Williams & Wilkins, 1969, p. 66.

155. Mulholland MW. Duodenal ulcer. In: Greenfield LJ (ed). Surgery: Scientific Principles and Practice (2nd ed). Philadelphia: Lippincott-Raven, 1997, pp. 759-772.

156. Rittenhouse M, McFee AS, Aust JB. Gastric devascularization: an alternate approach to the surgical treatment of massive diffuse hemorrhage from gastritis. South Med J 69:892, 1976. [PubMed: 1085036]

157. Richardson JD, Aust JB. Gastric devascularization: a useful salvage procedure for massive hemorrhagic gastritis. Ann Surg 185: 649, 1977. [PubMed: 301014]

158. Kelessis NG, Vassilopoulos PP, Lambrinakis PM, Zissis CG, Efremidou AI. Treatment-related acute gastric bleeding managed successfully with surgical devascularization. J Surg Oncol (Suppl) 2000;74:138-140. [PubMed: 10914824]

159. Schlein M, Saadia R. Postoperative gastric ischaemia. Br J Surg 76:844-848, 1989.

160. Jacobsen JD. The circulation of the stomach. Gastroenterology 48:85-110, 1965.

161. Isabella V, Marotta E, Bianchi F. Ischemic necrosis of proximal gastric remnant following subtotal gastrectomy with splenectomy. J Surg Oncol 25:124-132, 1984. [PubMed: 6420621]

162. Casey KM, Quigley TM, Kozarek RA, Raker EJ. Lethal nature of ischemic gastropathy. Am J Surg 165:646-649, 1993. [PubMed: 8488953]

163. Costantino MJ. Ruptured gastric artery aneurysm complicating pneumococcal sepsis: a case report and review. J Med Assoc Ga 79(9):677-678, 1990.

164. Witte JT, Hasson JE, Harms BA, Corrigan TE, Love RB. Fatal gastric artery dissection and rupture occurring as a paraesophageal mass: a case report and literature review. Surgery 107(5): 590-594, 1990.

165. Salam AA, Warren WD. Anatomic basis of the surgical treatment of portal hypertension. Surg Clin North Am 1974;54:1247-1257. [PubMed: 4610845]

166. Purcell HK, Conner JJ, Alexander WF, Scully NM. Observations on the major radicles of the extrahepatic portal systems. AMA Arch Surg 62:670, 1951. [PubMed: 14818540]

167. Douglass BE, Baggenstoss AH, Hollinshead WH. The anatomy of the portal vein and its tributaries. Surg Gynecol Obstet 1950; 91:562. [PubMed: 14787850]

168. Hollinshead WH. Anatomy for Surgeons: Vol 2. New York: Hoeber-Harper, 1956.

169. Takayasu K, Aoki K, Ichikawa T, Onmura T, Sekiguchi R, Terauchi T, Takayama T. Aberrant right gastric vein directly communicating with left portal vein system. Acta Radiol 31:575-577, 1990. [PubMed: 2278780]

170. Miyaki T, Yamada M, Kumaki K. Aberrant course of the left gastric vein in the human. Acta Anat 130:275-279, 1987. [PubMed: 3434179]

171. Voiglio EJ, Boutillier du Retail C, Neidhardt JPH, Caillot JL, Barale F, Mertens P. Gastrocolic vein: Definition and report of two cases of avulsion. Surg Radiol Anat 1998;20:197-201. [PubMed: 9706679]

172. Coller FA, Kay EB, McIntyre RS. Regional lymphatic metastases of carcinoma of the stomach. Arch Surg 43:748, 1941.

173. Garlock JH. The problem of carcinoma of the cardiac end of the stomach. Surg Gynecol Obstet 73:244, 1941.

174. Berry REL, Rottschafer W. The lymphatic spread of cancer of the stomach observed in operative specimens removed by radical surgery including total pancreatectomy. Surg Gynecol Obstet 104: 269, 1957. [PubMed: 13422144]

175. Weinberg J, Greaney EM. Identification of regional lymph nodes by means of a vital staining dye during surgery of gastric cancer. Surg Gynecol Obstet 90:561, 1950. [PubMed: 15418425]

176. Mulholland MW. Gastric anatomy and physiology. In Greenfield LJ. Surgery: Scientific Principles and Practice (2nd ed). Philadelphia: JB Lippincott, 1997, pp. 745-759.

177. Harkins HN, Nyhus LM. Surgery of the Stomach and Duodenum, 2nd ed. Boston: Little, Brown, 1969.

178. Balfe DM, Mauro MA, Koehler RE, Lee JK, Weyman PJ, Picus D, Peterson RR. Gastrohepatic ligament: Normal and pathologic CT anatomy. Radiology 1984;150:485-490. [PubMed: 6691106]

179. Rosenberg JC, Schwade JG, Vaitkevicius VK. Cancer of the esophagus. In: DeVita VT Jr, Hellman S, Rosenberg SA (eds). Cancer: Principles and Practice of Oncology. Philadelphia: JB Lippincott, 1982, p. 504.

180. Guernsey JM, Knudsen DF. Abdominal exploration in the evaluation of patients with carcinoma of the thoracic esophagus. J Thorac Cardiovasc Surg 59:62, 1970. [PubMed: 4982937]

181. Jaehne J, Meyer HJ, Maschek H, Geerlings H, Bruns E, Pichlmayr R. Lymphadenectomy in gastric carcinoma. Arch Surg 127: 290-294, 1992. [PubMed: 1372495]

182. Smith JW, Shiu MH, Kelsey L, Brennan MF. Morbidity of radical lymphadenectomy in the curative resection of gastric carcinoma. Arch Surg 126:1469-1473, 1991. [PubMed: 1842175]

183. Gilbertson VA. Results of treatment of stomach cancer: An appraisal of efforts for more extensive surgery and a report of 1983 cases. Cancer 1969;23:1305-1308.

184. Dent DM, Madden MV, Price SK. Randomized comparison of R1 and R2 gastrectomy for gastric carcinoma. Br J Surg 1988;75: 110-112. [PubMed: 3349293]

185. Bonenkamp JJ, Songun I, Hermans J, Sasako M, Welvaart K, Plukker JTM, van Elk P, Obertop H, Gouma DJ, Taat CW, van Lanshot J, Meyer S, de Graff PW, von Meyenfeldt MF, Tilanus H, van de Velde CJH. Randomized comparison of morbidity after D1 and D2 dissection for gastric cancer in 996 Dutch patients. Lancet 1995;345:745-748. [PubMed: 7891484]

186. Cuschieri A, Fayers P, Fielding J, Craven J, Bancewicz J, Joypaul V, Cook P, for the Surgical Cooperative Group. Postoperative morbidity and mortality after D1 and D2 resections for gastric cancer: Preliminary results of the MRC randomized controlled trial. Lancet 1996;347:995-999. [PubMed: 8606613]

187. Iriyama K, Asakawa T, Koike H, Nishiwaki H, Suzuki H. Is extensive lymphadenectomy necessary for surgical treatment of intramucosal carcinoma of the stomach? Arch Surg 124:309-311, 1989. [PubMed: 2919964]

188. Otsuji E, Toma A, Kobayashi S, Okamoto K, Hagiwara A, Yamagishi H. Outcome of prophylactic radical lymphadenectomy with gastrectomy in patients with early gastric carcinoma without lymph node metastasis. Cancer 2000;89:1425-1430. [PubMed: 11013354]

189. Zinninger MM, Collins WT. Extension of carcinoma of the stomach into the duodenum and esophagus. Ann Surg 1949;130: 557.

190. Marvin CP. The intraduodenal spread of malignant gastric lesions [thesis]. Rochester MN: Mayo Foundation, University of Minnesota Graduate School, 1947.

191. de Manzoni G, Verlato G, Guglielmi A, Laterza E, Tomezzoli A, Pelosi G, Di Leo A, Cordiano C. Classification of lymph node metastases from carcinoma of the stomach: Comparison of the old (1987) and new (1997) TNM systems. World J Surg 1999;23: 664-669.

192. Boku T, Nakane Y, Okusa T, Hirozane N, Imabayashi N, Hioki K, Yamamoto M. Strategy for lymphadenectomy of gastric cancer. Surgery 1989;105:585-592. [PubMed: 2705096]

193. Shchepotin I, Evans SRT, Shabahang M, Cherny V, Buras RR, Zadorozhny A, Nauta RJ. Radical treatment of locally recurrent gastric cancer. Am Surg 61(4):371-376, 1995.

194. Jakl RJ, Miholic J, Koller R, Markis E, Wolner E. Prognostic factors in adenocarcinoma of the cardia. Am J Surg 169:316, 1995. [PubMed: 7879834]

195. Adachi Y, Oshiro T, Okuyama T, Kamakura T, Mori M, Maehara Y, Sugimachi K. A simple classification of lymph node level in gastric carcinoma. Am J Surg 169:382-385, 1995. [PubMed: 7694976]

196. Namieno T, Koito K, Higashi T, Sato N, Uchino J. General pattern of lymph node metastasis in early gastric carcinoma. World J Surg 20:996-1000, 1996. [PubMed: 8798355]

197. Mak K, Hui P, Chan W, Leung K. Mucosal lymphangiectasia in gastric adenocarcinoma. Arch Surg 131:1135, 1996.

198. Kodera Y, Yamamura Y, Torii A, Uesaka K, Hirai T, Yasui K, Morimoto T, Kato T, Kito T. Surgical treatment of Borrmann type IV gastric carcinoma: relevance of lymphadenectomy in improving survival. J Am Coll Surg 183:480-485, 1996. [PubMed: 8912617]

199. Kodera Y, Yamamura Y, Shimizu Y, Torii A, Hirai T, Yasui K, Morimoto T, Kato T, Kito T. The number of metastatic lymph nodes: A promising prognostic determinant for gastric carcinoma in the latest edition of the TNM classification. J Am Coll Surg 1998;187:597-603. [PubMed: 9849732]

200. Gabella G (ed). Cardiovascular system. In: Williams PL (ed). Gray’s Anatomy, 38th Ed. New York: Churchill Livingstone, 1995, p. 1619.

201. Visalli JA, Grimes OF. An embryologic and anatomic approach to the treatment of gastric cancer. Surg Gynecol Obstet 103:401, 1956. [PubMed: 13360646]

202. Grimes OF, Visalli JA. The embryologic approach to the surgical management of carcinoma of the upper stomach. Surg Clin North Am 44:1227, 1964.

203. Sakaguchi T, Watanabe A, Sawada H, Yamada Y, Tatsumi M, Fujimoto H, Emoto K, Nakano H. Characteristics and clinical outcome of proximal-third gastric cancer. J Am Coll Surg 187: 352-357, 1998. [PubMed: 9783780]

204. Ellis FH Jr, Williamson WA, Heatley GJ. Cancer of the esophagus and cardia: does age influence treatment selection and surgical outcomes? J Am Coll Surg 187:345-351, 1998. [PubMed: 9783779]

205. Virchow R. Zur Diagnose der Krebse im Unterleibe. Die Medicinische Reform, 45:248, 1849.

206. Skandalakis JE, Godwin JT, Androulakis JA, Gray SW. The differential diagnosis of tumors of the neck. In: Ariel IM (ed). Progress in Clinical Cancer, Vol IV. New York: Grune & Stratton, 1969, pp. 141-159.

207. Skandalakis JE, Gray SW, Takakis NC, Godwin JT, Poer DH. Tumors of the neck. Surgery 48(2):375-384, 1960.

208. Weiss L, Gilbert HA, Ballon SC (eds). Lymphatic System Metastasis. Boston: GK Hall Medical Publishers, 1980, p.126.

209. Kawaura Y. Radical operation for gastric cancer involving the esophagus. Surg Rounds 12(5):17-31, 1989.

210. Nishi M. Special comment: the present state of gastric cancer in Japan. In: Wastell C, Nyhus LM, Donahue PE. Surgery of the Esophagus, Stomach, and Small Intestine, 5th Ed. Boston: Little, Brown and Company, 1995, p. 605.

211. Roukos D, Encke A. Letter to the editor. World J Surg 20(1): 116-117, 1996.

212. Tanaka Y, Takao T, Watanabe H, Kido T, Ogawa N. Early stage gastric lymphoma: is operation essential? World J Surg 18:896, 1994. [PubMed: 7846915]

213. Moynihan BGA. The surgical treatment of cancer of sigmoid flexure and rectum. Surg Gynecol Obstet 6:463, 1908.

214. Jentschura D, Heubner C, Manegold BC, Rumstadt B, Winkler M, Trede M. Surgery for early gastric cancer: a European one-center experience. World J Surg 21:845-849, 1997. [PubMed: 9327676]

215. Wanebo HJ, Kennedy BJ, Winchester DP, Fremgen A, Stewart AK. Gastric carcinoma: does lymph node dissection alter survival? J Am Coll Surg 183:616-624, 1996. [PubMed: 8957465]

216. Namieno T, Koito K, Higashi T, Takahashi M, Yamashita K, Kondo Y. Assessing the suitability of gastric carcinoma for limited resection: endoscopic prediction of lymph node metastases. World J Surg 1998;22:859-864. [PubMed: 9673559]

217. Namieno T, Koito K, Higashi T, Shimamura T, Yamashita K, Sato, N, Kondo Y. Assessing the suitability of gastric carcinoma for limited resection: histologic differentiation of endoscopic biopsy. World J Surg 1998;22:865-868. [PubMed: 9673560]

218. Kodera Y, Yamamura Y, Shimuzu Y, Torii A, Hirai T, Yasui K, Morimoto T, Kato T. Adenocarcinoma of the gastroesophageal junction in Japan: relevance of Siewert’s classification applied to 177 cases resected at a single institution. J Am Coll Surg 1999; 189:594-601.

219. Siewert JR, Stein HJ. Carcinoma of the cardia: carcinoma of the gastroesophageal junction – classification, pathology and extent of resection. Dis Esophagus 1996;9:173-182.

220. Siewert JR, Stein HJ, Sendler A, Fink U. Surgical resection for cancer of the cardia. Sem Surg Oncol 1999;17:125-131. [PubMed: 10449684]

221. Elder JB, Deakin M. Gastric physiology. In: Wastell C, Nyhus LM, Donahue PE (eds). Surgery of the Esophagus, Stomach, and Small Intestine, 5th Ed. Boston: Little, Brown and Co, 1995, p. 426.

222. Waisbren SJ, Modlin IM. Lester R. Dragstedt and his role in the evolution of therapeutic vagotomy in the United States. Am J Surg 167:344-359, 1994. [PubMed: 8160911]

223. Dragstedt LR, Fournier HJ, Woodward ER, Tovee EB, Harper PV Jr. Transabdominal gastric vagotomy. A study of the anatomy and surgery of the vagus nerves at the lower portion of the esophagus. Surg Gynecol Obstet 85:461, 1947.

224. Griffith CA. A new anatomical approach to the problem of incomplete vagotomy. Surg Clin North Am 44:1239, 1964.

225. Weinberg JA, Campbell GW, Voelker RB. Vagus nerve section in the treatment of peptic ulcer. J Thorac Surg 17:743, 1948.

226. Legros G, Griffith CA. The anatomic pathways of the vagal fibers to the antral mucosa. Surgery 66:751, 1969. [PubMed: 5822762]

227. Loeweneck H. Functional anatomy of the vagus nerves in the upper abdomen. In: Holle F, Andersson S (eds). Vagotomy. New York: Springer-Verlag, 1974.

228. Payne JT. The significance of vascular landmarks in gastric resection. West J Surg 71:161, 1963. [PubMed: 14057470]

229. Grassi G. Highly selective vagotomy with intraoperative acid secretion test of completeness of vagal section. Surg Gynecol Obstet 140:259, 1975. [PubMed: 1124480]

230. Donahue PE, Yoshida J, Richter HM, Liu K, Bombeck CT, Nyhus LM. Proximal gastric vagotomy with drainage for obstructing duodenal ulcer. Surgery 104(4):757-764, 1988.

231. Donahue PE, Yoshida J, Richter HM, Bombeck CT, Nyhus LM, Maroske D, Thon KP, Roeher HD. Can the use of endoscopic Congo red test decrease the incidence of incomplete proximal gastric vagotomy? Gastrointest Endosc 33(6):427-431, 1987.

232. Donahue PE, Bombeck CT, Yoshida Y, Nyhus LM. Endoscopic Congo red test during proximal gastric vagotomy. Am J Surg 153: 249-255, 1987. [PubMed: 3826505]

233. Donahue PE, Griffith C, Richter HM. A 50-year perspective upon selective gastric vagotomy. Am J Surg 172:9-12, 1996. [PubMed: 8686811]

234. Bruckner WL. Gastric secretion and motility after highly selective sympathectomy. In: Holle F, Andersson S (eds). Vagotomy. New York: Springer-Verlag, 1974, p. 89.

235. Absolon KA. The surgical school of Theodor Billroth. Surgery 50(4):697-715, 1961.

236. Stevens J. In: McCullough D. The Path Between the Seas: The Creation of the Panama Canal, 1870-1914. New York: Simon and Schuster, 1977, p. 617.

237. Waugh JM, Hood RT Jr. Gastric operations: a historic review — I. Q Rev Surg Obstet Gynecol 10(4):201-214, 1953.

238. Teague WJ, Ackroyd R, Watson DI, Devitt PG. Changing patterns in the management of gastric volvulus over 14 years. Br J Surg 2000;87:358-361. [PubMed: 10718808]

239. Sharma D, Saxena A, Rahman H, Raina VK, Kapoor JP. “Free omental plug”: a nostalgic look at an old and dependable technique for giant peptic perforations. Dig Surg 2000;17:216-218. [PubMed: 10867452]

240. Blomgren LGM. Perforated peptic ulcer: long-term results after simple closure in the elderly. World J Surg 21:412-415, 1997. [PubMed: 9143574]

241. Wysocki A, Biesiada Z, Beben P, Budzynski A. Perforated gastric ulcer. Dig Surg 2000;17:132-137. [PubMed: 10781975]

242. Donahue PE, Nyhus LM. Exposure of the periesophageal space. Surg Gynecol Obstet 152:218-220, 1981.

243. Healey JE Jr, Schwartz SI. Surgical anatomy. In: Schwartz SI. Surgical Diseases of the Liver. New York: McGraw-Hill, 1964, p. 12.

244. Hemingway AP, Gibson RN. Radiological anatomy of the liver and biliary tract. In: Blumgart LH (ed). Surgery of the Liver and Biliary Tract. Edinburgh: Churchill Livingstone, 1988, p. 33.

245. Chassin JL. Operative Strategy in General Surgery, 2nd ed. New York: Springer-Verlag, 1994.

246. Griffith CA. Anatomic considerations in gastroduodenal surgery. In: Nyhus LM, Baker RJ, Fischer JE (eds). Mastery of Surgery, 3rd ed, Vol 1. Boston: Little, Brown and Co, 1997.

247. Ohgami M, Otani Y, Kumai K, Kubota T, Kim Y-I, Kitajima M. Curative laparoscopic surgery for early gastric cancer: five years experience. World J Surg 1999;23:187-193. [PubMed: 9880430]

248. Hayes N, Ng EKW, Raimes SA, Crofts TJ, Woods SDS, Griffin SM, Chung SCS. Total gastrectomy with extended lymphadenectomy for “curable” stomach cancer: experience in a non-Japanese Asian center. J Am Coll Surg 1999;188:27-32. [PubMed: 9915239]

249. Eguchi T, Takahashi Y, Yamagata M, Kasahara M, Fujii M. Gastric cancer in young patients. J Am Coll Surg 1999;188:22-26. [PubMed: 9915238]

250. Rodriguez-Sanjuán JC, Alvarez-Cañas C, Casado F, García-Castillo L, Casanova D, Val-Bernal F, Naranjo A. Results and prognostic factors in stage IE-IIE primary gastric lymphoma after gastrectomy. J Am Coll Surg 1999;188:296-303.

251. Schumpelick V, Farthmann E, Schreiber HW. Chirurgie des Magens: Historisches and Entwicklungstendenzen. Med Welt 48/50:3-87, 1976.

252. Ovaska J, Kruuna O, Saario I, Schröder T, Lempinen M. Surgical treatment of gastric carcinoma. Am J Surg 158:467-471, 1989. [PubMed: 2817231]

253. Kwon SJ, Members of the Korean Gastric Cancer Study Group. Prognostic impact of splenectomy on gastric cancer: results of the Korean Gastric Cancer Study Group. World J Surg 21:837-844, 1997. [PubMed: 9327675]

254. Schwarz A, Büchler M, Usinger K, Rieger H, Glasbrenner B, Friess H, Kunz R, Beger HG. Importance of the duodenal passage and pouch volume after total gastrectomy and reconstruction with the Ulm pouch: prospective randomized clinical study. World J Surg 20:60-67, 1996. [PubMed: 8588415]

255. Heimbucher J, Fuchs KH, Freys SM, Clark GWB, Incarbone R, DeMeester TR, Bremner CG, Thiede A. Motility in the Hunt-Lawrence pouch after total gastrectomy. Am J Surg 168:622-626, 1994. [PubMed: 7978007]

256. Schuricht AL, Gowen GF, Azurin DJ. Thoracoscopic vagotomy as a safe adjunct to remedial gastric surgery. Am Surg 63(6):540-542, 1997.

257. Donahue PE. Ulcer surgery and highly selective vagotomy – Y2K. Arch Surg 1999;134:1373-1377. [PubMed: 10593337]

258. Sapala MA, Sapala JA, Resto Soto AD, Bouwman DL. Cholelithiasis following subtotal gastric resection with truncal vagotomy. Surg Gynecol Obstet 1979;148:36. [PubMed: 758694]

259. Rehnberg O, Haglund U. Gallstone disease following antrectomy and gastroduodenectomy with or without vagotomy. Ann Surg 1984;201:315.

260. Lorusso D, Misciagna G, Noviello MR, Tarantino S. Cholelithiasis after Billroth II gastric resection. Surgery 1988;103:579. [PubMed: 3283982]

261. Cipollini F, Mecozzi V, Altilia F. Increased risk for gallstone disease in subjects operated on for partial gastrectomy with gastrojejunostomy (BII operation) Ital J Gastroenterol 1991;23: 351. [PubMed: 1742525]

262. Ise H, Kitayama O, Hayasaka H, Matsuno S. [Post-gastrectomy gallstone disease]. Jpn J Clin Med 1993;51:1875. [PubMed: 8366612]

263. Inoue K, Fuchigami A, Higashide S, Sumi S, Kogire M, Suzuki T, Tobe T. Gallbladder sludge and stone formation in relation to contractile function after gastrectomy. Ann Surg 1992;215:19. [PubMed: 1731646]

264. Deitel M. Jejunocolic and jejunoileal bypass: an historical perspective. In: Deitel M (ed). Surgery for the Morbidly Obese Patient. Philadelphia: Lea & Febiger, 1989, pp. 81-90.

265. Benotti PN, Forse RA. The role of gastric surgery in the multidisciplinary management of severe obesity. Am J Surg 169:361-367, 1995. [PubMed: 7879845]

266. Curry TK, Carter PL, Porter CA, Watts DM. Resectional gastric bypass is a new alternative in morbid obesity. Am J Surg 175:367-370, 1998. [PubMed: 9600279]

267. Nehoda H, Weiss H, Labeck B, Hourmont K, Lanthaler M, Oberwalder M, Aigner F. Results and complications after adjustable gastric banding in a series of 250 patients. Am J Surg 2001;181:12-15. [PubMed: 11248168]

268. Higa KD, Boone KB, Ho T, Davies OG. Laparoscopic Roux-en-Y gastric bypass for morbid obesity: technique and preliminary results of our first 400 patients. Arch Surg 2000;135:1029-1033. [PubMed: 10982506]

269. Spiro RH, Bains MS, Shah JP, Strong EW. Gastric transposition for head and neck cancer: a critical update. Am J Surg 162:348-352, 1991. [PubMed: 1951887]

270. Ruangtrakool R, Spitz L. Early complications of gastric transposition operation. J Med Assoc Thailand 2000;83:352-357. [PubMed: 10808693]

271. Whelan SL, Parkin DM, Masuyer E. Trends in cancer incidence and mortality. Lyon, France: IARC Scientific Publications, 1993.

272. Kimura H, Kanno M, Takamura H, Arakawa H, Maeda K, Uogishi M, Sodani H, Kewashima A. Case report of a patient with multiple lesions of the stomach, including multiple cancers and an adenomatous polyp. J Gastroenterol 30(1):108-112, 1995.

273. Evrensel T, Manavoglu O, Ozyardimci C, Orhan B, Gulten M, Nak SG, Yerci O. Subtypes of intestinal metaplasia and their relationship to Helicobacter pylori. J Environ Pathol Toxicol Oncol 15(2-4):221-224, 1996.

274. Evrensel T, Manavoglu O, Ozyardimci C, Gulten M, Nak SG, Yerci O. Helicobacter pylori and intestinal metaplasia. J Environ Pathol Toxicol Oncol 15(2-4):215-219, 1996.

275. Watanabe H. The pathological sequence in the development of gastric cancer: II. Scand J Gastroenterol Suppl 214:36-39, discussion 40-43, 1996.

276. McCloy RF, Arnold R, Bardhan KD, Cattan D, Klinkenberg-Knol E, Maton PN, Riddell RH, Sipponen P, Walan A. Pathophysiological effects of long-term acid suppression in man. Dig Dis Sci 40(2 Suppl):96S-120S, 1995.

277. Weston AP, Krmpotich PT, Cherian R, Dixon A, Topalovski M. Prospective evaluation of intestinal metaplasia and dysplasia within the cardia of patients with Barrett’s esophagus. Dig Dis Sci 42(3): 597-602, 1997.

278. Morales TG, Sampliner RE, Bhattacharyya A. Intestinal metaplasia of the gastric cardia. Am J Gastroenterol 92(3):414-418, 1997.

279. Simeone DM, Hassan A, Scheiman JM. Giant peptic ulcer: a surgical or medical disease? Surgery 1999;126:474-478. [PubMed: 10486598]

280. Hansson LE. Risk of stomach cancer in patients with peptic ulcer disease. World J Surg 2000;24:315-320. [PubMed: 10658066]

281. Donahue PE. Parietal cell vagotomy versus vagotomy-antrectomy: ulcer surgery in the modern era. World J Surg 2000;24: 264-269. [PubMed: 10658059]

282. Jordan PH Jr, Thornby J. Twenty years after parietal cell vagotomy or selective vagotomy-antrectomy for treatment of duodenal ulcer. Final report. Ann Surg 220(3):283-293; discussion 293-296, 1994.

283. Martin IG, Diament RH, Dixon MF, Axon AT, Johnston D. Helicobacter pylori and recurrent ulceration after highly selective vagotomy. Eur J Gastroenterol Hepatol 7(3):207-209, 1995.

284. Johnson AG. Proximal gastric vagotomy: does it have a place in the future management of peptic ulcer? World J Surg 2000;24: 259-263. [PubMed: 10658058]

285. Huang JQ, Hunt RH. pH, healing rate and symptom relief in acid-related diseases. Yale J Biol Med 69(2):159-174, 1996.

286. Khulusi S, Badve S, Patel P, Lloyd R, Marrero JM, Finlayson C, Mendall MA, Northfield TC. Pathogenesis of gastric metaplasia of the human duodenum: role of Helicobacter pylori, gastric acid, and ulceration. Gastroenterology 110(2):452-458, 1996.

287. Kikuchi S, Arai Y, Kobayashi N, Tsukamoto H, Shimao H, Sakakibara Y, Hiki Y, Kakita A. Is extended lymphadenectomy valuable in palliatively gastrectomized patients with gastric cancer and simultaneous peritoneal metastasis? Hepatogastroenterology 47(32):563-566, 2000.

288. Karpeh MS, Leon L, Klimstra D, Brennan MF. Lymph node staging in gastric cancer: is location more important than number? An analysis of 1,038 patients. Ann Surg 2000;232:362-371. [PubMed: 10973386]

289. Nurnberger HR, Awwad E, Lohlein D. [Is splenectomy “en principe” necessary for radical gastrectomy with systematic lymphadenectomy?] [German]. Zentralbl Chir 121(2):144-147, 1996.

290. Pierie JPEN, Choudry U, Muzikansky A, Yeap BY, Souba WW, Ott MJ. The effect of surgery and grade on outcome of gastrointestinal stromal tumors. Arch Surg 2001;136:383-389. [PubMed: 11296107]

291. Kasakura Y, Fujii M, Mochizuki F, Kochi M, Kaiga T. Is there a benefit of pancreaticosplenectomy with gastrectomy for advanced gastric cancer? Am J Surg 2000;179:237-242. [PubMed: 10827328]

292. Wanebo HJ, Kennedy BJ, Winchester DP, Stewart AK, Fremgen AM. Role of splenectomy in gastric cancer surgery: adverse effect of elective splenectomy on longterm survival. J Am Coll Surg 185(2):177-184, 1997.

293. Bosing NM, Goretzki PE, Roher HD. Gastric cancer: which patients benefit from systemic lymphadenectomy? Eur J Surg Oncol 2000;26:498-505. [PubMed: 11016473]

294. Tsujitani S, Katano K, Oka A, Ikeguchi M, Maeta M, Kaibara N. Limited operation for gastric cancer in the elderly. Br J Surg 83(6): 836-839, 1996.

295. Ishigami S, Natsugoe S, Saihara T, Hokita S, Tokushige M, Watanabe T, Iwaschige H, Aikou T. Clinical and pathologic features of early gastric cancer in elderly patients. Hepatogastroenterology 44(16):1164-1168, 1997.

296. Hundahl SA, Stemmermann GN, Oishi A. Racial factors cannot explain superior Japanese outcomes in stomach cancer. Arch Surg 131(2):170-175, 1996.

297. Roukos DH, Lorenz M, Encke A. Evidence of survival benefit of extended (D2) lymphadenectomy in Western patients with gastric cancer based on a new concept: a prospective long-term follow-up study. Surgery 123:573-578, 1998. [PubMed: 9591011]

298. Meyer HJ, Jähne J. Lymph node dissection for gastric cancer. Sem Surg Oncol 1999;17:117-124. [PubMed: 10449683]

299. Maehara Y, Tomoda M, Tomisaki S, Ohmori M, Baba H, Akazawa K, Sugimachi K. Surgical treatment and outcome for node-negative gastric cancer. Surgery 121:633-639, 1997. [PubMed: 9186463]

300. Wu CW, Hsieh MC, Lo SS, Lui WY, P’eng FK. Results of curative gastrectomy for carcinoma of the distal third of the stomach. J Am Coll Surg 183(3):201-207, 1996.

301. Chan WH, Wong WK, Khin LW, Chan HS, Soo KC. Significance of a positive oesophageal margin in stomach cancer. Aust N Z J Surg 2000;70:700-703. [PubMed: 11021482]

302. Rüdiger Siewert J, Feith M, Werner M, Stein HJ. Adenocarcinoma of the esophagogastric junction: results of surgical therapy based on anatomical/topographic classification in 1,002 consecutive patients. Ann Surg 2000;232:353-361. [PubMed: 16456907]

303. Kitamura K, Yamaguchi T, Taniguchi H, Hagiwara A., Sawai K, Takahashi T. Analysis of lymph node metastasis in early gastric cancer: rationale of limited surgery. J Surg Oncol 64(1):42-47, 1997.

304. Shimizu S, Uchiyama A, Mizumoto K, Morisaki T, Nakamura K, Shimura H, Tanaka M. Laparoscopically assisted distal gastrectomy for early gastric cancer: is it superior to open surgery? Surg Endosc 2000;14:27-31. [PubMed: 10653231]

305. Imada T, Rino Y, Takahashi M, Suzuki M, Tanaka J, Shiozawa M, Kabara K, Hatori S, Ito H, Yamamoto Y, Amano T. Postoperative functional evaluation of pylorus-preserving gastrectomy for early gastric cancer compared with conventional distal gastrectomy. Surgery 123:165-170, 1998. [PubMed: 9481402]

306. Nakane Y, Akehira K, Inoue K, Iiyama H, Sato M, Masuya Y, Okumura S, Yamamichi K, Hioki K. Postoperative evaluation of pylorus-preserving gastrectomy for early gastric cancer. Hepatogastroenterology 2000;47:590-595. [PubMed: 10791245]

307. Seto Y, Nagawa H, Muto T. Impact of lymph node metastasis on survival with early gastric cancer. World J Surg 21:186-190, 1997. [PubMed: 8995076]

308. Otsuji E, Yamaguchi T, Sawai K, Okamoto K, Takahashi T. End results of simultaneous pancreatectomy, splenectomy and total gastrectomy for patients with gastric carcinoma. Br J Cancer 75 (8):1219-1223, 1997.

309. Kyzer S, Binyamini Y, Melki Y, Ohana G, Koren R, Chaimoff C, Wolloch Y. Comparative study of the early postoperative course and complications in patients undergoing Billroth I and Billroth II gastrectomy. World J Surg 21: 763-767, 1997. [PubMed: 9276708]

310. Shibao K, Okamoto K, Nagata N, Kuroda Y, Itoh H, Ohsato K. [The clinicopathological features and operative results in cancer after partial gastrectomy]. [Japanese]. Sangyo Ika Daigaku Zasshi 19(2):147-156, 1997.

311. Schepotin IB, Chorny VA, Nauta RJ, Shabahang M, Buras RR, Evans SRT. Extended surgical resection in T4 gastric cancer. Am J Surg 175:123-126, 1998.

312. Graham AJ, Finley RJ, Clifton JC, Evans KG, Fradet G. Surgical management of adenocarcinoma of the cardia. Am J Surg 175: 418-421, 1998. [PubMed: 9600291]

313. Kodera Y, Yamamura Y, Shimizu Y, Torii A, Hirai T, Yasui K, Morimoto T, Kato T, Kito T. Lack of benefit of combined pancreaticosplenectomy in D2 resection for proximal-third gastric carcinoma. World J Surg 21(6):622-627, discussion 627-628, 1997.

314. Newman E, Brennan MF, Hochwald SN, Harrison LE, Karpeh MS. Gastric remnant carcinoma: just another proximal gastric cancer or a unique entity? Am J Surg 173:292-297, 1997. [PubMed: 9136783]

315. Lo SS, Wu CW, Hsieh MC, Lui WY. Is gastric remnant cancer clinically different from primary gastric cancer? Hepatogastroenterology 44(13):299-301, 1997.

316. Lo SS, Tsay SH, Wu CW, Hsieh MC, Li AF, Lui WY. Intestinal-type tumour in resected gastric remnant cancer. J Gastroenterol Hepatol 1997;12:434-436. [PubMed: 9195400]

317. Herrington JL Jr, Sawyers JL. Remedial operations. In: Wastell C, Nyhus LM, Donahue PE. Surgery of the Esophagus, Stomach, and Small Intestine, 5th Ed. Boston: Little, Brown and Co., 1995.

318. Farrands PA, Blake JRS, Ansell ID. Endoscopic review of patients who have had gastric surgery. Br Med J 286:755, 1983. [PubMed: 6402234]

319. Nagamachi Y. [Pedicle transplantation of a transverse colon segment after total gastrectomy.] [Japanese] Nippon Geka Gakkai Zasshi. J Jpn Surg Soc 1997;98:542-548. [PubMed: 9255805]

320. Tsujitani S, Oka S, Saito H, Kondo A, Ikeguchi M, Maeta M, Kaibara N. Less invasive surgery for early gastric cancer based on the low probability of lymph node metastasis. Surgery 1999;125:148-154. [PubMed: 10026747]

321. Greene FL. Management of gastric remnant carcinoma based on the results of a 15-year endoscopic screening program. Ann Surg 223(6):701-706; discussion 706-708, 1996.

322. Sano T, Sasako M, Kinoshita T, Katai H, Maruyama K, Takenaka T, Nakanishi Y. Total gastrectomy for primary gastric lymphoma at stages IE and IIE: a prospective study of fifty cases. Surgery 121(5):501-505, 1997.

323. Vaillant JC, Ruskone-Fourmestraux A, Aegerter P, Gayet B, Rambaud JC, Valleur P, Parc R. Management and long-term results of surgery for localized gastric lymphomas. Am J Surg 2000;179:216-222. [PubMed: 10827324]

324. Mulholland MW. Gastric neoplasms. In: Greenfield LJ (ed). Surgery: Scientific Principles and Practice (2nd ed). Philadelphia: Lippincott-Raven, 1997, pp. 795-804.

325. Schindl M, Kaserer K, Niederle B. Treatment of gastric neuroendocrine tumors. Arch Surg 2001;136:49-54. [PubMed: 11146777]

326. Hepworth CC, Menzies D, Motson RW. Minimally invasive surgery for posterior gastric stromal tumors. Surg Endosc 2000;14:349-353. [PubMed: 10790553]

327. Basso N, Rosato P, De Leo A, Picconi T, Trentino P, Fantini A, Silecchia G. Laparoscopic treatment of gastric stromal tumors. Surg Endosc 2000;14:524-526. [PubMed: 10890957]

328. Harrison LE, Karpeh MS, Brennan MF. Total gastrectomy is not necessary for proximal gastric cancer. Surgery 123:127-130, 1998. [PubMed: 9481396]

329. Pickleman J, Watson W, Cunningham J, Fisher SG, Gamelli R. The failed gastrointestinal anastomosis: an inevitable catastrophe? J Am Coll Surg 1999;188:473-482. [PubMed: 10235574]

330. Abuksis G, Mor M, Segal N, Shemesh I, Plout S, Sulkes J, Fraser GM, Niv Y. Percutaneous endoscopic gastrostomy: high mortality rates in hospitalized patients. Am J Gastroenterol 2000;95:3.

331. Cosentini EP, Sautner T, Gnant M, Winkelbauer F, Teleky B, Jakesz R. Outcomes of surgical, percutaneous endoscopic, and percutaneous radiologic gastrostomies. Arch Surg 1998;133: 1076-1083. [PubMed: 9790204]

332. Shahbani DK, Goldberg R. Peritonitis after gastrostomy tube replacement in the emergency department. J Emerg Med 2000; 18:45-46.

333. Kobak GE, McClenathan DT, Schurman SJ. Complications of removing percutaneous endoscopic gastrostomy tubes in children. J Pediatr Gastroenterol Nutr 2000;30:404-407. [PubMed: 10776951]

334. Hulka F, Harrison MW, Campbell TJ, Campbell JR. Complications of pyloromyotomy for infantile hypertrophic pyloric stenosis. Am J Surg 173:450-452, 1997. [PubMed: 9168087]

335. Gasparri MG, Pipinos II, Kralovich KA, Margolin DA. Retrograde jejunogastric intussusception. South Med J 2000;93:499-500. [PubMed: 10832950]

336. Fromm D. Complications of Gastric Surgery. New York: John Wiley & Sons, 1977, p. 110.

337. Zittel TT, Zeeb B, Maier GW, Kaiser GW, Zwirner M, Liebich H, Starlinger M, Becker HD. High prevalence of bone disorders after gastrectomy. Am J Surg 174:431-438, 1997. [PubMed: 9337169]

338. Golden R. Functional obstruction of efferent loop of jejunum following partial gastrectomy. JAMA 148:721, 1952. [PubMed: 14897633]

339. Shiino Y, Filipi CJ, Hinder RA, Tomonaga T, Awad ZT, Marsh RE. Persistent dysphagia after laparoscopic vagotomy. Surg Endosc 2000;14:330-335. [PubMed: 10790549]

340. Sugerman H. Morbid obesity. In: Greenfield LJ, Mulholland MW, Oldham KT, Zelenock GB. Surgery: Scientific Principles and Practice. Philadelphia: JB Lippincott, 1993, p. 707.

341. Weiss HG, Nehoda H, Labeck B, Peer-Kühberger R, Klingler P, Gadenstätter A, Aigner F, Wetscher GJ. Treatment of morbid obesity with laparoscopic adjustable gastric banding affects esophageal motility. Am J Surg 2000;180:479-482. [PubMed: 11182402]

342. Chareton B, Landen S, Manganas D, Meunier B, Launois B. Prospective randomized trial comparing Billroth I and Billroth II procedures for carcinoma of the gastric antrum. J Am Coll Surg 183:190-194, 1996. [PubMed: 8784310]

343. Markowitz AM. Retroanastomotic hernia. In: Nyhus LM, Harkins HN. Hernia. Philadelphia: JB Lippincott, 1964, p. 607.

344. Henley FA. Gastrectomy with replacement: A preliminary communication. Br J Surg 1952;40:118. [PubMed: 12987641]

345. Henley FA. Gastrectomy with replacement. Ann R Coll Surg Engl 1953;13:141. [PubMed: 13080952]

346. Hays RP. Anatomic and physiologic reconstruction following total gastrectomy by the use of a jejunal food pouch. Surg Forum 1953;4:291. [PubMed: 13187286]

347. Shackman R. Jejunogastric intussusception. Br J Surg 1940;27: 475.

348. Tompkins RK (ed). Reoperative Surgery. Philadelphia: JB Lippincott, 1988, p. xi.

349. White TT, Mulholland MW, Harrison RC. Reoperative Gastrointestinal Surgery. Norwalk, CT: Appleton & Lange, 1989.

350. McQuarrie DG, Humphrey EW. Reoperative General Surgery. St. Louis: Mosby Year Book, 1992.

351. Bar-Natan M, Larson GM, Stephens G, Massey T. Delayed gastric emptying after gastric surgery. Am J Surg 172:24-28, 1996. [PubMed: 8686797]

352. Dankmeijer J, Miete M. Le développement précoce de l’estomac chez l’embryo humain. C R Ass Anat 1959; 103:341-344.

Copyright ©2006 The McGraw-Hill Companies. All rights reserved.
Privacy Notice. Any use is subject to the Terms of Use and Notice. Additional Credits and Copyright Information.

Leave a Reply

Time limit is exhausted. Please reload the CAPTCHA.


apply_now Pepperstone Group Limited