In 1817, Ephraim McDowell,1 of Danville Kentucky, reported the first successful transabdominal removal of a pelvic mass, launching therapeutic laparotomy. Wide application of laparotomy was delayed until Morton’s innovation of ether anesthesia as reported by Bigelow2 in 1846. Despite the fact that the important innovations of endotracheal tubes, electric lights, intravenous fluids, intraoperative pharmacologic paralysis, antibiotics, and intravenous nutrition, would slowly develop over the subsequent 150 years, it was general anesthesia that facilitated the expansive future of laparotomy. From the 1840s until the 1880s, therapeutic laparotomy was used to treat pathology as long as the surgical management simply involved drainage of infections or removal of the offending organ (mass) by ligation to control bleeding and abdominal attachments. The most common use of laparotomy, termed “abdominal section” in the 1800s, was for gynecologic pathology.3 Laparotomy was also used in the mid-1800s for other abdominal abnormalities including abdominal wall masses,4 abdominal abscess,5 splenomegaly,6 bowel obstruction,7 intussusception,8 kidney trauma,9 and cholecystitis.10
Despite the previous description of enteral anastomosis in the published literature, it was rarely used in therapeutic laparotomy prior to the introduction of gastrojejunostomy in 1881. In the era before gastrojejunostomy, there was significant controversy as to the best technique for enteric anastomosis. Although suture technique had been introduced to the surgical literature, the type of sutures was actively debated. In addition, prosthetic devices to aid in the anastomosis were frequently used with varying success. Metal, bone or dissolvable tubes or spools were inserted in-between 2 pieces of bowel and held together mechanically or by sutures to allow an anastomosis to mature. These prosthetics were often used because they were much faster than suturing.11 This lack of consensus concerning operative technique and inconsistent outcomes likely delayed the expansion of the use of enteral anastomosis. The first successful gastrojejunostomy was reported prior to worldwide consensus had been reached about the technique of enteral anastomosis. The eventual wide adoption of sutured gastrojejunostomy established this technique as the standard for all luminal anastomoses.
In the 1800s, gastric outlet obstruction from any cause was a lethal diagnosis. If an individuals could not eat or drink, they died quickly of dehydration, or slowly of malnutrition. In this era, the 2 leading causes of gastric outlet obstruction were peptic ulcer disease and cancer; there were no curative medical or surgical treatment for either. The prospect of operative gastrojejunostomy created hope for a fatal condition. This is likely the reason that surgeons around the world pursued a solution for gastric outlet obstruction and why gastrojejunostomy, once it was reported in the literature, was rapidly adopted. The main barrier to the wide adoption of gastrojejunostomy was the need for a fast dependable anastomotic technique that general surgeons could learn and teach.
The first successful gastrojejunostomy was reported from Theodor Billroth’s Clinic in Vienna. Billroth’s clinic was considered to be the most progressive and successful center for innovations related to abdominal (and other) operations. Billroth was born in Prussia in 1829, he completed medical school in Berlin at the Frederick-Wilhelms University in 1852. He was an assistant to Bernhard von Langenbeck and Carl Langenbuch until 1860, when he became the chair of surgery at the University of Zurich (Fig. 1). His success in Zurich led to his appointment at the Professor of Surgery in Vienna in 1867 at the Vienna General Hospital.12 His disciples and trainees include some of the greatest surgeons of the late 19th and early 20th century including John B. Murphy, Johann von Mikulicz-Radecki, Viktor Von Hacker, Vincenz Czerny, and Anton Wolfler.13
Many of the operations that were pioneered at this Clinic actually were created by these “students” of Billroth. Although it has often been stated that these surgeons were “pupils,” in fact, they were senior surgeons at the time and eventually left Billroth to become chairs of their own departments. One of his more prominent protégées was Anton Wolfler, who was born in 1850 in the village of Kopezten in what is the Czech Republic today. He attended the University of Vienna, where he earned a medical degree. He became an assistant and eventually a colleague with Billroth. Later in his career, he became the Professor of Surgery at the University of Graz in 1885 and finally the Professor of Surgery at the Charles University in Prague in 1896.14
Billroth and his colleagues anticipated an operation for obstruction of the pylorus to include gastric resection and a sutured anastomosis between the stomach and duodenum. They prepared for this innovation by practicing this operation on 10 living dogs. In this experiment, they were evaluating their resection and their suturing technique, tracking their technical success and its effect on mortality.15 Billroth’s first successful gastric resection occurred on January 29, 1881, and included the resection of a tumor at the pylorus plus an anastomosis later termed “Billroth 1.”16 In September of 1881, Wolfler operated on a 38-year-old male patient with malignant gastric outlet obstruction. The patient originally reported to Dr Billroth’s clinic, where Wolfler offered laparotomy to remove the patient’s gastric cancer. At operation, the tumor was unresectable, so Wolfler proceeded with a sutured gastrojejunostomy to relieve the obstruction. After Dr Wolfler’s palliative operation, the patient recovered well and was able to eat by mouth. The case report of this novel operation appeared in April of 1882.17
SUTURES, BUTTONS, AND BONES
The importance of Wolfler’s successful gastrojejunostomy in the management of gastric outlet obstruction was recognized and adopted by a variety of surgical centers around the world.18–20 During this early period of adoption of gastrojejunostomy, surgeons continued to debate the best approach for creating an enteral anastomosis either by sutures or prosthetic devices. Nicholas Senn, a prominent Chicago surgeon and president of the American Surgical Association, stated that the “modern” approach to sutured bowel anastomosis began with Lembert’s report in 1826.11,21 From the 1820s to the 1890s, most major surgical centers in the Europe and the United States published differing suture technique for bowel anastomosis. These included invagination techniques of Jobert (1827), 2-layer modifications by Czerney and Wolfler (1881), Halsted’s horizontal mattress sutures (1887), and Connell’s running mattress suture (1892).11 On the other hand, there were several prosthetic-based anastomotic techniques that were either bone-based or metal devices that were placed at the interface at the anastomosis. The bone-based devices included Senn’s decalcified bone plates, Robson’s bone bobbin, and Paul’s bone tube. These techniques appeared in the literature but never had wide acceptance because they were difficult to place or did not heal predictably.22 Several other techniques using metal devices were well described, but the most successful was Murphy’s button.
In the first 30 years of gastrojejunostomy, the most common technique for this anastomosis used Murphy’s Button.23 The inventor of the Button, John Benjamin Murphy (Fig. 2), was born in Wisconsin in 1857 and graduated from Rush Medical College in 1879. He trained as a surgeon at Cook County Hospital in Chicago and in Vienna under Billroth. He served on faculty at many of Chicago’s medical colleges and was the Chair of Surgery at Mercy Hospital.24 Murphy often hosted visiting surgeons in his clinics, the notes of which were transcribed and published as the Surgical Clinics of John B. Murphy (later named the Surgical Clinics of North America).25 He is known for several innovations including “Murphy’s Sign,” the physical finding in acute cholecystitis described as inspiratory arrest during palpation of the right upper quadrant. In addition to his novel approach to surgical problems, Murphy had an outsized personality. In 1912, after the attempted assassination of Theodore Roosevelt in Milwaukee, the ex-president was transported to Chicago under the care of Murphy. The extrovert, Roosevelt, was easily matched by the personality of Murphy.26
Murphy’s Button was a 2-part spool made out of metal with flanged ends, which were secured in the enterotomies by a purse string and the Button was snapped together (Fig. 3). Once the anastomosis healed, the Button would be passed either proximally or distally. This anastomosis could be created with a very short operative time and was technically easier then suturing. The Button had a major impact on the evolution of gastrojejunostomy. The speed and ease of placement of the Button between the stomach and jejunum allowed early adoption of the new operation. Unfortunately, as the use of the Button grew, complication directly related to the Button was reported and use of the device started to diminish. After 1925, the Button was rarely used for gastrojejunostomy as most surgical leaders were recommending a sutured anastomosis. In fact, in 1924, in his presidential address to the Royal College of Medicine, Mr Herbert J. Patterson referred to the Button as “long since discarded.”27
COMPLICATIONS AND MODIFICATIONS OF GASTROJEJUNOSTOMY
Soon after Wolfler’s successful antecolic gastrojejunostomy, a second such patient was operated on by Dr Billroth himself, but the patient did not recover after developing bilious vomiting. At autopsy, the gastrojejunostomy was sufficiently twisted that the efferent limb was obstructed.17 Thus, the very beginning of the history of gastrojejunostomy was plagued by efferent obstruction a problem that still occurs today.28 Throughout the early history of gastrojejunostomy, technical errors were encountered and solved by innovative persistence. The early observation from a variety of surgical centers during the late 1800s was that cyclical bilious vomiting was usually due to efferent limb obstruction at the site of the gastrojejunostomy, which often resulted in early postoperative mortality due to dehydration. Alternatively, if the afferent limb was obstructed, the patient was also troubled by vomiting, but it occurred chronically and presented weight loss and failure to thrive. Once the presumptive diagnosis was made, both afferent and efferent limb obstruction prompted reoperation and modification of the gastrojejunostomy.
The early experience from the Billroth clinic prompted a debate concerning advantages and disadvantages of antecolic versus retrocolic gastrojejunostomy. This debate began as complications were seen with antecolic gastrojejunostomy, pushing surgeons to consider Anton von Hacker’s more complicated posterior gastrojejunostomy, reported in 1885. Von Hacker was born in 1852 in Vienna and went to medical school at the University of Vienna, where he graduated in 1878. By 1880, he was a “student” under Billroth. In 1903 he was named the Professor and Chair of the Department of Surgery at Karl-Franzens University School of Medicine in Graz Austria.29 The retrocolic gastrojejunostomy was technically more challenging than the antecolic version of the operation as reported by William Mayo in 1905.30 The retrocolic approach required an opening through the colonic mesentery and was associated with a longer operative time.31 By the turn of the century, 20 years after the initial report of Wolfler’s operation, most of the elite surgeons of the world believed that a retrocolic anastomosis between the back of the stomach and the very first portion of the jejunum would create a gastroenterostomy with the least chance of anatomic complication despite its longer operative time.
The complication of efferent limb obstruction was originally identified by Wolfler, but it was a German surgeon, Heinrich Braun, who devised an enteroenterostomy between the afferent and efferent limb to treat/avoid the complication.32 By 1904, William Mayo of the Mayo Clinic reported treating afferent and efferent limb obstruction by creation of a “Braun” anastomosis. Mayo recommended surgical suture obstruction of the afferent limb just proximal to the gastrojejunal anastomosis, which would prevent the “Braun” enteroenterostomy from stricturing due to nonuse.33 Mayo also used the “Y” anastomosis, championed by the Swiss surgeon, Cesar Roux, as a modification that would avoid cyclical vomiting. Roux credited Wolfler for the original conception of the “y” anastomosis. It had the obvious advantage that it eliminated cyclical bilious vomiting but was criticized as long and technically challenging in an era where speed of an operation was considered as perhaps the most important feature.34,35
Marginal ulceration at the gastrojejunal anastomosis was likely first noted by Braun before 1900 but fully characterized by Sir David Wilkie 10 years later.36 Early on, it was thought to be rare but more common in patients with peptic ulcer disease. Ulceration occurred on the wall of the jejunum opposite the anastomosis and also at the anastomosis. This complication presented as obstruction and perforation and occasionally bleeding. Remedial operations to treat marginal ulceration were complex and often required resection of the ulcer and revision of the gastrojejunostomy.37 The addition of vagotomy was not recommended until Lester Dragstedt’s seminal paper in 1945.38
Obstruction from true anastomotic strictures, not related to anatomic kinking or ulceration, was often associated with Murphy’s Button. Occasionally the Button would lodge in the anastomosis or leave a stricture as it passed.30 In addition, gastrojejunostomies were prone to stricture if the native gastric outlet was incompletely obstructed. If normal enteral flow through the pylorus was present, the gastrojejunostomy would stricture from lack of use. To avoid this complication some surgeons recommended that the pylorus be surgically closed with suture to keep the gastrojejunostomy open and functioning.39 Small bowel obstruction was recognized early in the history of gastrojejunostomy. Murphy’s Button was responsible for bowel obstruction if the device lodged at the ileocecal valve as it attempted to pass distally. The specific complication of herniation of a loop of intestine behind an antecolic gastrojejunostomy was observed by a variety of authors but bears the name of Walther Petersen after his report in 1900.40
Perhaps best depiction of a “Petersen hernia” was made by William Mayo in his 1902 Annals of Surgery article, which also includes a description of all of the complications of gastrojejunostomy noted above.41 William J. Mayo (Fig. 4) was born in June of 1861 and along with his younger brother, Charles Mayo, and their father, William Worrall Mayo, created the Rochester Minnesota-based Mayo Clinic.42 Between 1890 and 1921, W. J. Mayo was a major contributor to the world’s literature documenting the indication, techniques and outcome of gastric surgery through the publication of at least 37 single-author articles in the medical literature. In the first 40 years after Wolfler’s description of gastrojejunostomy, the Annals of Surgery had a large role in documenting the evolution of this operation due to the publication of 165 articles on the topic. Of those 165 articles, Mayo Clinic surgeons (W. J. Mayo, C. H. Mayo, Donald C. Balfour) authored 22 dealing with gastrojejunostomy and its complication. Their large case series was unmatched in the world literature in that era.30 In multiple publications, predominantly in the Annals of Surgery, Mayo surgeons made gastrojejunostomy common place in the management of gastric cancer and peptic ulcer disease.
The safe surgical treatment of gastric outlet obstruction was anticipated in the mid-19th century but did not come to fruition until 1881 with Wolfler’s successful antecolic gastrojejunostomy in a patient with unresectable antral cancer. Between 1882 and 1890, Billroth and his colleagues, Wolfler and von Hacker, described a variety of complications that occurred after gastrojejunostomy including afferent/efferent limb obstruction, and small bowel obstruction into the space behind an antecolic gastrojejunostomy. Billroth’s disciples also reported on various modification of the original gastroenterostomy, which included retrocolic gastrojejunostomy, the “y” modification and the enteroenterostomy below a loop gastrojejunostomy. After Billroth and his colleagues described these operations, American surgeons such as J. B. Murphy and William Mayo had a major influence in making this new procedure common place in abdominal operations. As technical challenges were managed and patient selection improved, the mortality steadily improved. Unfortunately, with all of the various modification that were tried, the complication of efferent obstruction, afferent obstruction, marginal ulceration, strictured anastomosis, and bilious vomiting have persisted and still plague today’s surgeons.
While surgeons created solutions for some of the unfortunate consequences of gastrojejunostomy, perhaps the more important byproduct of this work was the perfection and validation of the sutured enteral anastomosis. Early laparotomy was limited to operations that were accomplished with removal of diseased organs by simple ligation, its attachment and blood supply (think of appendectomy or removal of pelvic masses). Therapeutic laparotomy to treat many of the diseases of the gastrointestinal tract advanced slowly due to the lack of consensus on the best enteral anastomosis. The first gastrojejunostomy was done via a suture technique between the stomach and jejunum in a patient with gastric outlet obstruction. This was a good environment to further develop the suture techniques that eventually became the norm for all enteral anastomoses. Gastrojejunostomy for outlet obstruction was optimal for this anastomotic trial due to the healthy thickness of the stomach, which was always in close proximity to a normal caliber jejunum. Other operations such as small bowel obstruction, colonic tumors, bowel perforation, and other intestinal emergencies would require emergent enteral anastomosis, but the bowel was often thin or diseased and not optimal for anastomosis. Outside of gastrojejunostomy, early success with enteral anastomosis was limited. In addition, many of the disease of the gut in that era which might have been managed by surgeons were difficult to diagnose at an early enough stage to end up with successful surgical management. On the other hand, gastric outlet obstruction was often easy to diagnose. The history and physical exam were often diagnostic. Patient was usually very thin and tumors in the gastric outlet were often palpable. Distended, obstructed stomach was palpable, but the rest of the abdomen was not distended. Gastric lavage temporarily relieved symptoms and insufflation of air into the stomach did not produce diffuse abdominal distension.
In the first 4 decades of gastrojejunostomy, many types of anastomotic techniques were tried. Initially, speed was of primary importance. Prior to the use of general anesthetics, when patients were awake and moving during a laparotomy, the surgeon had to be fast or the operation would fail. In the 1880s, the speed of an operation was still considered one of the main features of a successful operation.43 Because of the importance of speed, the Murphy Button was used frequently in the early era of gastrojejunostomy. It was reported by J. B. Murphy that with little practice, he could complete a gastrojejunostomy with a Button in 6 minutes.44 Over time, after every feasible technique of anastomosis had been tried, and complications of the Button became apparent; sutured anastomosis became the standard of care. The successes of sutured gastrojejunostomy, opened the door to success in many other abdominal operations that required intestinal anastomosis. In the early 1900s, as technique for gastrojejunostomy was being defined, operations to treat bleeding and obstruction of other portions of the gastrointestinal tract were reported in the surgical literature with greater frequency with sutured anastomosis as the key technical advance.45–47 Gastrojejunostomy can be considered a “gateway procedure” since it was the operation that showed the efficacy of sutured anastomoses, which made other luminal operations safe and within the capabilities of most general surgeons.
In the 19th and early 20th century, there were several innovations that facilitated the expanded indication for therapeutic laparotomy. The development of gastrojejunostomy had a major impact on the willingness of surgeons to treat abdominal pathology that required luminal anastomosis. Prior to gastrojejunostomy, surgeons limited therapeutic laparotomy to extirpation of abdominal pathology. As gastrojejunostomy evolved, surgeons became facile with enteric anastomosis. Gastrojejunostomy started as a sutured anastomosis at the Billroth Clinic, was widely adopted through the use of the Murphy Button and popularized by a large experience from the Mayo Clinic. Experience with and use of sutured enteric anastomosis grew largely after its successful application for gastrojejunostomy.
1. McDowell E. Three cases of extirpation of diseased ovaria. Eclectic Repertory Anal Rev. 1817;7:242–244. (This article can be found at Am J Ob Gyn
2. Bigelow HJ. Insensibility during surgical operations produced by inhalation. Boston Med Surg J. 1846;35:309–317.
3. Thornton JK. One hundred and seventy-two antiseptic abdominal section. Medico-Chirurgical Trans. 1881;64:139–165.
4. Cheever DW. A remarkable recovery from abdominal section. Boston Medical Surgical J. 1874;90:181–183.
5. Tait L. Treatment of pelvic suppuration. Medico-Chirurgical Trans. 1880;63:307–316.
6. Martin A. A successful case of splenotomy. Brit Med J. 1878;1:191–192.
7. Jessop TR. Abdominal section in intestinal obstruction. Br Med J. 1879;2:488–490.
8. Hutchinson J. A successful case of abdominal section for intussusception. Medico-Chirurgical Trans. 1874;57:31–74.
9. Barker AE. Nephrectomy by abdominal section. Med Chir Trans. 1880;63:181–215.
10. Tait L. Case of cholecystotomy performed for dropsy of the gall-bladder due to impaction of a gall-stone. Medico-Chirurgical Trans. 1880;63:17–22.
11. Senn N. Enterorrhaphy; its history, technique, and present status. JAMA. 1893;21:215–235.
12. Fong ZV, Lavu H, Rosato EL, et al. Christian Albert Theodor Billroth, M.D., founding father of abdominal surgery (1829-1894). Am Surg. 2012;78:280–281.
13. Androutsos G. Theodor Billroth (1829-1894) and other protagonists of gastric surgery for cancer. J BUON. 2004;9:215–220.
14. Sigua BV, Zemlyanoy VP, Danilov AM, et al. Anton Wolfler-forgotten genius of surgery. Bull Russian Military Med Acad. 2020;69:224–227.
15. Gaertner F. Vivisection. Am Nat. 1891;25:864–869.
16. Wolfler A. On the removal of the pylorus for cancer, in appendix III of Clinical Surgery, extract from the reports of surgical practice (Theodore Billroth). New Sydenham Society, London; 1881:494–510. (retrieved on July 9, 2022 from, https://www.google.com/books/edition/Clinical_surgery/tj0JAAAAIAAJ?hl=en&gbpv=1&dq=billroth+pylorus&pg=PR1&printsec=frontcover
17. Wölfler A. Gastro-enterostomie. Zentralbl Chir. 1881;8:705. (Translation available at, Half-yearly compendium of Medical Science edited by DG Brinton. 1882;30:261–262,article 140).
18. Koplik H. Gastro-enterostomy in Luecke’s Clinic. Ann Surg. 1888;7:199–203.
19. Renton JC. Notes of a case of gastro-enterostomy for dilatation of the stomach-recovery. Glasgow Med J. 1892;38:424–427.
20. Maunsell HW. A new method of intestinal surgery. Am J of Med Sci. 1892;103:245–258.
21. Lembert A. Nouveau procedre de enterorraphie. Repertoire Gen Anat Physiol Path. 1826;2:3–11. (Translation available at Dis Colon Rectum
22. Edmunds W, Ballance CA. Observations and experiments on intestinal and gastrointestinal anastomosis. Med Chir Trans. 1896;79:255–311.
23. Murphy JB. Cholecysto-intestinal, gastro intestinal, entero-intestinal anastomosis, and approximation without sutures. Med Rec. 1892;42:665–676.
24. Dr. John B. Murphy, Noted Surgeon Dies. New York Times
. August 12, 1916, page 9.
25. Rutkow IM. A history of the surgical clinics of North America. Surg Clin North Am. 1987;67:1217–1239.
26. Barr J, Pappas TN. Stricken before election: presidential health crises from 1880 to 2020. Ann of Surg Open. 2022;3:e126.
27. Paterson HJ. The lure of the new: president’s address. Proc R Soc Med. 1925;18(Surg Sect):1–9.
28. Lim D, Bain K, Sinha P. Successful endoscopic management of efferent loop syndrome after Billroth II distal gastrectomy. BMJ Case Rep. 2018;2018:bcr2018.
29. Stanger O. Victor von Hacker (1852-1933)–recollections of the life and works of an important Billroth student. Chirurg. 2000;71:478–484.
30. Mayo WJ. I. A review of five hundred cases of gastro-enterostomy, including pyloroplasty, gastroduodenostomy, and gastrojejunostomy
. Ann Surg. 1905;42:641–655.
31. Mayo WJ. V. Malignant diseases of the stomach and pylorus. Ann Surg. 1900;32:251–277.
32. Czerny V. Heinrich braun. Dtsch Med Wochenschr. 1911;37:1037–1038.
33. Mayo WJ. X. Duodenal ulcer: a clinical review of fifty-eight operated cases, with some remarks on gastrojejunostomy
. Ann Surg. 1904;40:900–908.
34. Roux C. Les anastomoses intestinales et gastrointestinales. Rev Gynecologie Chirurgie Abdominale. 1900;4:787–796.
35. Vauthey JN, Maddern GJ, Gertsch P. César Roux–Swiss pioneer in surgery. Surgery. 1992;112:946–950.
36. Wilkie DPD. “Gastro-jejunal and jejunal ulceration following gastro-enterostomy.” Edinburgh Med J. 1910;5:316–327.
37. Erdmann JF. Marginal, gastrojejunal or peptic ulcer subsequent to gastroenterostomy. Ann Surg. 1921;73:434–440.
38. Dragstedt LR. Vagotomy for gastroduodenal ulcer. Ann Surg. 1945;122:973–989.
39. Mayo WJ. V. The technique of gastrojejunostomy
. Ann Surg. 1906;43:537–542.
40. Petersen W. Concerning twisting of the intestines following a gastroenterostomy. Arch Klin Chir. 1900;62:94–114.
41. Mayo WJ. VII. Complications following gastro-enterostomy. Ann Surg. 1902;36:231–244.
42. Dr. William Mayo of Clinic is Dead. New York Times
. July 29, 1939, Section O page 18.
43. Schlich T. ‘The days of brilliancy are past’: skill, styles and the changing rules of surgical performance, ca. 1820-1920. Med Hist. 2015;59:379–403.
44. Murphy JB. An analysis of the cases operated upon with the Murphy Button up to date. Lancet. 1895;145:1040–1047.
45. Macfee WF. Resection with aseptic end-to-end anastomosis for carcinoma of the colon. Ann Surg. 1937;106:701–714.
46. Workman EW, Miller GG. A clinical review of two hundred and forty-one cases of obstruction of the small bowel. Can Med Assoc J. 1934;30:141–148.
47. Walton AJ. Reconstruction of the common bile-duct. Brit J Surg. 1921;34:169–178.