Powis SJA, BARNES AD, DAWSON-. EDWARDS P, et al: Ileocolonic prob- lems after cadaveric renal transplanta- tion. Br Med J 1: 99, 1972. 38. DEMLING RH ...
Gastrointestinal complications of renal transplantation. 1. The upper gastrointestinal tract STUART D. ARCHIBALD, MD, FRCS[C]; DENNIS W. JIRSCH, MD, M SC, PH D, FRCS[C]; ROBERT A. BEAR, MD, FRCP[C], FACP
In 95 consecutive cases of cavaderic renal transplantation followed up for I to 83 months (mean 23.1 months) 17 complications developed in the upper gastrointestinal tract of 15 patients; these included duodenal ulcer in 12 and gastric ulcer, esophagitis, hemorrhagic gastritis, small-bowel obstruction and small-bowel perforation in I each. The occurrence of a complication was not related to the patient's age, sex, blood group or use of cigarettes or alcohol, the duration of hemodialysis before transplantation, the tissue match or the number of infusions of immunosuppressive medication. One patient died of the complication. The peptic ulcers that developed after transplantation were successfully managed conservatively in 690/o of cases. Since surgical treatment in patients whose immune response has been suppressed is associated with an increased frequency of complications such as disruption of suture lines, it is preferable to reserve it for those in whom complications develop that are unresponsive to conservative measures. Dans 95 cas consecutifs de greffe d.un rein cadaverique surveilles pendant I a 83 mois (23.1 mois en moyenne) 17 complications du tractus gastrointestinal superieur sont survenues chez 15 patients; ceci comprend un ulc.re duodenal chez 12 patients et un ulcere From the departments of surgery and medicine (nephrology), St. Michael's Hospital and the University of Toronto Reprint requests to: Dr. Dennis W. Jirsch, 30 Bond St., Toronto, Ont. M5B 1W8
gastrique, une oesophagite, une gastrite hemorragique, une obstruction du petit intestin et une perforation du petit intestin dans I patient chacun. L'apparition de complications na pas ete reliee a l'ige, au sexe, au groupe sanguin ou a l'usage de Ia cigarette ou l'alcool, au temps cumulatif passe sous hemodialyse avant Ia greffe, a Ia compatibilite tissulaire du greffon pour le receveur ou au nombre de perfusions de medicaments immunosuppresseurs. Un patient est mort de ses complications. Les ulceres peptiques qui sont apparus apres Ia greffe ont ete traites avec succ&s de fa.on conservatrice dans 69% des cas. Comme le traitement chirurgical est associe a une frequence accrue de complications, telles que Ia rupture de Ia ligne de suture, chez les patients dont Ia reponse immunitaire est diminuee, il vaut mieux le reserver pour ceux dont les complications sont resistantes a des mesures conservatrices.
Renal allotranspiantation and antirejection therapy are associated with numerous complications affecting the upper gastrointestinal tract. Moore and Hume' reported that gastrointestinal hemorrhage was the single most lethal complication of transplantation: all their patients thus affected died. Although such complications are not rare, there is no consensus as to their pathogenesis and optimal management. Accordingly, we reviewed the experience with 95 consecutive cases of cadaveric renal transplantation at St. Michael's Hospital, Toronto, to determine the fre-
quency, nature, clinical import and method of treatment of complications affecting the upper gastrointestinal tract after renal transplantation. For comparative purposes we surveyed the relevant English-language literature. Patients and methods In our centre renal transplantation is performed in selected patients2 by the same surgical team using standard surgical techniques. Currently immunosuppressive medication consists of daily maintenance doses of prednisone (100 mg tapered to 15 mg by 1 year) and azathioprine (2 to 3 mg/kg of body weight) and five intravenous infusions of 750 mg of methylprednisolone and 400 mg of cyclophosphamide during the first 21 days after transplantation and as needed thereafter to control rejection. Twelve of the first renal transplant recipients received no such infusions, but only oral therapy with prednisone and azathioprine. Roentgenographic studies of the upper gastrointestinal tract after administration of barium were performed before transplantation in patients with either a significant past history or current symptoms of disease in this area of the body. The charts of all recipients of cadaveric renal transplants from November 1969 to June 15, 1977 were reviewed. All patients were followed up for at least 4 months or until
CMA JOURNAL/DECEMBER 9, 1978/VOL. 119
Table I-Complications affecting tile upper gastrointestinal tract among reclpie,.ts of 95 cadaverlcrenaltrausplants No. of No. of Complication complications deaths Esophagitls 1 It Gastritis I I Peptic ulcer GastrIc 1 it Duodenal 12 2t Small-bowel disease Obstruction II 0 Penforation 1 1 'Tho patient also had. duodenal ulcer. fTheobstruction recurred. tThe complication was pot a major causative factor In the death.
and required emergency vagotomy and pyloroplasty (no. 9). Of the three patients (nos. 11, 12 and 13) with bleeding ulcers treated with antacids and cimetidine, two experienced bleeding again after initial control was achieved (nos. 12 and 13). Thus, in this small series there was a 50% failure rate for cimetidine in the control of peptic ulceration. The nine patients whose complication was controlled with conservative treatment remained ulcer-free for a mean follow-up period of 19 months. Surgical treatment of peptic ulceration was performed only as an emergency procedure. In two patients (nos. 9 and 13) massive hemorrhage from the ulcer necessitated an operation. In case 9 emergency vagotomy and pyloroplasty controlled the bleeding, which had occurred after 1 month of antacid and cimetidine therapy, but, as was found at autopsy some months later, the pyloroplasty did not heal (Table IV). In the other case (no. 13) a 50-year-old man with severe hemorrhage from the upper gastrointestinal tract was treated by emergency vagotomy and pyloroplasty followed by cimetidine therapy. Despite initial success, hemorrhage from a persistent giant duo-
denal ulcer occurred a month later, follow-up was marred by recurrent necessitating emergency oversewing sepsis and general debility, but renal of the ulcer and a Billroth II opera- function was maintained with minition. Bleeding ceased, but persistent mal immunosuppression. drainage from a lateral duodenosEsophagitis was diagnosed within tomy site was a problem (Table IV). a month after transplantation in a One week after the operation a duo- 43-year-old obese man with diabetes denocolic fistula was evident between who had retrosternal and epigastric the duodenostomy site and the hepa- pain, and iron deficiency anemia sectic flexure. Since the colon was not ondary to chronic blood loss. The thought to have been injured during diagnosis was made endoscopically, the operation, the fistula was con- and treatment with antacids gave sidered to be due to the presence of symptomatic relief, but the patient the duodenostomy tube. Some months died 2 weeks later of unrelated later an enteroenterostomy between causes. afferent and efferent loops and an Hemorrhagic gastritis was found ileocolostomy were performed. The in a 33-year-old woman preterminpatient's course during the period of ally. Hyperacute rejection necessiCMA JOURNAL/DECEMBER 9, 1978/VOL. 119
tated removal of the renal allograft 5 days after transplantation. The doses of immunosuppressive agents were tapered and then the medications were discontinued by 1 week. Her clinical course was complicated by systemic sepsis, recurrent episodes of hypotension with massive retroand intraperitoneal bleeding from the nephrectomy site, and episodic gastric hemorrhage treated conservatively. Postmortem examination revealed multiple acute ulcers of the lesser curvature of the stomach. There were two small-bowel complications. In case 5 a 44-year-old man had two episodes of incomplete small-bowel obstruction; in each instance the obstruction was secondary to an ileal Richter's hernia in a pentoneal defect initially created and repaired at the time of renal transplantation. Repair was performed twice before closure was maintained. In case 9 vague abdominal pain, fever, ileus and a recurrent, tender, left-sided abdominal mass developed over several weeks, 12 months after transplantation. Laparotomy revealed four small-bowel perforations communicating with an abscess cavity. The perforations were included in two short small-bowel resections, and continuity was restored with primary anastomoses. Enterocutaneous fistulas developed in both suture lines and a duodenal ulcer developed subsequently. The patient suffered a fatal cardiac arrhythmia 3 months later. Discussion The development of a gastric or duodenal ulcer is the most common gastrointestinal problem seen after renal transplantation,3 having been found in up to 18% of patients.4 In the review by Owens and colleagues4 more than half the reported cases of peptic ulcer presented with hemorrhage; bleeding was a more common complication of duodenal ulcer, and perforation a more common complication of gastric ulcer. In the patients in whom an ulcer developed after transplantation the mortality was 40%, and nearly 90% if an emergency operation was required. In our series no patients died directly of peptic ulceration; however, the number of patients was small. Some authors have recommended a prophylactic ulcer operation before
renal transplantation,1'.7 believing that patients who may require an operation for peptic ulcer after transplantation can be identified before transplantation and should be treated preventively. According to these authors the morbidity and mortality are lower with this approach than with nonoperative treatment or emergency operation. Selection of these patients is difficult. Some authors7 have reported high accuracy in selecting patients simply by means of a positive initial history or abnormal roentgenograms of the upper gastrointestinal tract made after barium administration. Most clinicians, however, consider this method inadequate3 because of the well recognized difficulties in interpreting such roentgenograms in patients with chronic renal failure.3-10 In our series we did not routinely study patients roentgenographically before transplantation. Only four of those in whom ulcers developed after transplantation were thought to warrant roentgenographic investigation before transplantation, and the results were negative in three. Documentation of gastric hypersecretion has been considered by some to be a sufficient indication for prophylactic vagotomy or antrectomy before renal transplantation,5'6 although the value of this approach has not been satisfactorily demonstrated. Gastric acid studies in general are of limited use,11 and they are even less predictable in patients with chronic renal failure.12 Most patients receiving hemodialysis have acid hypersecretion3'1316 and high serum gastrin values.11'17'18 Ulcer disease in patients with azotemia1921 may be related to hypergastrinemia, although this has been disputed.18 Hypergastrinemia before transplantation should not be considered an indication for a prophylactic ulcer operation since serum gastrin values usually become normal in patients with a functioning renal transplant,18 yet peptic ulceration may still develop.. Not only have selection criteria for a prophylactic ulcer operation not been established for these patients, but also there is controversy, even among proponents of such an operation, regarding the best operation. Both vagotomy and drainage, and vagotomy and gastrectomy have been proposed, but the data are inconclusive. The effectiveness of panietal cell
1294 CMA JOURNAL/DECEMBER 9, 1978/VOL. 119
vagotomy has not been studied in these patients. In spite of these deficiencies, a low operative mortality and a low reported rate of recurrence of peptic ulceration after transplantation4 seem to favour performance of a prophylactic operation. The data of Spanos and colleagues7 provide the best support for this approach, but these investigators failed to present objective evidence of peptic ulcer in 15 of the 30 patients selected before transplantation from 377 prospective candidates for transplantation. All 15 patients with verified ulcers received an ulcer operation before transplantation, so that there was no conservatively managed control group. After the operation either hemorrhage occurred before transplantation or the ulcer recurred after transplantation in five of these patients. Two patients died, either from ulcer recurrence or from a complication of laparotomy. In comparison, in 4 of our 13 patients in whom peptic ulceration developed the ulcer recurred with conservative treatment but no deaths were attributable to the ulcer. On the basis of these limited figures we consider the role of a prophylactic operation for peptic ulcer in transplant recipients to be undetermined. Another therapeutic approach to peptic ulcer in transplant recipients involves manipulation of the type, the route of administration or the dosage of immunosuppressive medications. This presupposes that these medications, particularly corticosteroids, are implicated etiologically. Although this theory has been challenged,23 it is probably valid14 when steroids are administered for more than 30 days, when the daily dose exceeds 20 mg of prednisone, when the total steroid dose exceeds 1000 mg of prednisone or when there is a history of prior peptic ulceration. Renal transplant recipients generally meet these criteria. Altering the type of medication from prednisone to methylprednisolone, and the route of administration from oral to intravenous is claimed by Woods and associates11 to have greatly reduced the frequency of gastrointestinal hemorrhage. Administering steroids every second day is purported to reduce the frequency of gastrointestinal side effects..A While these claims are noteworthy, the numbers of such cases reported has
been small. In our series the mean number of steroid infusions received by patients in whom peptic ulcers developed (5.2) was not significantly different from the mean number received by the remainder of the transplant population over the same mean period of follow-up (6.0). A third approach has involved the addition of H2-receptor antagonists, such as cimetidine, to antacid therapy. Our results with this agent were disappointing; further studies with varied dosages and routes of administration and with documentation of factors such as the gastric pH are required. A study that would establish the appropriate prophylactic management of peptic ulcer disease would involve renal transplant candidates with endoscopically verified ulcers and would compare those undergoing a prophylactic operation with those treated conservatively with antacids or cimetidine or both. Without such a study, clinicians' approaches to the prevention of peptic ulceration remain anecdotal and open to criticism. Similarly, a study is needed to determine the best method of managing peptic ulcers that develop after transplantation. However, because of the precarious medical status of many of these patients, a controlled study is unlikely. Our data support the concept that conservative management of symptomatic ulcers is generally adequate, and that surgical treatment should be reserved for otherwise unmanageable complications. Our experience of a 1 % frequency of ulcerative esophagitis in renal transplant recipients contrasts with the 5% frequency previously reported.3'9 Documented cases have been attributed to gastric acid reflux,9 Candida albicans infection9'27 and cytomegalovirus infection,27'28 (the last especially when other gastrointestinal ulcerations are present).29 Esophageal ulcers (occasionally perforated) have been linked with steroid therapy20'3' and, rarely, with uremia,32 and may be suspected when dysphagia, retrosternal pain or occult blood loss is present. Roentgenography after a barium swallow is helpful in excluding other lesions, but diagnosis requires esophagoscopy or esophageal brushing, or both, or biopsy. Treatment is generally symptomatic;28 antacid therapy may be sufficient, as
in our patient. The efficacy of H2receptor antagonists is unknown. The possibility of candidal ulceration can be greatly reduced by restricting the use of broad-spectrum antibiotics; amphotericin B should not be given prophylactically, although it may be required for the treatment of established lesions.3 No antiviral treatment is available for suspected cytomegalovirus infection. Erosive gastritis causing bleeding is easily recognized in the early period after transplantation;3'7'27'33 the hemorrhage may range from trivial to torrential. The diagnosis is suggested by gastrointestinal hemorrhage in an asymptomatic or mildly symptomatic patient, and must be verified by gastroscopy. Gastritis can be due to many factors, including operative stress, uremia, acid-peptic digestion or mucosal compromise as an effect of steroid or antacid administration. Treatment depends on the degree of hemorrhage and the patient's general condition. With minor bleeding, conservative treatment is appropriate. Recently, H2-receptor antagonist therapy in a small series of renal transplant recipients effectively stopped hemorrhage caused by erosive gastritis.34 The use of arterial infusion of vasopressin has not yet been evaluated. Surgical treatment usually consists of vagotomy and drainage or variable gastric resection. The role of gastric devascularization has not been determined. Complications affecting the small bowel are uncommon following renal transplantation.27'33'. Although bowel obstruction secondary to bands or adhesions is possible whenever the peritoneal cavity is entered, other reported causes of mechanical obstruction include fecal impaction36 and hernia (in our case via an iatrogenic peritoneal defect). Spontaneous single or multiple perforations of small bowel have been described following renal transplantation.33 Although in one case perforation occurred in a closed-loop obstruction secondary to adhesive bands,3 in other cases of perforation and necrosis, vasculitis of unknown cause has been described.37 Large-vessel thrombosis or embolism, especially in patients with arteriosclerosis, and perhaps following an episode of hypotension, can cause bowel necrosis and perforation.37 Our patient did not have significant arterial disease and
had not been hypotensive, but still had distal mesenteric arterial thromboses and perforations, the basis of which remains unclear. Uremia has not been shown to cause these perforations. The clinical presentation of intestinal perforation in renal transplant recipients may include abdominal distension, pain, paralytic ileus, fever and systemic sepsis.37'38 Laparotomy is essential to confirm the diagnosis and resect the affected portion of bowel. Primary anastomosis avoids the fluid and electrolyte disturbances of possibly multiple enterostomies, but is hazardous if performed in the presence of peritoneal contamination,3 since the risk of suture-line disruption is increased. Even when suturing of uncompromised gastrointestinal tissue can be done in an uncontaminated field, the risk of disruption is increased in these patients, whose immune response has been suppressed and who may be debilitated. This is suggested by the pyloroplasty dehiscence in one of our cases. When suturing is done in the presence of contamination (as with the two small-bowel anastomoses in the same case) the likelihood of suture line disruption rises. 3 However, this risk may be unavoidable. Furthermore, it appears that intraperitoneal drains should be used cautiously, since injury to the bowel may result, as in one of our cases. Conclusion Complications of cadaveric renal transplantation affecting the upper gastrointestinal tract are relatively common and may be life-threatening. The patients should therefore be observed closely for the development of such problems. If they occur, close cooperation between physician and surgeon is essential. We thank Drs. M. Johnson and M.B. Goldstein for permitting their patients to be included in this study. References 1. MOORE TC, HUME DM: The period and nature of hazard in clinical renal transplantation. I. The hazard to patient survival. Ann Surg 170: 1, 1969 2. KJELLSTRAND
BUSELMETER TJ, et al: Renal transplantation: recipient selection, medical management, and dialysis, in Transplantation, 1st ed, NAJARIAN JS, SIMMONS RL (eds), Lea & Febiger, Philadelphia, 1972, pp 418-45
CMA JOURNAL/DECEMBER 9, 1978/VOL. 119
3. HADJIYANNAKIS EJ, EVANS DB, SMELLIE WAB, et al: Gastrointestinal complications after renal transplantation. Lancet 2: 781, 1971 4. OWENS ML, PAsSARo E, WILSON SE,
et al: Treatment of peptic ulcer disease in the renal transplant patient. Ann Surg 186: 17, 1977 5. GORDON EM, JOHNSON AG, WILLIAMS
22. CANAVAN JSF, BRIOGS JD, BELL PRF: Gastric acid secretion following renal transplantation. Br J Surg 62: 737, 1975 23. CONN HO, BLITZER BL: Nonassociation of adrenocorticosteroid therapy and peptic ulcer. N Engi J Med 294: 437, 1976
G: Gastric assessment of prospective renal-transplant patients. Lancet 1: 226, 1972
24. GREEN SB, GAIL MH, BYAR DP:
6. PICHLMAIER H, lABOUR A, FAUL P,
25. Woons JE, ANDERSON CF, JOHNSON
et al: Nieren transplantation. Arch Kim Med 214: 306, 1968
WJ, et al: Experience with renal transplantation in high risk patients. Surg Gynecal Obstet 137: 393, 1973
7. SPANOS PK, SIMMONS RL, RATTAZZI
LC, et al: Peptic ulcer disease in the transplant recipient. Arch Surg 109: 193, 1974 8. LEwICKI AM, SAITO S, MERRILL JP: Gastrointestinal bleeding in the renal transplant patient. Radiology 102:
533, 1972 9. PETERSON R: Gastrointestinal abnormalities in renal homotransplant pa-
tients. J Can Assoc Radio! 27: 240, 1976 10. WIENER SN, VERTES V, SHAPIRO H: The upper gastrointestinal tract in patients undergoing chronic dialysis. Ra-
diology 92: 110, 1969 11. LANDOR JH: Gastric secretory tests
and their relevance to surgeons. Surgery 65: 523, 1969 12. GINGELL JC, BURNS GP, CHISHOLM GD: Gastric acid secretion in chronic uremia and after renal transplantation.
Br Med J 4: 424, 1968 13. MCCONNELL JB, STEWART WK, THJODLEIFSSON B, et al: Gastric function in chronic renal failure. Effects of maintenance haemodialysis. Lancet
2: 1121, 1975 14. SHEPHERD AMM, STEWART WK, WORMSLEY KG: Peptic ulceration in chronic renal failure. Lancet 1: 1357,
Steroids and peptic ulcer (C). Ibid, p 1291
26. SIEGAL RR, LUKE RG, HELLEBUSCH
AA: Reduction of toxicity of corticosteroid therapy after renal transplantation. Am J Med 53: 159, 1972 27. ALDRETE JS, STERLING WA, HATH-
AWAY BM, et al: Gastrointestinal and hepatic complications affecting patients with renal allografts. Am J Surg 129: 115, 1975 28. DIETHELM AG: Surgical management of complications of steroid therapy. Ann Surg 185: 251, 1977 29. HENSON D: Cytomegalovirus inclusion bodies in the gastrointestinal tract. Arch Pathol 93: 477, 1972 30. GLENN F, GRAFE WR .j.: Surgical
complications of adrenal steroid therapy. Ann Surg 165: 1023, 1967 31. SAUER
LAEGER EE: Serious untoward gastrointestinal manifestations possibly related to administration of cortisone and corticotropin. Proc Staff Meet
Mayo Gun 28: 641, 1953 32. JAFFE RH, LAING DR: Changes of the digestive tract in uremia: pathologic anatomic study. Arch intern Med 53: 851, 1934
15. SULLIVAN SN, TUSTANOFF E, SLAUGH-
33. PENN I, GROTH CG, BRETTSCHNEIDER
TER DN, et al: Hypergastrinemia and
L, et al: Surgically correctable intraabdominal complications before and after renal homotransplantation. Ann Surg 168: 865, 1968
gastric acid hypersecretion in uremia.
Clin Nephrol 5: 25, 1976 16. VENTKATESWARAN PS, JEFFERS A, HOCKEN AG: Gastric acid secretion in chronic renal failure. Br Med J 4:
22, 1972 17. DURKIN MG, ESSIG LI, NOLPH KD:
Gastrin removal during peritoneal dialysis (abstr). Gun Res 19: 657, 1971 18. KORMAN MG, LAyER MC, HANSKY J: Hypergastrinaemia in chronic renal
failure. Br Med J 1: 209, 1972 19. GOLDSTEIN H, MURPHY D, SOKOL A,
et al: Gastric acid secretion in patients undergoing chronic dialysis. Arch intern Med 120: 645, 1967 20. MCLEOD LE, MANDIN H, DAVIDMAN M, et al: Intermittent hemodialysis in terminal chronic renal failure. Can
Med Assoc J 94: 318, 1966 21. PENDRAS JP, ERICKSON RV: Heinodialysis: a successful therapy for chronic uremia. Ann Intern Med 64: 9Q. 1Q.
34. MACDONALD AS, STEELE JB, BOTTOM-
LEY MG: Treatment of stress-induced upper gastrointestinal haemorrhage with metiamide. Lancet 1: 68, 1976 35. JULIEN PJ, GOLDBERG HI, MARGULIS AR, et al: Gastrointestinal complica-
tions following renal transplantation. Radiology 117: 37, 1975
An APMR goes to school for an extra year to improve his knowledge and understanding of the fundamentals of the health sciences. APMR stands for Accredited Pharmaceutical Manufacturer Representative. To become accredited, he must complete his company's training program and he must complete a one-year course of study developed and administered by our association. Passing the final examinations held in Canadian University settings earns him a well-deserved diploma and professional accreditation. The purpose of this program is to increase the representative's overall professional expertise. Now he's better qualified to converse with you in the fields of physiology, pathology, pharmaco-therapeutics and other related subjects. Your APMR is more able to answer those tough questions you put to him. We think you'll notice the difference.
36. PENN I, BRETrSCHNEIDER L, SIMPSON K, et al: Major colonic problems in
human homotransplant recipients. Arch Surg 100: 61, 1970 37. Powis SJA, BARNES AD, DAWSONEDWARDS P, et al: Ileocolonic problems after cadaveric renal transplanta-
tion. Br Med J 1: 99, 1972 38. DEMLING RH, SALVATIERRA 0, BELZER FO: Intestinal necrosis and perforation after renal transplantation. Arc/i
Surg 110: 251, 1975
1296 CMA JOURNAL/DECEMBER 9, 1978/VOL. 119
Council for the Accreditation of Pharmaceutical Manufacturers Representatives of Canada.