Small Intestinal By Douglas
A. Hale, Kathleen
Transplantation A. Waldorf,
James
in Nonhuman
Kleinschmidt,
Richard
Primates
H. Pearl, and Alan E. Seyfer
Washington, DC and Portland, Oregon 0 Small intestinal transplantation represents a potentially therapeutic procedure for individuals with short gut syndrome. The purpose of this study.was to develop a model for small intestioal transplantation in primates that is: technically feasible without microsurgery; consistent in the prevention of allograft rejection; functional in terms of nutrient absorption; and compatible with harvest for multiple organ procurement. First, autotransplantations on four rhesus monkeys were performed in order to study a vtiety of harvesting techniques and vascular anastomoses. Then, a study was performed with 14 heterotopic allotransplants in 4 baboons and 10 rhesus primates. The successful donor model consisted of division of the pancreas, harvesting the small bowel with a superior mesenteric artery and portal vein pedicle. The allograft vascular pedicle was anastomosed to the recipient’s common iliac vessels in end-to-side fashion. The graft was transplanted as an out-of-continuity loop, both ends being exteriorized as stomas providing access for absorption studies and biopsy. Three immunosuppressive regimens were tested: (1) cyclosporine A (CyA) 20 mg/kg/d, solumedrol (SML) 2 mg/kg/d, and graft irradiation (150 rad) (n = 4); (2) CyA 20 mg/kg/d and SML 2 mg/kg/d (n = 3); and (3) CyA 40 mglkgld, SML 2 mgikgld. and azathioprine 5 mg/kg/d (n = 3). There were 4 deaths due to technical error in the first 24 hours. Weekly graft biopsy, serum CyA levels, complete blood count, and automated 2Cchannel serum analysis were performed. Grafts surviving greater than 14 days underwent absorption study via luminal perfusion with sucrose, maltose, dextrose, Pregestimil, xylose, and cyclosporine. Average graft survival was 7.5 (range, 6 to g), 26.3 (range, 15 to 41). and 75.3 (range, 68 to 84) days, respectively,for the three immunosuppressive regimens. Absorption of carbohydrate was documented by serial measurement of serum glucose in fasted primates after infusion of maltose, dextrose, sucrose, and Pregestimil via the allograft. Absorption of xylose was similarly documented. Absorption of CyA was documented by serial serum CyA levels after luminal infusion of 40 mg/kg. A clinically applicable model for small intestinal harvest was developed that is consistent with multiple organ retrieval. A reliable allotransplantation technique and immunosuppressive regimen was developed resulting in long-term graft survival and functional absorption of carbohydrates and CyA. Copyright o 1991 by W.B. Saunders Company INDEX WORDS:
Intestinal transplantation.
From the General and Plastic Surgery Services, Walter Reed Army Medical Center, Washington, DC, and the Department of Surgery, Oregon Health Sciences University, Portland, OR Supported in part by US Army Grant No. 01288 (Dr Seyfer). Presented at the Jens G. Rosenkrantz Resident Competition at the 42nd Annual Meeting of the Surgical Section of the American Academy of Pediatrics, Boston, Massachusetts, October 6-7, 1990. The opinions expressed are those of the authors and do not reflect the opinions of the United States Army or the Department of Defense. Address reprint requests to Alan E. Seyfe, MD, FACS, Professor of Surgery, Head, Division of l&tic and Reconstructive Surgery, 3181 SWSamJackson Park Rd, Portland, OR 97201. Cop+@ o 1991 by @?B. Saunders Company 0022-3468/91/2608-0007$03.00/0
914
S
HORT GUT SYNDROME remains a serious medical problem for which there is no safe and effective therapy. Etiologic factors leading to the short gut state include necrotizing enterocolitis, midgut volvulus, trauma, embolic phenomenon, and Crohn’s disease. The current mainstay of therapy remains total parenteral nutrition, which can accommodate the nutritional demands of the short gut individual but is associated with significant morbidity and mortality from complications such as hepatic failure and sepsis. It is our belief that small intestinal transplantation represents a potentially significant therapeutic option for individuals with short gut syndrome. Experiments in small intestinal allografting have been successfully performed in rat, canine, and porcine animal models.‘” To date, no experimental allografting in a nonhuman primate model has been reported. We chose this model in an attempt to duplicate as much as possible the human anatomical and physiological milieu. In addition to studying a primate model, the purpose of this study was threefold: (1) to develop a reliable arid reproducible small intestinal transplant procedure compatible with a multiple organ procurement scenario; (2)‘to develop a consistently successful immunosuppressive regimen; and (3) to assess the function of successful allografts by measuring absorption of various carbohydrates and cyclosporine through the graft. MATERIALS
AND
METHODS
A pilot study was conducted by performing small intestinal autografts in four rhesus monkeys. A variety of vascular anastomosis were studied including graft superior mesenteric artery (SMA) and superior mesenteric vein (WV) to recipient splenic artery and vein in end-to-side fashion; graft SMA and SMV to recipient aorta and portal vein (PV); graft SMA and SMV to recipient aorta and vena cava; and graft SMA and SMV to recipient common iliac artery (CIA) and common iliac vein (CIV). The model selected consisted of an end-to-side anastomosis of the graft SMA and SMV to the recipient CIA and CIV. This model was selected for its relative technical ease and based on data demonstrating that only minimal metabolic differences result as a consequence of systemic versus portal venous anastomosis.’ A total of 14 small bowel allotransplants were performed between three pairs of baboons and 11 pairs of rhesus monkeys, each pair. consisting of one donor and one recipient animal. The animals were divided into three immunosuppressive groups. The first group consisted of’3 pairs of baboons and 3 pairs of rhesus monkeys. T?Y& immunosuppressive regimen for this group consisted of ex vivo graft irradiation of 150 rad,.cyclosporine at a dose of 20 mg/kg/d in one dose, and solumedrol2 mgikg/d in one dose.
JournalofPediatric Surgery, Vol26,
No 8 (August), 1391: pp 314-920
SMALL INTESTINE TRANSPLANT
915
IN PRIMATES
The second group consisted of 4 pairs of rhesus monkeys. The immunosuppressive regimen for this group consisted of cyclosporine at a dose of 40 mg/kg d in one dose and solumedrol at apose of 2 mg/kg/d in one dose. The third group consisted of 4 pairs of rhesus monkeys. The immunosuppressive regimen for this group consisted of cyclosporine at a dose of 40 mgikg/d in one dose, solumedrol at a dose of 2 mgkgld in one dose, and azathioprine at a dose of 5 mg/kg/d in one dose. All animals in each of the immunosuppressive regimens received one dose of cyclosporine on the evening prior to surgery and all recipients received 15 mg/kg of solumedrol intravenouslyintraoperatively prior to the restoration of blood flow to the graft. In the perioperative period, the immunosuppressive medications were given via intramuscular injection. We attempted to minimize this route of administration as much as possible because it was associated with a significant incidence of abscesses at the injection site. Once the animals were tolerating a diet, medications were given orally. Multiple attempts to persuade the primates to voluntarily take the cyclosporine by mixing it with their food were unsuccessful; ultimately, it was necessary to hand-catch the animals daily and administer the medications through nasogastric tubes. The day prior to surgery, both donor and recipient animals were maintained on a clear liquid diet. Roth primates were NPO after midnight. The donor was immobilized initially under ketamine anesthesia then brought to the operating room for a conventional inhalational anesthetic (oxygen, nitrous oxide, and isoforane) with endotracheal intubation. The anesthesia was administered by a veterinary technician and supervised by a veterinarian. The donor operation consisted of isolation of the entire small bowel on a SMA and PV pedicle formed after division of the donor’s pancreas. The aorta was cannulated distal to the take off of the SMA and cross-clamped proximal to the celiac trunk, thus allowing for simultaneous flushing of the liver, kidneys, and small intestine. An in situ flush of the graft was then performed using 500 mL of Ringer’s lactate containing 500 mg of kanamycin and lO,OtlOU of heparin, which was cooled to 4°C. After completion of the flush, 40 cm of distal jejunum and proximal ileum were excised on a vascular pedicle consisting of the SMA and PV. A cuff of aorta was taken at the proximal aspect of the SMA to allow for a Carrel patch in the recipient. Once the graft was removed, the donor primate was euthanized with a standard euthanasia solution. The lumen of the. graft was flushed with 500 mL of Ringer’s lactate at 4°C containing: 500 mg of kanamycin. The graft was then maintained in moistened gauze at 4°C until implantation.
The recipient primate was immobilized under ketamine anesthesia, brought to the operating room, and conventional endotracheal inhalational anesthesia was initiated. The recipient received 30 mg/kg of cefoxitin IV with the establishment of the anesthetic intravenous (IV) line. The abdomen was opened using a midline incision and the right CIA and CIV were isolated. Proximal and distal vascular control were obtained and the primate was systemically heparinized with 100 U/kg of heparin. Using a running 8-O prolene suture, an end-to-side anastomosis of the graft PV to the donor CIV was performed. The PV was clamped and venous outflow of the right lower extremity was restored. With running 7-O prolene, an end-to-side anastomosis was then performed between the graft SMA and the recipient’s CIA. After restoration of blood flow to the graft, the proximal and distal ends were matured as enterostomies in the right and left lower quadrants of the primate’s abdominal wall. Five hundred milliliters of warm (40°C) Ringer’s lactate was then placed into the peritoneal cavity on closing to prevent dehydration on the first postoperative day. The animals were then begun on a clear liquid diet on the first postoperative day and the diet was advanced as tolerated. The recipient animals were monitored daily for general appearance of both the primate and the stomas. A biopsy of the stomas was performed weekly to monitor for evidence of rejection. Whole blood cyclosporine levels, serum chemistries, and complete blood counts were also performed on a weekly basis. The blood draw and biopsy were performed under ketamine anesthesia. Tests of allograft absorption were performed by immobilizing the primate with ketamine anesthesia and drawing a baseline blood sample. Following this, a known quantity of the substance for absorption study was infused into one of the enterostomies. This was accomplished by threading a 1OF pediatric feeding catheter into the lumen of the allograft and injecting the substance via syringe through the feeding tube. Repeat blood draws were then performed at 30 and 60 minutes to assess for the appearance of increased quantities of the substance in the blood.
RESULTS
Survival There were four technical failures resulting in the early death (Iess than 24 hours) of the primate (Table 1). Necropsy evaluationof the primate? immediately after death demonstrated that the cause of death was venous anastomotic thrombosis and graft necrosis in
Table 1. Animal Survival Animal ID No.
Group
Species
Survival (days1
Explanation
EPA 1
1
Papio
6
Graft excised secondary to rejection
EPA 2
1
Papio
7
Graft excised secondary to rejection
EPA4
1
Papio
1
Venous thrombosis
827~
1
Macaque
0
Cause of death unknown
EPA 5
1
Papio
9
Graft excised secondary to rejection
49Y
1
Macaque
a
Graft excised secondary to rejection
47R
2
Macaque
15
Graft excised secondary to rejection
A931
2
Macaque
1
A941
2
Macaque
41
Graft rejected
Al29
2
Macaque
23
Stomas retracted, graft rejected
Venous thrombosis
x01
3
Macaque
1
Al 23
3
Macaque
68
Euthenized for presumed infection, graft viable
0765
3
Macaque
84
Euthanized for presumed infection, graft viable
0727
3
Macaque
74
Euthanized for presumed infection, graft viable
lntraabdominal hemorrhage
916
HALE ET AL
two primates, and massive intraabdominal hemorrhage from the arterial anastomosis in one primate. One primate died of unknown causes immediately after an apparently successful transplantation. Graft viability was assessed grossly on a daily basis by visual inspection of the enterostomies. Further confirmation of graft viability was the presence of bleeding and histologically viable tissue at the sites of partial-thickness biopsies, which were performed weekly on the stomas. Finally, absorption of carbohydrates through the allograft further confirmed its viability. In most instances, graft rejection was obvious on gross inspection of the stomas that appeared necrotic. There was no bleeding from the biopsy sites and permanent sections of the graft at this point demonstrated lymphoplasma cell infiltration of the submucosa, coagulation necrosis of the mucosa and thrombosis of small vessels. When possible, rejected grafts were excised and immunosuppressive medications were discontinued. However, in a few instances this was not possible, and graft necrosis led to the death of the primate. Excluding technical failures, of which there were two, primates in the first immunosuppressive regimen (cyclosporine, 20 mg/kg/d, solumedrol, 2 mg/kg/d, and ex vivo graft irradiation of 150 rad) demonstrated graft survival averaging 7.5 days (Fig 1). The stomas were generally edematous and hyperemic by the 4th postoperative day, and this universally led to frank necrosis of the allografts, which were then excised. Again excluding technical failures, of which there was one, primates in the second immunosuppressive regimen (cyclosporine, 40 mg/kg/d, solumedrol, 2 mg/kg/d) demonstrated graft survivals averaging 26.3 days. The allografts appeared entirely normal both grossly and microscopically until approximately 4 100
a0 60 Survival ,-e-mb”-,_,
(Days) 4.
0 Group I
Group 2
Group 3
Fig 1. Mean graft survival between immunosuppressive Numbers in parenthesis represent the standard deviation.
regimens.
days prior to the finding of frank necrosis of the stomas. On excision of these allografts, the small intestine was noted to be markedly shrunken and fibrotic in appearance. The increased dose of cyclosporine resulted in a mean whole blood level of 1070.6 ng/mL versus 592.3 ng/mL found in those primates comprising the first immunosuppressive regimen. It seems reasonable to expect that the increased allograft survival was due to the increased dose of cyclosporine, although deletion of the ex vivo radiation is a confounding factor. Graft survival averaged 75.3 days in the primates comprising the third immunosuppressive regimen (cyclosporine, 40 mg/kg/d, solumedrol, 2 mg/kg d, azathioprine, 5 mg/kg/d), again excluding one technical failure. However, interestingly enough, in this group there was no evidence, grossly or microscopically, of graft rejection. All three long-term survivors were euthanized for what appeared to be an inexorably deteriorating condition manifested by lethargy and failure to take adequate nutrition or water. All grafts were viable at necropsy. The clinical impression is that the primates died of an opportunistic infection of unknown identity. Bacterial cultures obtained at the time of euthanasia were negative. Unfortunately, viral cultures were not obtained. Absorption
Absorption of carbohydrates through the graft was studied by perfusing the lumen of the allograft with a known quantity of a monosacchande disaccharide, or polysaccharide, and measuring the primates’ serum glucose level at 0, 30, and 60 minutes. Absorption studies were performed under ketamine anesthesia and the anesthetic and absorption test procedure did not affect serum glucose levels in control animals (Fig 2). First to be studied was the monosaccharide dextrose. Absorption of dextrose occurs directly as a result of carrier protein transport and requires no further action by brush border enzymes to facilitate this. Primates 0765, A123, and 0727 were subjected to the absorption procedure on postoperative days 35, 53, and 14, respectively. Baseline serum glucose levels were drawn. An appropriate volume of a D,,W solution yielding a total dextrose load of 2 g/kg was infused into the allograft ostomy. Serum glucose levels were then repeated at 30 and 60 minutes, demonstrating a consistent near doubling of the baseline levels (Fig 3). D-xylose is a relatively poorly absorbed 5-carbon sugar commonly used to test for mucosal absorption capability. Primates 0765 and 0727 were tested for D-xylose absorption on postoperative days 71 and 50,
SMALL INTESTINE TRANSPLANT
IN PRIMATES
917
80 -’ Serum Xylose mgldl
60 -I 40 -’ 20 -’
0 Min
30 Min
60 Min
0 Min
Time
60 bin Time
Fig 2. Serial serum glucose levels in primates undergoing the absorption testing procedure but having no material infused into the allograft. 0, Al23 (POD 48); l, 0785 (POD 30).
Fig 4. Serial serum D-xylose levels in primates following infusion of 4 g/kg of D-xylose in the allograft lumen. 0.0765 (POD 71); +, 0727 (POD 50).
respectively. D-xylose (4 g/kg) was dissolved in 30 mL of water and infused into the allograft lumen. Baseline serum samples demonstrated no detectable D-xylose. After 60 minutes, serum specimens were redrawn demonstrating significantly elevated levels (Fig
glucose levels were again elevated at 30 and 60 minutes in all primates, demonstrating intact mucosal function (Fig 5). The disaccharide maltose was then studied in primates 0765, 0727, Al29 and twice in primate A941 on postoperative days 64, 43, 12, 14, and 35, respectively. Again, elevation in serum glucose levels was noted at both 30 and 60 minutes in all primates (Fig 6). Carbohydrate absorption was then studied after the administration of 50 mL of full-strength Pregestimil to primates 0765 and 0727 on postoperative days 49 and 28, respectively. As its carbohydrate source, Pregestimil (Mead-Johnson, Evansville, IN) contains 85% corn syrup solids (dextrose, maltose,
4). The disaccharide sucrose was next studied in primates 0765, 0727, and Al23 on postoperative days 43, 22, and 61, respectively. Sucrose is broken down by brush border enzymes prior to being absorbed as the monosaccharides glucose and fructose. These brush border enzymes are produced by intestinal mucosal cells and an intact mucosal enzymatic production capability can be inferred from successful absorption of disaccharides and oligosaccharides. Serum 110
Serum Glucose
120
Serum Glucose mgldl
80
8o
60 60 50 0 An
30 kn
60 ‘Yin
Time Fig 3. Serial serum glucose levels in primates following infusion of 2 g/kg of dextrose in the allograft lumen. 0.0765 (POD 35); l, Al23 (POD 53); n , 0727 (POD 14).
40 0 Min
30 Min
60 Yin
Time Fig 5. Serial serum glucose levels in primates following infusion of 2 g/kg of sucrose in the ellograft lumen. 0, 0765 (POD 43); l , 0727 (POD 22); I, Al23 (POD 61).
HALE ET AL
918
Serum
Glucose mgldl
I
I
I
0 Min
30 tdin
60 Yin
4
0765(64)
-
0727(43)
-
A941(14)
-o-
A!J41(35)
v
A129(12)
f
Fig 6. Serial serum glucose levels in primates following infusion of 2 g/kg of maitose in the allograft lumen.
Time other glucose polymers) and 15% tapioca starch. Each 50-mL allotment contains 4.55 g of carbohydrate, accounting for a total dose of 0.8 g/kg. Again, significant absorption of the carbohydrates was documented at 30 and 60 minutes (Fig 7). Realizing that individuals with small intestinal allograft would some day depend on that very allograft to absorb the immunosuppressive medications necessary to prevent rejection, we next studied the absorption of 40 mgikg of oral cyclosporine suspension through the graft in primates Al23 and 0765 on postoperative days 48 and 30, respectively. Both primates had not received any cyclosporine in the 24 hours preceding the absorption test. Whole blood levels of cyclosporine were elevated approximately
25% over baseline after 60 minutes in both primates (Fig 8). DISCUSSION
The,earliest investigations of small bowel transplantation were conducted by Lillehei et al in 1959. They pioneered the surgical techniques for small bowel transplantation and graft preservation that resulted in the long-term survival of autografted segments of bowel in dogs.’ Despite the early success of Lillehei et al, advancement in small bowel allografting was elusive for decades, the major obstacles being immunologic reactions mounted by and against the organ recipient. Work conducted by Manax et al in 1966 demon-
Whole Blood
600
W ng/ml
.~-,__~~--
,__“~_^__
,,,,_-~~~I___x-,-”
4aO
WV
0 Min
200 30 Min
60 Min
Time Fig 7. Serial serum glucose levels in primates following infusion of 50 mL of full-strength Pregestimil into the allograft lumen. 0, 0755 (POD 49); 9.0727fPOD 28).
0 Min
60 Min Time
Fig 8. Serial whole blood cyclosporine levels in primate following the infusion of 40 mg/kg of oral cyclosporine suspension into the ellograft lumen. 0.0785 (POD 30); 9, Al23 (POD 48).
SMALL INTESTINE TRANSPLANT
IN PRIMATES
strated survival following allografting in otherwise untreated dogs of only 8 to 15 days’ duration.6 On autopsy, marked mesenteric lymphadenopathy of the allograft led the investigators to suspect a graft-versushost reaction. This was further supported by the discovery that short segments of the allograft were uniformly rejected by the recipient whereas large allografts inevitably caused recipient death prior to rejection. It remained for Monchik and Russel to verify the existence of both immunologic reactions in a series of studies using highly inbred strains of rats and hybrid Fl offspring.7 Using this model, they were able to study unilateral reactions of rejection and graft-versus-host disease. Furthermore, they were able to demonstrate the elimination of graft-versushost disease in allografted dogs following irradiation of the graft with 150 rad of external beam radiation.’ Prior to the advent of cyclosporine in the 198Os, successful small bowel allotransplantation remained elusive. The availability of cyclosporine rekindled the interest in small bowel transplantation, especially in light of the reports that it may prevent acute graftversus-host disease in bone marrow transplants.9~10 The use of cyclosporine has resulted in marked increase in both small intestinal allograft and recipient survival as demonstrated by recent reports citing prolonged (8 months to 1 year) recipient survival and allograft viability in both rats and dogs following meticulous cyclosporine therapy.‘.’ Although these results have not been completely verified, they provide ample hope that successful small bowel transplantation will soon be a reality. The lack of a significant number of long-term survivors of small bowel allografting has presented little opportunity for the study of allograft function. The one available study with long-term survival (> 1 year) in mongrel dog isotopic allografts demonstrates that the nutritional status of these animals was significantly better than that of control dogs with iatrogenic short guts but worse when compared with normal controls.’ These conclusions were based on the comparative absorption of D-xylose and fat along with serial measurements of serum albumin. In addition, Lillehei’s original studies demonstrated that isotopic isografts were capable of maintaining the nutritional status in animal models. It remains to be shown whether transplanted small intestine can provide adequate nutritional function for the life of an animal or, perhaps more importantly, whether it can support normal growth and development in young animals. Also not available is information concerning the effects of rejection, acute or chronic, on nutritional function.
919
Kimura et al have recently demonstrated long term survival and nutritional support of outbred pigs with enterectomy and small bowel allografts using only cyclosporine for immunosuppression.3 This demonstration of long-term nutritional support is encouraging. Long-term graft survival in a porcine model was also recently demonstrated in a study by Kaneko et al.” This study again confirmed the importance of cyclosporine, in relatively high doses, in preventing rejection. In addition, systemic venous drainage of the graft did not appear to cause any significant metabolic abnormalities. To the best of our knowledge, the present study is the only reported experience with small intestinal allotransplantation in a primate model. Although it was possible to attain prolonged graft survival, the amount of immunosuppression necessary to accomplish this presumably led to the death of the primates. This appears to be a recurring theme in small intestinal allografting. Because the allografts were not functional, we cannot infer whether these grafts could have made a significant contribution to the nutritional support of a short gut primate. However, it has been demonstrated that the enzymatic machinery of the allograft was functional and allowed for the absorption of carbohydrates. The fact that cyclosporine was also absorbed through the allograft demonstrates that recipients of a small intestinal transplant should be able to be maintained on oral immunosuppression. This remains to be proven conclusively. Small intestinal transplantation has been attempted in a handful of human patients, in many instances accompanied by hepatic transplantation as well. Starzl et al” and Williams et alI3 have reported their experience with transplantation of multiple abdominal viscera to include liver, stomach, pancreas, and small bowel. A total of four transplants were performed on children with short gut syndrome and subsequent liver failure as a result of parenteral nutrition. In each report the first patient survived less than 5 days. The second patients in each study survived for 109 (Williams et al) and 193 (Starzl et al) days. Both patients suffered multiple septic episodes that were presumably due to the translocation of bacteria across the small bowel allograft mucosa. Both children died of complications due to an EpsteinBarr virus-related lymphoproliferative disorder after the intestinal grafts had been functional for a short period of time. Unequivocal success has remained elusive, although there has been a recent report of an apparently successful hepatic/small intestinal allotransplantation.14 It appears that we are at the dawn of an age where
HALE ET AL
920
small intestinal allotransplantation can be used to treat short gut individuals prior to the time at which they are experiencing a catastrophic complication of total parenteral nutrition. Further studies must be
done in order to characterize an immunosuppressive regimen that prevents rejection of the allograft but, conversely, does not overimmunosuppress the patient.
REFERENCES 1. Lee KK, Schraut WH: Structure and function of orthotopic small bowel allografts in rats treated with Cyclosporine. Am J Surg 151:55-59,1986 2. Diliz-Perez H, McClure J, Bedetti C, et al: Successful small bowel allotransplantation dogs with cyclosporine and prednisone. Transplantation 37:126-129,1984 3. Kimura K, LaRosa CA, Blank MA, et al: Successful segmental transplantation in enterectomized pigs. Ann Surg 211:158-163, 1990 4. Schraut W, Abraham S, Lee K: Portal versus caval venous drainage of small bowel allografts: Technical and metabolic consequences. Surgery 99:193-197,1986 5. Lillehei R, Goot B, Miller F: The physiological response of the small bowel of the dog to ischemia including prolonged in vitro preservation of the bowel with successful replacement and survival. Ann Surg 150:543-560,1959 6. Manax W, Lyons G, Lillehei R: Transplantation of the small bowel and stomach. Adv Surg 2:371-4181966 7. Monchik G, Russell P: Transplantation of small bowel in the
rat: Technical and immunologic considerations. Surgery 70:693702.1971 8. Cohen Z, MacGregor A, Moore K, et al: Canine small bowel transplantation: A study of the immunological responses. Arch Surg 111:248-253,1976 9. Morris P: Cyclosporine A. Transplantation 32:349-354,198l 10. Powles R, Clink H, Spence D, et al: Cyclosporine A to prevent graft vs. host disease in man after allogeneic bone marrow transplantation. Lancet 1:327-329,198O 11. Kaneko H, Hamcock W, Schweizer RT: Progress in experimental porcine small-bowel transplantation. Arch Surg 124:587592,1989 12. Starzl TE, Rowe MI, Todo S, et al: Transplantation of multiple abdominal viscera. JAMA261:1449-1457,1989 13. Williams JW, Sankary HN, Foster PF, et al: Splanchnic transplantation: An approach to the infant dependent on parenteral nutrition who develops irreversible liver disease. JAMA 261:1458-1462,1989 14. Grant D, Wall W, Mimeault R, et al: Successful small-bowel/ liver transplantation. Lancet 335:181-184,199O
Discussion A. Guttman (Montreal, Quebec): I’d like to know how much work was involved in your preliminary studies. As you know, Lillehei who first conceived small bowel transplantation, described the problem, in large animals, of a high technical failure rate. He recommended tacking the mesentery of the transplanted bowel to the back wall of the abdomen. Westbroeck reported a 30% technical failure rate in large animals. How long did it take you to get this technical proficiency and what was your failure rate? R Heineman
(Rotterdam, The Netherlands):
You
used three different immunosuppressive regimens. I’m missing a nonimmunosuppressive treatment control group. I’d like to know if you have included this essential control group. The second question is related to the first one. In the first experimental group, you irradiated the graft. I want to know if you studied the forementioned control group with irradiation. The background of this question is that I’m curious to know if you have noticed in such a group that the graft-versus-host reaction has been abolished by the irradiation and as a consequence, the rejection reaction took the upper hand and caused an accelerated rejection. Finally, have you studied the effect of irradiation on small bowel in these primates? D. Hale (response): In terms of control, we didn’t run a formal control group. We had an informal
control group in which we performed a transplant in one animal, and in which due to miscommunication the animal received no immunosuppressive medications. That animal rejected its graft in 4 days. But we did not include that in the study. We didn’t notice any difference in overall graft histology or behavior on the basis of radiation in terms of reducing graft-versushost disease (GVH). There was no evidence of GVH at the time of autopsy in the animal that had received radiation but nor was there any evidence of it in the autopsies of all the other animals, so I don’t think that would have been cause for early rejection of that group. In terms of the effect of radiation on the bowel, this has not been studied in primates but multiple studies have been performed in a canine model by Dr Schraut gauging the amount of radiation that is safe and effective in preventing GVH and, by the same token, does not harm the bowel; the threshold appears to be 1,000 rad. In terms of technical proficiency, our pilot study consisted of a total of 7 animals. We had quite a bit of practice but there was a fairly steep learning curve, mainly in terms of the venous anastomosis that caused a great degree of difficulty in the outset. But I’d say that after approximately four animals we had the technique down fairly well to where we could perform these in about 90 minutes.
A. Hale, Kathleen
Transplantation A. Waldorf,
James
in Nonhuman
Kleinschmidt,
Richard
Primates
H. Pearl, and Alan E. Seyfer
Washington, DC and Portland, Oregon 0 Small intestinal transplantation represents a potentially therapeutic procedure for individuals with short gut syndrome. The purpose of this study.was to develop a model for small intestioal transplantation in primates that is: technically feasible without microsurgery; consistent in the prevention of allograft rejection; functional in terms of nutrient absorption; and compatible with harvest for multiple organ procurement. First, autotransplantations on four rhesus monkeys were performed in order to study a vtiety of harvesting techniques and vascular anastomoses. Then, a study was performed with 14 heterotopic allotransplants in 4 baboons and 10 rhesus primates. The successful donor model consisted of division of the pancreas, harvesting the small bowel with a superior mesenteric artery and portal vein pedicle. The allograft vascular pedicle was anastomosed to the recipient’s common iliac vessels in end-to-side fashion. The graft was transplanted as an out-of-continuity loop, both ends being exteriorized as stomas providing access for absorption studies and biopsy. Three immunosuppressive regimens were tested: (1) cyclosporine A (CyA) 20 mg/kg/d, solumedrol (SML) 2 mg/kg/d, and graft irradiation (150 rad) (n = 4); (2) CyA 20 mg/kg/d and SML 2 mg/kg/d (n = 3); and (3) CyA 40 mglkgld, SML 2 mgikgld. and azathioprine 5 mg/kg/d (n = 3). There were 4 deaths due to technical error in the first 24 hours. Weekly graft biopsy, serum CyA levels, complete blood count, and automated 2Cchannel serum analysis were performed. Grafts surviving greater than 14 days underwent absorption study via luminal perfusion with sucrose, maltose, dextrose, Pregestimil, xylose, and cyclosporine. Average graft survival was 7.5 (range, 6 to g), 26.3 (range, 15 to 41). and 75.3 (range, 68 to 84) days, respectively,for the three immunosuppressive regimens. Absorption of carbohydrate was documented by serial measurement of serum glucose in fasted primates after infusion of maltose, dextrose, sucrose, and Pregestimil via the allograft. Absorption of xylose was similarly documented. Absorption of CyA was documented by serial serum CyA levels after luminal infusion of 40 mg/kg. A clinically applicable model for small intestinal harvest was developed that is consistent with multiple organ retrieval. A reliable allotransplantation technique and immunosuppressive regimen was developed resulting in long-term graft survival and functional absorption of carbohydrates and CyA. Copyright o 1991 by W.B. Saunders Company INDEX WORDS:
Intestinal transplantation.
From the General and Plastic Surgery Services, Walter Reed Army Medical Center, Washington, DC, and the Department of Surgery, Oregon Health Sciences University, Portland, OR Supported in part by US Army Grant No. 01288 (Dr Seyfer). Presented at the Jens G. Rosenkrantz Resident Competition at the 42nd Annual Meeting of the Surgical Section of the American Academy of Pediatrics, Boston, Massachusetts, October 6-7, 1990. The opinions expressed are those of the authors and do not reflect the opinions of the United States Army or the Department of Defense. Address reprint requests to Alan E. Seyfe, MD, FACS, Professor of Surgery, Head, Division of l&tic and Reconstructive Surgery, 3181 SWSamJackson Park Rd, Portland, OR 97201. Cop+@ o 1991 by @?B. Saunders Company 0022-3468/91/2608-0007$03.00/0
914
S
HORT GUT SYNDROME remains a serious medical problem for which there is no safe and effective therapy. Etiologic factors leading to the short gut state include necrotizing enterocolitis, midgut volvulus, trauma, embolic phenomenon, and Crohn’s disease. The current mainstay of therapy remains total parenteral nutrition, which can accommodate the nutritional demands of the short gut individual but is associated with significant morbidity and mortality from complications such as hepatic failure and sepsis. It is our belief that small intestinal transplantation represents a potentially significant therapeutic option for individuals with short gut syndrome. Experiments in small intestinal allografting have been successfully performed in rat, canine, and porcine animal models.‘” To date, no experimental allografting in a nonhuman primate model has been reported. We chose this model in an attempt to duplicate as much as possible the human anatomical and physiological milieu. In addition to studying a primate model, the purpose of this study was threefold: (1) to develop a reliable arid reproducible small intestinal transplant procedure compatible with a multiple organ procurement scenario; (2)‘to develop a consistently successful immunosuppressive regimen; and (3) to assess the function of successful allografts by measuring absorption of various carbohydrates and cyclosporine through the graft. MATERIALS
AND
METHODS
A pilot study was conducted by performing small intestinal autografts in four rhesus monkeys. A variety of vascular anastomosis were studied including graft superior mesenteric artery (SMA) and superior mesenteric vein (WV) to recipient splenic artery and vein in end-to-side fashion; graft SMA and SMV to recipient aorta and portal vein (PV); graft SMA and SMV to recipient aorta and vena cava; and graft SMA and SMV to recipient common iliac artery (CIA) and common iliac vein (CIV). The model selected consisted of an end-to-side anastomosis of the graft SMA and SMV to the recipient CIA and CIV. This model was selected for its relative technical ease and based on data demonstrating that only minimal metabolic differences result as a consequence of systemic versus portal venous anastomosis.’ A total of 14 small bowel allotransplants were performed between three pairs of baboons and 11 pairs of rhesus monkeys, each pair. consisting of one donor and one recipient animal. The animals were divided into three immunosuppressive groups. The first group consisted of’3 pairs of baboons and 3 pairs of rhesus monkeys. T?Y& immunosuppressive regimen for this group consisted of ex vivo graft irradiation of 150 rad,.cyclosporine at a dose of 20 mg/kg/d in one dose, and solumedrol2 mgikg/d in one dose.
JournalofPediatric Surgery, Vol26,
No 8 (August), 1391: pp 314-920
SMALL INTESTINE TRANSPLANT
915
IN PRIMATES
The second group consisted of 4 pairs of rhesus monkeys. The immunosuppressive regimen for this group consisted of cyclosporine at a dose of 40 mg/kg d in one dose and solumedrol at apose of 2 mg/kg/d in one dose. The third group consisted of 4 pairs of rhesus monkeys. The immunosuppressive regimen for this group consisted of cyclosporine at a dose of 40 mgikg/d in one dose, solumedrol at a dose of 2 mgkgld in one dose, and azathioprine at a dose of 5 mg/kg/d in one dose. All animals in each of the immunosuppressive regimens received one dose of cyclosporine on the evening prior to surgery and all recipients received 15 mg/kg of solumedrol intravenouslyintraoperatively prior to the restoration of blood flow to the graft. In the perioperative period, the immunosuppressive medications were given via intramuscular injection. We attempted to minimize this route of administration as much as possible because it was associated with a significant incidence of abscesses at the injection site. Once the animals were tolerating a diet, medications were given orally. Multiple attempts to persuade the primates to voluntarily take the cyclosporine by mixing it with their food were unsuccessful; ultimately, it was necessary to hand-catch the animals daily and administer the medications through nasogastric tubes. The day prior to surgery, both donor and recipient animals were maintained on a clear liquid diet. Roth primates were NPO after midnight. The donor was immobilized initially under ketamine anesthesia then brought to the operating room for a conventional inhalational anesthetic (oxygen, nitrous oxide, and isoforane) with endotracheal intubation. The anesthesia was administered by a veterinary technician and supervised by a veterinarian. The donor operation consisted of isolation of the entire small bowel on a SMA and PV pedicle formed after division of the donor’s pancreas. The aorta was cannulated distal to the take off of the SMA and cross-clamped proximal to the celiac trunk, thus allowing for simultaneous flushing of the liver, kidneys, and small intestine. An in situ flush of the graft was then performed using 500 mL of Ringer’s lactate containing 500 mg of kanamycin and lO,OtlOU of heparin, which was cooled to 4°C. After completion of the flush, 40 cm of distal jejunum and proximal ileum were excised on a vascular pedicle consisting of the SMA and PV. A cuff of aorta was taken at the proximal aspect of the SMA to allow for a Carrel patch in the recipient. Once the graft was removed, the donor primate was euthanized with a standard euthanasia solution. The lumen of the. graft was flushed with 500 mL of Ringer’s lactate at 4°C containing: 500 mg of kanamycin. The graft was then maintained in moistened gauze at 4°C until implantation.
The recipient primate was immobilized under ketamine anesthesia, brought to the operating room, and conventional endotracheal inhalational anesthesia was initiated. The recipient received 30 mg/kg of cefoxitin IV with the establishment of the anesthetic intravenous (IV) line. The abdomen was opened using a midline incision and the right CIA and CIV were isolated. Proximal and distal vascular control were obtained and the primate was systemically heparinized with 100 U/kg of heparin. Using a running 8-O prolene suture, an end-to-side anastomosis of the graft PV to the donor CIV was performed. The PV was clamped and venous outflow of the right lower extremity was restored. With running 7-O prolene, an end-to-side anastomosis was then performed between the graft SMA and the recipient’s CIA. After restoration of blood flow to the graft, the proximal and distal ends were matured as enterostomies in the right and left lower quadrants of the primate’s abdominal wall. Five hundred milliliters of warm (40°C) Ringer’s lactate was then placed into the peritoneal cavity on closing to prevent dehydration on the first postoperative day. The animals were then begun on a clear liquid diet on the first postoperative day and the diet was advanced as tolerated. The recipient animals were monitored daily for general appearance of both the primate and the stomas. A biopsy of the stomas was performed weekly to monitor for evidence of rejection. Whole blood cyclosporine levels, serum chemistries, and complete blood counts were also performed on a weekly basis. The blood draw and biopsy were performed under ketamine anesthesia. Tests of allograft absorption were performed by immobilizing the primate with ketamine anesthesia and drawing a baseline blood sample. Following this, a known quantity of the substance for absorption study was infused into one of the enterostomies. This was accomplished by threading a 1OF pediatric feeding catheter into the lumen of the allograft and injecting the substance via syringe through the feeding tube. Repeat blood draws were then performed at 30 and 60 minutes to assess for the appearance of increased quantities of the substance in the blood.
RESULTS
Survival There were four technical failures resulting in the early death (Iess than 24 hours) of the primate (Table 1). Necropsy evaluationof the primate? immediately after death demonstrated that the cause of death was venous anastomotic thrombosis and graft necrosis in
Table 1. Animal Survival Animal ID No.
Group
Species
Survival (days1
Explanation
EPA 1
1
Papio
6
Graft excised secondary to rejection
EPA 2
1
Papio
7
Graft excised secondary to rejection
EPA4
1
Papio
1
Venous thrombosis
827~
1
Macaque
0
Cause of death unknown
EPA 5
1
Papio
9
Graft excised secondary to rejection
49Y
1
Macaque
a
Graft excised secondary to rejection
47R
2
Macaque
15
Graft excised secondary to rejection
A931
2
Macaque
1
A941
2
Macaque
41
Graft rejected
Al29
2
Macaque
23
Stomas retracted, graft rejected
Venous thrombosis
x01
3
Macaque
1
Al 23
3
Macaque
68
Euthenized for presumed infection, graft viable
0765
3
Macaque
84
Euthanized for presumed infection, graft viable
0727
3
Macaque
74
Euthanized for presumed infection, graft viable
lntraabdominal hemorrhage
916
HALE ET AL
two primates, and massive intraabdominal hemorrhage from the arterial anastomosis in one primate. One primate died of unknown causes immediately after an apparently successful transplantation. Graft viability was assessed grossly on a daily basis by visual inspection of the enterostomies. Further confirmation of graft viability was the presence of bleeding and histologically viable tissue at the sites of partial-thickness biopsies, which were performed weekly on the stomas. Finally, absorption of carbohydrates through the allograft further confirmed its viability. In most instances, graft rejection was obvious on gross inspection of the stomas that appeared necrotic. There was no bleeding from the biopsy sites and permanent sections of the graft at this point demonstrated lymphoplasma cell infiltration of the submucosa, coagulation necrosis of the mucosa and thrombosis of small vessels. When possible, rejected grafts were excised and immunosuppressive medications were discontinued. However, in a few instances this was not possible, and graft necrosis led to the death of the primate. Excluding technical failures, of which there were two, primates in the first immunosuppressive regimen (cyclosporine, 20 mg/kg/d, solumedrol, 2 mg/kg/d, and ex vivo graft irradiation of 150 rad) demonstrated graft survival averaging 7.5 days (Fig 1). The stomas were generally edematous and hyperemic by the 4th postoperative day, and this universally led to frank necrosis of the allografts, which were then excised. Again excluding technical failures, of which there was one, primates in the second immunosuppressive regimen (cyclosporine, 40 mg/kg/d, solumedrol, 2 mg/kg/d) demonstrated graft survivals averaging 26.3 days. The allografts appeared entirely normal both grossly and microscopically until approximately 4 100
a0 60 Survival ,-e-mb”-,_,
(Days) 4.
0 Group I
Group 2
Group 3
Fig 1. Mean graft survival between immunosuppressive Numbers in parenthesis represent the standard deviation.
regimens.
days prior to the finding of frank necrosis of the stomas. On excision of these allografts, the small intestine was noted to be markedly shrunken and fibrotic in appearance. The increased dose of cyclosporine resulted in a mean whole blood level of 1070.6 ng/mL versus 592.3 ng/mL found in those primates comprising the first immunosuppressive regimen. It seems reasonable to expect that the increased allograft survival was due to the increased dose of cyclosporine, although deletion of the ex vivo radiation is a confounding factor. Graft survival averaged 75.3 days in the primates comprising the third immunosuppressive regimen (cyclosporine, 40 mg/kg/d, solumedrol, 2 mg/kg d, azathioprine, 5 mg/kg/d), again excluding one technical failure. However, interestingly enough, in this group there was no evidence, grossly or microscopically, of graft rejection. All three long-term survivors were euthanized for what appeared to be an inexorably deteriorating condition manifested by lethargy and failure to take adequate nutrition or water. All grafts were viable at necropsy. The clinical impression is that the primates died of an opportunistic infection of unknown identity. Bacterial cultures obtained at the time of euthanasia were negative. Unfortunately, viral cultures were not obtained. Absorption
Absorption of carbohydrates through the graft was studied by perfusing the lumen of the allograft with a known quantity of a monosacchande disaccharide, or polysaccharide, and measuring the primates’ serum glucose level at 0, 30, and 60 minutes. Absorption studies were performed under ketamine anesthesia and the anesthetic and absorption test procedure did not affect serum glucose levels in control animals (Fig 2). First to be studied was the monosaccharide dextrose. Absorption of dextrose occurs directly as a result of carrier protein transport and requires no further action by brush border enzymes to facilitate this. Primates 0765, A123, and 0727 were subjected to the absorption procedure on postoperative days 35, 53, and 14, respectively. Baseline serum glucose levels were drawn. An appropriate volume of a D,,W solution yielding a total dextrose load of 2 g/kg was infused into the allograft ostomy. Serum glucose levels were then repeated at 30 and 60 minutes, demonstrating a consistent near doubling of the baseline levels (Fig 3). D-xylose is a relatively poorly absorbed 5-carbon sugar commonly used to test for mucosal absorption capability. Primates 0765 and 0727 were tested for D-xylose absorption on postoperative days 71 and 50,
SMALL INTESTINE TRANSPLANT
IN PRIMATES
917
80 -’ Serum Xylose mgldl
60 -I 40 -’ 20 -’
0 Min
30 Min
60 Min
0 Min
Time
60 bin Time
Fig 2. Serial serum glucose levels in primates undergoing the absorption testing procedure but having no material infused into the allograft. 0, Al23 (POD 48); l, 0785 (POD 30).
Fig 4. Serial serum D-xylose levels in primates following infusion of 4 g/kg of D-xylose in the allograft lumen. 0.0765 (POD 71); +, 0727 (POD 50).
respectively. D-xylose (4 g/kg) was dissolved in 30 mL of water and infused into the allograft lumen. Baseline serum samples demonstrated no detectable D-xylose. After 60 minutes, serum specimens were redrawn demonstrating significantly elevated levels (Fig
glucose levels were again elevated at 30 and 60 minutes in all primates, demonstrating intact mucosal function (Fig 5). The disaccharide maltose was then studied in primates 0765, 0727, Al29 and twice in primate A941 on postoperative days 64, 43, 12, 14, and 35, respectively. Again, elevation in serum glucose levels was noted at both 30 and 60 minutes in all primates (Fig 6). Carbohydrate absorption was then studied after the administration of 50 mL of full-strength Pregestimil to primates 0765 and 0727 on postoperative days 49 and 28, respectively. As its carbohydrate source, Pregestimil (Mead-Johnson, Evansville, IN) contains 85% corn syrup solids (dextrose, maltose,
4). The disaccharide sucrose was next studied in primates 0765, 0727, and Al23 on postoperative days 43, 22, and 61, respectively. Sucrose is broken down by brush border enzymes prior to being absorbed as the monosaccharides glucose and fructose. These brush border enzymes are produced by intestinal mucosal cells and an intact mucosal enzymatic production capability can be inferred from successful absorption of disaccharides and oligosaccharides. Serum 110
Serum Glucose
120
Serum Glucose mgldl
80
8o
60 60 50 0 An
30 kn
60 ‘Yin
Time Fig 3. Serial serum glucose levels in primates following infusion of 2 g/kg of dextrose in the allograft lumen. 0.0765 (POD 35); l, Al23 (POD 53); n , 0727 (POD 14).
40 0 Min
30 Min
60 Yin
Time Fig 5. Serial serum glucose levels in primates following infusion of 2 g/kg of sucrose in the ellograft lumen. 0, 0765 (POD 43); l , 0727 (POD 22); I, Al23 (POD 61).
HALE ET AL
918
Serum
Glucose mgldl
I
I
I
0 Min
30 tdin
60 Yin
4
0765(64)
-
0727(43)
-
A941(14)
-o-
A!J41(35)
v
A129(12)
f
Fig 6. Serial serum glucose levels in primates following infusion of 2 g/kg of maitose in the allograft lumen.
Time other glucose polymers) and 15% tapioca starch. Each 50-mL allotment contains 4.55 g of carbohydrate, accounting for a total dose of 0.8 g/kg. Again, significant absorption of the carbohydrates was documented at 30 and 60 minutes (Fig 7). Realizing that individuals with small intestinal allograft would some day depend on that very allograft to absorb the immunosuppressive medications necessary to prevent rejection, we next studied the absorption of 40 mgikg of oral cyclosporine suspension through the graft in primates Al23 and 0765 on postoperative days 48 and 30, respectively. Both primates had not received any cyclosporine in the 24 hours preceding the absorption test. Whole blood levels of cyclosporine were elevated approximately
25% over baseline after 60 minutes in both primates (Fig 8). DISCUSSION
The,earliest investigations of small bowel transplantation were conducted by Lillehei et al in 1959. They pioneered the surgical techniques for small bowel transplantation and graft preservation that resulted in the long-term survival of autografted segments of bowel in dogs.’ Despite the early success of Lillehei et al, advancement in small bowel allografting was elusive for decades, the major obstacles being immunologic reactions mounted by and against the organ recipient. Work conducted by Manax et al in 1966 demon-
Whole Blood
600
W ng/ml
.~-,__~~--
,__“~_^__
,,,,_-~~~I___x-,-”
4aO
WV
0 Min
200 30 Min
60 Min
Time Fig 7. Serial serum glucose levels in primates following infusion of 50 mL of full-strength Pregestimil into the allograft lumen. 0, 0755 (POD 49); 9.0727fPOD 28).
0 Min
60 Min Time
Fig 8. Serial whole blood cyclosporine levels in primate following the infusion of 40 mg/kg of oral cyclosporine suspension into the ellograft lumen. 0.0785 (POD 30); 9, Al23 (POD 48).
SMALL INTESTINE TRANSPLANT
IN PRIMATES
strated survival following allografting in otherwise untreated dogs of only 8 to 15 days’ duration.6 On autopsy, marked mesenteric lymphadenopathy of the allograft led the investigators to suspect a graft-versushost reaction. This was further supported by the discovery that short segments of the allograft were uniformly rejected by the recipient whereas large allografts inevitably caused recipient death prior to rejection. It remained for Monchik and Russel to verify the existence of both immunologic reactions in a series of studies using highly inbred strains of rats and hybrid Fl offspring.7 Using this model, they were able to study unilateral reactions of rejection and graft-versus-host disease. Furthermore, they were able to demonstrate the elimination of graft-versushost disease in allografted dogs following irradiation of the graft with 150 rad of external beam radiation.’ Prior to the advent of cyclosporine in the 198Os, successful small bowel allotransplantation remained elusive. The availability of cyclosporine rekindled the interest in small bowel transplantation, especially in light of the reports that it may prevent acute graftversus-host disease in bone marrow transplants.9~10 The use of cyclosporine has resulted in marked increase in both small intestinal allograft and recipient survival as demonstrated by recent reports citing prolonged (8 months to 1 year) recipient survival and allograft viability in both rats and dogs following meticulous cyclosporine therapy.‘.’ Although these results have not been completely verified, they provide ample hope that successful small bowel transplantation will soon be a reality. The lack of a significant number of long-term survivors of small bowel allografting has presented little opportunity for the study of allograft function. The one available study with long-term survival (> 1 year) in mongrel dog isotopic allografts demonstrates that the nutritional status of these animals was significantly better than that of control dogs with iatrogenic short guts but worse when compared with normal controls.’ These conclusions were based on the comparative absorption of D-xylose and fat along with serial measurements of serum albumin. In addition, Lillehei’s original studies demonstrated that isotopic isografts were capable of maintaining the nutritional status in animal models. It remains to be shown whether transplanted small intestine can provide adequate nutritional function for the life of an animal or, perhaps more importantly, whether it can support normal growth and development in young animals. Also not available is information concerning the effects of rejection, acute or chronic, on nutritional function.
919
Kimura et al have recently demonstrated long term survival and nutritional support of outbred pigs with enterectomy and small bowel allografts using only cyclosporine for immunosuppression.3 This demonstration of long-term nutritional support is encouraging. Long-term graft survival in a porcine model was also recently demonstrated in a study by Kaneko et al.” This study again confirmed the importance of cyclosporine, in relatively high doses, in preventing rejection. In addition, systemic venous drainage of the graft did not appear to cause any significant metabolic abnormalities. To the best of our knowledge, the present study is the only reported experience with small intestinal allotransplantation in a primate model. Although it was possible to attain prolonged graft survival, the amount of immunosuppression necessary to accomplish this presumably led to the death of the primates. This appears to be a recurring theme in small intestinal allografting. Because the allografts were not functional, we cannot infer whether these grafts could have made a significant contribution to the nutritional support of a short gut primate. However, it has been demonstrated that the enzymatic machinery of the allograft was functional and allowed for the absorption of carbohydrates. The fact that cyclosporine was also absorbed through the allograft demonstrates that recipients of a small intestinal transplant should be able to be maintained on oral immunosuppression. This remains to be proven conclusively. Small intestinal transplantation has been attempted in a handful of human patients, in many instances accompanied by hepatic transplantation as well. Starzl et al” and Williams et alI3 have reported their experience with transplantation of multiple abdominal viscera to include liver, stomach, pancreas, and small bowel. A total of four transplants were performed on children with short gut syndrome and subsequent liver failure as a result of parenteral nutrition. In each report the first patient survived less than 5 days. The second patients in each study survived for 109 (Williams et al) and 193 (Starzl et al) days. Both patients suffered multiple septic episodes that were presumably due to the translocation of bacteria across the small bowel allograft mucosa. Both children died of complications due to an EpsteinBarr virus-related lymphoproliferative disorder after the intestinal grafts had been functional for a short period of time. Unequivocal success has remained elusive, although there has been a recent report of an apparently successful hepatic/small intestinal allotransplantation.14 It appears that we are at the dawn of an age where
HALE ET AL
920
small intestinal allotransplantation can be used to treat short gut individuals prior to the time at which they are experiencing a catastrophic complication of total parenteral nutrition. Further studies must be
done in order to characterize an immunosuppressive regimen that prevents rejection of the allograft but, conversely, does not overimmunosuppress the patient.
REFERENCES 1. Lee KK, Schraut WH: Structure and function of orthotopic small bowel allografts in rats treated with Cyclosporine. Am J Surg 151:55-59,1986 2. Diliz-Perez H, McClure J, Bedetti C, et al: Successful small bowel allotransplantation dogs with cyclosporine and prednisone. Transplantation 37:126-129,1984 3. Kimura K, LaRosa CA, Blank MA, et al: Successful segmental transplantation in enterectomized pigs. Ann Surg 211:158-163, 1990 4. Schraut W, Abraham S, Lee K: Portal versus caval venous drainage of small bowel allografts: Technical and metabolic consequences. Surgery 99:193-197,1986 5. Lillehei R, Goot B, Miller F: The physiological response of the small bowel of the dog to ischemia including prolonged in vitro preservation of the bowel with successful replacement and survival. Ann Surg 150:543-560,1959 6. Manax W, Lyons G, Lillehei R: Transplantation of the small bowel and stomach. Adv Surg 2:371-4181966 7. Monchik G, Russell P: Transplantation of small bowel in the
rat: Technical and immunologic considerations. Surgery 70:693702.1971 8. Cohen Z, MacGregor A, Moore K, et al: Canine small bowel transplantation: A study of the immunological responses. Arch Surg 111:248-253,1976 9. Morris P: Cyclosporine A. Transplantation 32:349-354,198l 10. Powles R, Clink H, Spence D, et al: Cyclosporine A to prevent graft vs. host disease in man after allogeneic bone marrow transplantation. Lancet 1:327-329,198O 11. Kaneko H, Hamcock W, Schweizer RT: Progress in experimental porcine small-bowel transplantation. Arch Surg 124:587592,1989 12. Starzl TE, Rowe MI, Todo S, et al: Transplantation of multiple abdominal viscera. JAMA261:1449-1457,1989 13. Williams JW, Sankary HN, Foster PF, et al: Splanchnic transplantation: An approach to the infant dependent on parenteral nutrition who develops irreversible liver disease. JAMA 261:1458-1462,1989 14. Grant D, Wall W, Mimeault R, et al: Successful small-bowel/ liver transplantation. Lancet 335:181-184,199O
Discussion A. Guttman (Montreal, Quebec): I’d like to know how much work was involved in your preliminary studies. As you know, Lillehei who first conceived small bowel transplantation, described the problem, in large animals, of a high technical failure rate. He recommended tacking the mesentery of the transplanted bowel to the back wall of the abdomen. Westbroeck reported a 30% technical failure rate in large animals. How long did it take you to get this technical proficiency and what was your failure rate? R Heineman
(Rotterdam, The Netherlands):
You
used three different immunosuppressive regimens. I’m missing a nonimmunosuppressive treatment control group. I’d like to know if you have included this essential control group. The second question is related to the first one. In the first experimental group, you irradiated the graft. I want to know if you studied the forementioned control group with irradiation. The background of this question is that I’m curious to know if you have noticed in such a group that the graft-versus-host reaction has been abolished by the irradiation and as a consequence, the rejection reaction took the upper hand and caused an accelerated rejection. Finally, have you studied the effect of irradiation on small bowel in these primates? D. Hale (response): In terms of control, we didn’t run a formal control group. We had an informal
control group in which we performed a transplant in one animal, and in which due to miscommunication the animal received no immunosuppressive medications. That animal rejected its graft in 4 days. But we did not include that in the study. We didn’t notice any difference in overall graft histology or behavior on the basis of radiation in terms of reducing graft-versushost disease (GVH). There was no evidence of GVH at the time of autopsy in the animal that had received radiation but nor was there any evidence of it in the autopsies of all the other animals, so I don’t think that would have been cause for early rejection of that group. In terms of the effect of radiation on the bowel, this has not been studied in primates but multiple studies have been performed in a canine model by Dr Schraut gauging the amount of radiation that is safe and effective in preventing GVH and, by the same token, does not harm the bowel; the threshold appears to be 1,000 rad. In terms of technical proficiency, our pilot study consisted of a total of 7 animals. We had quite a bit of practice but there was a fairly steep learning curve, mainly in terms of the venous anastomosis that caused a great degree of difficulty in the outset. But I’d say that after approximately four animals we had the technique down fairly well to where we could perform these in about 90 minutes.
