Acute cellular rejection grading scheme for human gastric allografts

Acute Cellular Rejection Grading Scheme for Human Gastric Allografts MONICA GARCIA, MD, VICTOR DELACRUZ, MD, ROQUE ORTIZ, MD, ALBERTO BAGNI, MD, DEBORAH WEPPLER, RN, MSN, TOMOAKI KATO, MD, ANDREAS TZAKIS, MD, AND PHILLIP RUIZ, MD, PHD The control of acute cellular rejection (ACR) in multivisceral transplantation improves long-term survival, but monitoring this process can be challenging because different allografts can display varying forms and degrees of rejection. Criteria for ACR of small bowel and liver have been established, but a systematic analysis for ACR in stomach is lacking. For this reason we have developed a comprehensive grading scheme for the evaluation of gastric allograft rejection. The grading scheme was designed to individually grade a variety of changes in the surface epithelium, lamina propria, and glandular structures. The individual values are cumulated, and the final score determines assignment of the rejection grade. The ACR grades range from no evidence of acute cellular rejection to severe rejection. We performed a retrospective study based on 70 gastric allograft biopsies from 20 patients who received multivisceral transplantation from 1995 to 2001. We found that the scoring system showed no significant interobserver variability and allowed for an accurate designation of

the ACR grade to the gastric allografts. We found with this grading system that neither clinical symptoms nor gastric endoscopic findings could serve as specific indicators of gastric ACR. Our results also showed that there were differences in the occurrence and intensity of acute rejection between the stomach and other transplanted organs, suggesting that ACR can occur independently among different allografts of the same host. In conclusion, we find that this scheme for grading ACR in gastric transplants is objective and reproducible. This grading system will likely allow for improved correlation between gastric ACR grade and clinical symptoms, as well as improve interobserver uniformity within and between institutions. HUM PATHOL 35: 343-349. © 2004 Elsevier Inc. All rights reserved. Key words: gastric transplant, gastric rejection. Abbreviations: MVTx, multivisceral transplantation; ACR, acute cellular rejection; HPF, high-power fields.

Multivisceral transplantation (MVTx) has become a lifesaving treatment for patients with intestinal and liver failure because of irreversible gastrointestinal disease or injury. It can include liver, intestine, pancreas, kidneys, and stomach; the latter is often a necessary addition, particularly when the recipient possesses severe adhesions or fistulas in the upper abdomen.1-5 Acute cellular rejection (ACR) remains the most important complication in MVTx patients and can be challenging to assess and manage because there is anatomic variability in the intensity of this process among different transplanted organs.6-8 There may be several clinical and endoscopic features displayed that raise suspicion of acute rejection of the small intestine with confirmation being reached by histological changes of the mucosal biopsy, the features of which we and others have described elsewhere.9-11 However, the potential for ACR to be localized or increased intensity in different organs of the MVTx patient necessitates that biopsy sampling of the different allograft organs be performed when acute rejection is being considered.12,13 Definitions of the histological criteria for acute rejection of small bowel have been established9,10; how-

ever, there is little relevant literature8,14 regarding diagnostic criteria for ACR in stomach. To our knowledge, there has not been a systematic or detailed analysis of ACR changes in human gastric allograft endoscopic samples. At our institution, we previously assessed ACR in the stomach with some of the criteria used for small intestine. However, we have found that there are often organ-specific and inflammatory alterations in stomach allografts. Thus, in this study we have developed a comprehensive semiquantitative grading scheme for the assessment of gastric allograft rejection based on the data inferred from experimental studies in animals14-17 and humans18-20 and from our experience at our transplant center.

From the Department of Pathology and Division of Transplantation, University of Miami School of Medicine, Miami, FL, USA and Department of Pathology, Anatomia Patologica Policlinico di Modena, Modena, Italy. Accepted for publication October 8, 2003. Address correspondence and reprint requests to Phillip Ruiz, MD, PhD, Department of Pathology, Jackson Memorial Hospital, 1611 NW 12 Ave, East Tower, Room 2101, Miami, FL 33136. 0046-8177/$—see front matter © 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.humpath.2003.10.011

MATERIALS AND METHODS Subjects Twenty patients who underwent MVTx from 1995 to 2001 because of irreversible intestinal and hepatic failure were included in this study; 18 were or are followed in our institution, and 2 have been followed in a transplant center in Italy. Patient age ranged from 0.8 to 48 years, with an average of 18.0 years, including 11 pediatric patients (6 female, 5 male) and 9 adults (5 female, 4 male). Retrospectively, we collected data from the patients, including symptomatology, clinical impression, and endoscopic findings at the time of biopsy, to correlate those with the degree of gastric rejection. Pathologic changes in biopsies of the other allograft organs were also assessed for comparison of the intensity of the rejection in the different allograft sites.

Endoscopic Biopsies A total of 70 endoscopic gastric allograft biopsies of the 20 patients, performed from 1995 to 2002, were included in

343

HUMAN PATHOLOGY

Volume 35, No. 3 (March 2004)

TABLE 1. Grading Sheet for Stomach Allograft Biopsies

this study. The tissue samples were adequately fixed in 10% formalin, embedded in paraffin, cut into multiple levels at 4 ␮m, and stained with hematoxylin and eosin for light-microscopy evaluation by three pathologists in a blinded fashion.

Grading Scheme A semiquantitative grading scheme was developed to evaluate a variety of changes in the surface epithelium, lamina propria, and glandular structures of stomach allografts (Table 1). The histopathologic parameters assessed in the surface epithelium included the type and intensity of the inflammatory infiltrate, the presence of erosion, ulceration, and mucosal flattening or hyperplasia. The changes in the lamina propria included edema, congestion, hemorrhage, and the amount of inflammatory infiltrate, which was estimated as follows: lymphocytes—scattered, frequent, or dense infiltrate with recognizable expansion of the lamina propria; eosinophils, neutrophils, plasma cells, and activated lymphocytes were scored for scattered cells versus frequent infiltrate or cluster formation. In the glandular epithelium, we assessed the overall architecture as normal architecture, as mild distortion, as moderate disarray, or as severe architectural changes; the number of epithelial apoptotic bodies per 10 high-power fields (HPF) was scored as follows: 0 to 3 per 10 HPF, 4 to 6 per 10 HPF, 7 to 10 per 10 HPF, ⬎10 per 10 HPF, or coalescent apoptotic bodies. The cellular changes included normal cytology; mild pleomorphism or reactive changes; moderate pleomorphism; atypia or mild vacuolization of the parietal cells; and severe pleomorphism, marked atypia, or marked vacuolization of the parietal cells.

RESULTS Demographics As shown in Table 2, the indications for multivisceral transplant in the 20 patients included the following: desmoid tumor (Gardner’s syndrome; 2), volvulus, megacystis-microcolon intestinal hypoperistalsis syndrome (3), protein C deficiency, Crohn’s disease, mesenteric thrombosis (2), scleroderma, intestinal atresia, necrotizing enterocolitis (2), gastroschisis (2), microvillous inclusion disease, and congenital intestinal pseudoobstruction (2). Five of the patients were retransplanted because of primary graft failure or uncontrollable rejection. Nine (45%) of 20 patients are currently alive, after 463 to 2588 days after transplant, with an average of 1398 days. Eleven patients died 44 to 1086 days after transplant, with an average of 312 days. From these patients, there were 70 endoscopic gastric biopsies, 41 duodenal biopsies, 34 biopsies from ileum, 26 colonic biopsies, 9 biopsies from jejunum, and 1 liver biopsy. The gastric biopsies were submitted from 26 to 1033 days after transplant, with an average of 259 days.

Changes 1. Surface epithelium A. Inflammatory infiltrate i. Absent ii. Scattered/infrequent iii. Frequent B. Erosion i. Absent ii. Present C. Ulceration i. Absent ii. Present D. Atrophy/hyperplasia i. Absent ii. Present Subtotal Changes 2. Lamina Propria A. Inflammatory infiltrate i. Absent ii. Scattered/infrequent iii. Frequent iv. Dense B. Edema i. Absent ii. Present C. Congestion i. Absent ii. Present D. Hemorrhage i. Absent ii. Present Subtotal Changes 3. Glandular Epithelium A. Inflammatory infiltrate i. Absent ii. Scattered/infrequent iii. Frequent B. Apoptotic bodies/ 10 HPF i. 0-3 ii. 4-6 iii. 7-10 iv. ⬎10 or coalescent C. Architectural disarray i. Normal ii. Mild disarray iii. Moderate iv. Severe D. Cellular changes i. Normal ii. Mild pleomorphism iii. Moderate pleomorphism iv. Atypia/vacuolization Subtotal Total

Value

Score

0 1 2 0 1 0 1 0 2

Value

Score

0 1 2 3 0 1 0 1 0 2

Value

Score

0 1 2 0 1 2 3 0 1 2 3 0 1 2 3

Grading Scheme The 70 biopsies for this study had been previously interpreted but had no standardized grading scheme— previous diagnoses ranged from no evidence of ACR to acute rejection, moderate. This preliminary interpretation was based on the criteria inferred from experimental studies and the experience in our transplant center,

with awareness of the necessity of a more reliable numerical scoring system. The pathologists blinded to the diagnosis assessed the histopathologic features outlined in Table 1, the biopsies were individually scored and cumulated, and the final score (Fig 1) was correlated with the previous pathological interpretation and clin-

344

ACR GRADING FOR HUMAN GASTRIC ALLOGRAFTS (Garcia et al)

TABLE 2. Patients’ Demographic Characteristics with Gastric Allograft Biopsies Patient number

Age (yr)

Sex

Primary disease

Number of biopsies

Mean time biopsy (d)

Outcome

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

29 13 1 22 34 47 25 48 9 4 8 2 2 1 32 1 1 16 28 37

M M F M F F F F F F F F M M M M F M F M

Gardner’s syndrome Desmoid tumor Volvulus Megacystic microcolon Protein C deficiency Crohn’s disease Mesenteric thrombosis Scleroderma Mesenteric thrombosis NEC Intestinal atresia Gastroschisis Megacystic microcolon Necrotizing enterocolitis Gastroschisis Trauma Microvillus inclusion disease Megacystic microcolon Congenital intestinal pseudoobstruction Pseudoobstruction Gardner’s syndrome

5 1 10 7 3 1 1 2 9 4 2 3 2 1 1 2 7 7 1 1

111.6 755 189.3 61 27 44 83 370.5 959.2 210 39 27 26 69 381 189.5 470.4 1033.1 90 59

Alive Alive Died—pneumonia, ARDS Died—peritonitis, sepsis Died—Klebsiella pneumonia Died—Citrobacter sepsis Died—sepsis Alive Alive Died—Pseudomonas pneumonia, GVHD Died—Adenovirus pneumonia Died—Serratia sepsis, severe rejection Died—severe rejection, adenovirus infection Alive Alive Died Alive Alive Died—sepsis Alive

Abbreviations: NEC, necrotizing enterocolitis; ARDS, adult respiratory distress syndrome; GVHD, graft versus host disease.

ical presentation. The acute cellular rejection divisions for different grades were obtained by using the average score, ⫾ 2 standard deviations to obtain a score range (Table 3). The result was several categories of ACR, representative examples of which are shown in Figs 2 through 4. Pearson coefficient was used to analyze the relationship between the scored histopathologic features and the degree of acute cellular rejection (Table 4). The histological changes in the lamina propria and

glandular epithelium showed a statistically significant positive correlation, particularly the inflammatory infiltrates (lymphocytes, eosinophils, and immunoblasts). Cellular architectural changes and the presence of apoptotic bodies, the latter being the feature with higher Pearson coefficient (r ⫽ 0.924, P ⬍0.01), also were associated with degree of acute rejection. The acute cellular rejection categories include the following: (1) indeterminate for acute rejection (score, 5-9; grade 1), which shows a scattered mixed inflamma-

FIGURE 1. Histograms showing the distribution of the acute rejection score for all of the biopsies evaluated in this study. The biopsies grouped into the five grades noted on the Y axis.

345

HUMAN PATHOLOGY

Volume 35, No. 3 (March 2004)

TABLE 3. Gastric Acute Rejection Scoring and Histopathologic Features Grade

Score

Histopathology

Grade 0—no rejection

0-4

Grade 1—indeterminate for ACR

5-9

Absent to very minimal inflammatory infiltrate, normal cytology and architecture (glandular structures arranged back to back) Scattered mixed inflammatory infiltrate, edema, or focal congestion; normal architecture; normal cytolog; and no increased apoptotic bodies Increased inflammation and apoptotic bodies, mild cytologic atypia, and mild architectural disarray Prominent mixed inflammatory infiltrate, increased cytologic atypia, vacuolization of parietal cells, erosion or ulceration of the surface epithelium, major architectural disarray, and increased or coalescent apoptotic bodies Significant distortion of the architecture with subtotal destruction of the glands and gastric pits, accompanied by ulceration

Grade 2—acute rejection, mild

10-15

Grade 3—acute rejection, moderate

16-20

Grade 4—acute rejection, severe

21-26

tory infiltrate associated with edema or mild focal congestion, normal architecture, normal cytology, and no increased number of apoptotic bodies (Fig 2A and B); (2) acute rejection, mild (score, 10-15; grade 2) is characterized by increased inflammation and number of apoptotic bodies per 10 HPF, mild cytological atypia, and mild architectural disarray (Fig. 3A-D); (3) acute rejection, moderate (score, 16-20, grade 3) shows a pronounced inflammatory cell infiltrate, increased cytological atypia, vacuolization of parietal cells, erosion or ulceration of the surface epithelium, and major architectural disarray associated with increased and/or coalescent apoptotic bodies (Fig 4A and B); and (4) acute rejection, severe (based on animal studies, grade 4), is characterized by distortion of the architecture with subtotal destruction of the glands and gastric pits accompanied by ulceration; these latter changes were not observed in any of our cases. We found that this scoring system showed no significant interobserver variability and allowed for an accurate designation of the ACR grade to the gastric allografts. We also observed that the extent of the in-

flammatory infiltrate (focal vs. diffuse) does not change the grading of the cellular rejection. A focal inflammatory infiltrate of significant intensity still warrants a diagnosis of rejection. Incidence of Rejection Mild and moderate ACR was observed in 31 (44%) of 70 cases, whereas acute rejection in the duodenum, ileum, jejunum, and colon was encountered in 30 (73%) of 41 cases, 25 (75%) of 33 cases, 4 (44%) of 9 cases, and 16 (62%) of 26 cases, respectively; this shows that acute cellular rejection is more common in duodenum and ileum. Comparison of the acute rejection grade between stomach and the other allograft organs revealed that the degree of acute rejection is more severe in ileum (60.6% of the biopsies) and colon (50% of the biopsies), followed by duodenum (41.6% of the cases; Table 5). The mean degree of rejection in the duodenum, ileum, and colon (1.89, 2.02, 1.60, respectively) compared with the stomach (1.40) was higher; however, we did not find statistically significant differ-

FIGURE 2. Photomicrographs of human gastric allografts graded as indeterminate for acute rejection (grade 1). (A) A mild interstitial infiltrate with congestion and minor glandular injury (hematoxylin and eosin [H&E], ⫻100). (B) A mild mixed chronic inflammatory cell infiltrate with epithelial infiltration by inflammatory cells and apoptotic changes (H&E, ⫻200).

346

ACR GRADING FOR HUMAN GASTRIC ALLOGRAFTS (Garcia et al)

FIGURE 3. Photomicrographs of human gastric allografts graded as acute rejection, mild (grade 3). (A-D) A significant mixed chronic inflammatory cell infiltrate. There is frequent cryptitis and apoptosis, interstitial hemorrhage (A), and epithelial morphological alterations (C). (In all panels: hematoxylin and eosin, ⫻200.)

ences (P ⫽ 0.3, 0.5, and 0.1, respectively). Moreover, we applied the Pearson coefficient to show the strength of the association between the degrees of rejection in the different gastrointestinal allograft sites and showed that there is not a correlation between them; the rejection grade in the stomach could occur independently from duodenum, jejunum, ileum, and colon (P vakue for Pearson [r] ⫽ 0.15, 0.027, 0.10, and 0.32, respectively). This finding supports the principle that acute rejection can be a segmental process and that there may be variable degrees of rejection; therefore, sampling of multiple allograft sites is necessary if rejection is clinically suspected. We found available information of presentation symptoms at the time of biopsy in 38 (54.2%) of the 70 gastric biopsies. In 7 (18.4%) cases, the patients were

asymptomatic at the time of the biopsy; none of these patients showed gastric ACR. The remaining 31 (71.6%) cases had 1 or more symptoms, including nausea, vomiting, diarrhea, fever, increased or decreased stoma output, abdominal pain or distension, dysphagia, gastrointestinal bleeding, or intolerance to feedings. Twelve (38.7%) of this latter group had gastric ACR, whereas 9 (29.0%) showed ACR in other gastrointestinal organs; 6 (19.3%) were found to have other causes (ie, bacteremia, strictures), and 4 (12.9%) cases had an undetermined cause. There was information about the endoscopic appearance of the gastric mucosa in 44 (62.8%) biopsies. Eight (80%) of 10 cases with normal mucosa revealed no ACR, whereas 5 (14.7%) and 12 (35.2%) of 34 cases with mucosal changes, including inflammation, nodu-

347

HUMAN PATHOLOGY

Volume 35, No. 3 (March 2004)

FIGURE 4. Photomicrographs of human gastric allografts graded as acute rejection, moderate (grade 3). (A; ⫻100, hematoxylin and eosin [H&E]) shows a significant, effacing chronic inflammatory cell infiltrate with abundant epithelial cell destruction. (B; ⫻200, H&E) shows confluent apoptosis of the glands.

larity, erythema, and friable gastric folds, showed moderate and mild gastric ACR, respectively. These findings showed that neither clinical symptoms nor gastric endoscopic findings are specific indicators for identification of gastric acute cellular rejection. All of the patients who had documented rejection were successfully treated at the time of the biopsy.

TABLE 4. Statistical Analysis Comparing Histopathological Features With Degree of Acute Cellular Rejection Histopathological features

Pearson (r )

P

0.445 0.233 NA NA 0.402 0.033

⬍0.001* 0.066 NA NA 0.001* 0.799

0.650* 0.310 0.694* 0.589* 0.057 0.290

⬍0.001* 0.309 ⬍0.001* ⬍0.001* 0.655 0.021*

0.548* 0.451 0.177 NA 0.924* 0.724* 0.849*

⬍0.001* ⬍0.001* 0.166 NA ⬍0.001* ⬍0.001* ⬍0.001*

Surface epithelium Lymphocytes Neutrophils Eosinophils Plasma cells Erosion or ulceration Atrophy or hyperplasia Lamina propria Lymphocytes Neutrophils Immunoblasts Eosinophils Edema Congestion Glandular epithelium Lymphocytes Neutrophils Eosinophils Plasma cells Apoptosis Architecture Cell changes Abbreviation: NA, not applicable. *Statistically significant.

DISCUSSION The diagnosis of ACR depends upon clinical impression, endoscopic examination, and the analysis of biopsies, for which microscopic examination continues to be the gold standard for the diagnosis and grading of this process. To date, histological criteria for grading gastric AR have not been developed. In 1967, Thompson et al14 performed ultrastructural studies in parietal cells from canine gastric homografts, revealing that rejection followed a progressive degenerative pattern of these cells with sequestration of polyribosomes and mitochondrial degeneration. In 1992, Schmid et al8 evaluated the histological pattern of gastric allograft rejection in rats with no immunosuppression, establishing 3 degrees of AR from phase I, which was characterized by submucosal edema and occasional mononuclear and polymorphonuclear leukocytes, to phase III, consisting of severe infiltration of all layers by mononuclear and polymorphonuclear leukocytes, altered mucosal architecture with subtotal destruction of the glands, and vasculitis of the intramuscular and subserosal vessels. In that context, we have developed a grading scheme for gastric allograft acute rejection that is TABLE 5. Comparison of the Rejection Degree Between Stomach and Other Allograft Sites Rejection grade

Same (%)

Stomach more severe (%)

Other organs more severe (%)

Stomach–duodenum Stomach–jejunum Stomach–ileum Stomach–colon Stomach–liver

36.5 44.4 15.1 23 0

22 44 24.2 26.9 100

41.6 11.1 60.6 50 0

348

ACR GRADING FOR HUMAN GASTRIC ALLOGRAFTS (Garcia et al)

simple, objective, and reproducible. This scheme includes the assessment of progressive changes in the surface epithelium, lamina propria, and glandular structures, which vary from minimal changes seen with indeterminate for AR (score, 5-9) to severe architectural changes with sloughing of the mucosa (score, 21-26). We believe that this grading system will allow for improved interobserver uniformity within and between institutions. Chronic gastritis potentially can share some of the features of acute rejection. Thus, some of the features that we score (Table 1) are also assessed to evaluate whether chronic gastritis could be occurring isolated or concomitantly. In this study, we did not find any evidence of chronic gastritis in our biopsies as supported by (among other features) an absence of significant glandular atrophy, organisms, intestinal or pseudopyloric metaplasia, fatty infiltration, and acute gastritis characteristics. Lymphocytic infiltration of glandular epithelial cells with subsequent apoptosis of these epithelial cells was the most specific finding of acute rejection, a finding analogous to changes in intestinal allografts.9,10 No vasculitis was observed in the superficial vessels of these biopsies, although it is reasonable to suspect that some of the congestion and erythrocyte extravasation12 observed may reflect a vascular component to the acute rejection.10 No significant interstitial11 or vascular fibrosis was observed, so that no evidence of chronic rejection was apparent at the time that this study was performed. In this study, we have shown and confirmed previous impressions that acute cellular rejection in the gastrointestinal tract (as with other organ allografts) is a regional process that may be confined to one anatomic region or be spotty in its distribution. We found that the most commonly involved areas in the GI tract for acute rejection were the ileum and duodenum. In some cases, the stomach was affected, whereas the other allograft sites showed no ACR. These results support the concept that the ileum is more susceptible to rejection than the stomach and other gastrointestinal organs, as reported in the literature.7,8 Our data also support that the gastric symptoms and endoscopic findings may be suggestive of potential acute rejection but that biopsy is necessary to establish that diagnosis. We did not find any association between histological degree of gastric AR and any other clinical features, although symptoms and endoscopic changes are important supportive information for the evaluating pathologist. For these reasons, to achieve an accurate diagnosis, sampling of macroscopically distorted as well

as normal areas and different allograft sites is encouraged.13 REFERENCES 1. Grant D. Intestinal transplantation: Current status. Transplant Proc 21:2869-2871, 1989 2. Starzl TE, Rowe M, Todo S, et al: Transplantation of multiple abdominal viscera. JAMA 261:1449-1457, 1989 3. Kato T, Ruiz P, Thompson JF, et al: Intestinal and multivisceral transplantation. World J Surg 26:226-237, 2002 4. Abu-Elmagd K, Reyes J, Bond G, et al: Clinical intestinal transplantation: A decade of experience at a single center. Ann Surg 234:404-417, 2001 5. Tzakis AG, Kato T, Nishida S, et al: Evolution of gastrointestinal transplantation at the University of Miami. Transplant Proc 33:1545-1549, 2001 6. Starzl TE, Todo S, Tzakis A, et al: Multivisceral and intestinal transplantation. Transplant Proc 24:1217-1223, 1992 7. Sigurdsson L, Reyes J, Todo S, et al: Anatomic variability of rejection in intestinal allografts after pediatric intestinal transplantation. J Pediatr Gastroenterol Nutr 27:403-406, 1998 8. Schmid T, Thaler W, Oberhuber G, et al: Pattern of rejection after multivisceral transplantation in stomach, small bowel, and pancreas. Transplant Proc 24:1215-1216, 1992 9. Lee RG, Nakamura K, Tsamandas AC, et al: Pathology of human intestinal transplantation. Gastroenterology 110:1820-1832, 1996 10. Ruiz P, Garcia M, Pappas P, et al: Mucosal vascular alterations in isolated small bowel allografts: Relationship to humoral sensitization. Am J Transplant 3:43-49, 2003 11. Ruiz P, Perez MT, Garcia M, et al: Semiquantitative measurement of mucosal fibrosis as a means of assessing chronic injury in bowel allografts. Transplant Proc 874:34-35, 2002 12. Madariaga JR, Reyes J, Mazariegos G, et al: The long-term efficacy of multivisceral transplantation. Transplant Proc 33:12191220, 2000 13. Goulet O, Revillon Y, Brousse N, et al: Successful small bowel transplantation in an infant. Transplantation 53:940-943, 1992 14. Thompson JC, Nemhauser GM, Davidson WD, et al: Mucosal changes incident to immune rejection of canine gastric homografts. Transplantation 8:941-946, 1967 15. Alessiani M, Spada M, Vaccarisi S, et al: Multivisceral transplantation in pigs: Technical aspects. Transplant Proc 30:2627-2628, 1998 16. Murase N, Demetris AJ, Kim DG, et al: Rejection of multivisceral allografts in rats: A sequential analysis with comparison to isolated orthotopic small bowel and liver grafts. Surgery 108:880-889, 1990 17. Oberhuber G, Schmid T, Thaler W, et al: The pattern of rejection after combined stomach, small bowel and pancreas transplantation in the rat. Transpl Int 6:296-298, 1993 18. Simeoni U, Boudjem K, Desprez P, et al: Functional and histological evolution of the grafts after pediatric multiple abdominal viscera transplantation. Transplant Proc 25:1371-1373, 1993 19. Olausson M, Krantz M, Gothberg G, et al: Multivisceral transplantation in Scandinavia: Experiences from first successful fiveorgan case. Transplant Proc 33:2501-2502, 2001 20. Jaffe R, Trage JD, Zeevi A, et al: Multivisceral intestinal transplantation: Surgical pathology. Ped Pathol 9:633-654, 1989

349