Small-Bowel Allograft Biopsies in the Management of Small-Intestinal and Multivisceral Transplant Recipients: Histopathologic Review and Clinical Correlations

Animal models of small-intestinal transplants were used by Lillehei et al1 in 1959 and Starzl and Kaupp2 in 1960. The first intestinal transplant in humans was performed by Deterling in Boston in 1964.3 The first reported case in humans was performed by Lillihei and colleagues4 in 1967. Before 1970, only 8 small-intestine transplants were reported worldwide, with a maximum graft survival time of 79 days.3 Early outcomes were quite disappointing with all patients dying of technical complications, sepsis, or rejection. The introduction of tacrolimus in 1990 improved actuarial graft and patient survival rates. Effective immunomodulatory therapy, in conjunction with improved surgical techniques and infection prophylaxis, has contributed to the clinical approval of small-intestinal and multivisceral transplants for patients who are totally parenteral nutrition–dependent and have permanent intestinal failure.5 In 2001, small-bowel transplant was approved as definitive therapy for short gut syndrome by Medicare. As of February 2012, there were 17 Medicare-approved intestinal and/or multivisceral transplant programs in the United States.6 

Intestinal transplant is considered for patients with irreversible short gut syndrome, functional intestinal failure, or severe complications of total parenteral nutrition. Indications for intestinal transplant in children include short gut following surgical resection for gastroschisis, volvulus, necrotizing enterocolitis, Hirschsprung disease, pseudo-obstruction, intestinal atresia, tufting enteropathy, and microvillous inclusion disease. In adults the most common indications include ischemia, Crohn disease, dysmotility, trauma, volvulus, and small-intestinal mesenteric tumor (eg, desmoid). Retransplantation following graft failure from severe acute or chronic rejection is another indication.7–9 

The small intestine can be transplanted in isolation or with other organs. There are 3 main types of intestinal transplants (Figure 1, A through C).10 The simplest is the isolated small-intestinal allograft, with or without colon. The colon may be included to increase absorptive capacity in patients from whom the entire native colon has had to be removed (eg, Hirschsprung disease and intestinal pseudo-obstruction). The composite liver–small-bowel allograft includes the liver, duodenum, pancreas, and small intestine, with or without colon. The most complex variant is the multivisceral graft, which includes the stomach in addition to the composite liver–small-intestine allograft, to replace a nonfunctional stomach in cases of pseudo-obstruction. A modified multivisceral allograft includes the stomach, pancreas, duodenum, and small bowel without the liver.11 

Isolated intestinal transplant (ITx) is indicated for patients with irreversible intestinal failure requiring permanent total parenteral nutrition with complications including limited central line access, multiple episodes of line sepsis, or deteriorating liver function secondary to total parenteral nutrition. Liver dysfunction in ITx is usually acute and reversible in nature. The composite liver–small-bowel transplant (LITx) is indicated for patients with intestinal failure and severe concurrent chronic liver disease or irreversible hepatic injury. The multivisceral transplant (MVTx) is reserved for patients with irreversible failure of 3 or more organs, including small bowel, with indications such as total occlusion of splanchnic circulation, visceral myopathy or neuropathy, extensive gastrointestinal polyposis, and a variety of abdominal malignancies.8,9,12,13 

The Intestinal Transplant Registry (ITR) maintains data on worldwide intestinal transplants.14 A summary of the data presented at the 10th annual ITR meeting was recently reviewed.9 In the period between April 1985 and May 2007 worldwide there were 1720 transplants including 746 small-intestine transplants (ITx), 594 combined liver and small-intestine transplants (LITx), and 380 multivisceral/modified multivisceral transplants (MVTx). Based on worldwide distribution of the different intestinal transplant types in pediatric recipients, the proportions are as follows: ITx, 36%; LITx, 35%; and MVTx, 29%. In adult recipients the proportions are ITx, 50%; LITx, 11%; and MVTx, 39%. Age at transplant ranged from less than 2 months to older than 65 years, with approximately 60% pediatric and 40% adult recipients. The short-term graft and patient survival rates have steadily increased in the past decade. Analyzing the period from 2005 to 2007, 1-year patient and graft survival rates for ITx were 90% and 80%, respectively. One-year patient and graft survival rates for LITx and MVTx were lower, both approximately 70%.14 

The Organ Procurement and Transplantation Network (OPTN) maintains transplant data for the United States. In data summarized in the OPTN 2009 annual report analyzing small-intestinal transplant data obtained on 1242 transplants performed from 1997 to 2007, the overall patient survival rates were 78% at 1 year, 57% at 5 years, and 44% at 10 years. The overall graft survival rates were 73% at 1 year, 47% at 5 years, and 34% at 10 years.7 With the implementation of new induction protocols and immunomodulatory procedures, short-term survival rates have improved. The expectation is that this will translate into increased long-term survival rates in the future. Improving the long-term outcome of small-bowel transplants is particularly challenging owing to the inability to completely control rejection, which results in graft loss, or alternatively, the development of lethal infections or posttransplant lymphoproliferative disease secondary to high immunosuppression required to control rejection.

The intestinal transplant pathologist plays a critical role within the transplant team in diagnosing the major complications of small-bowel transplants including acute and chronic rejection, infections, and posttransplant lymphoproliferative disorders. It is essential to interpret biopsy findings in the appropriate clinical context. Communication with the gastroenterologist and surgeon with respect to clinical information and endoscopic findings is essential. Clinical information includes history and clinical symptoms and abnormal laboratory values. Essential endoscopic information includes (1) locations of biopsies (jejunum versus ileum), (2) native organ versus graft and (3) description of all endoscopic findings (erosions, ulcers, irregular mucosa) and specific location of biopsies with respect to these lesions, including biopsies of endoscopically normal areas.

Protocol biopsies are obtained in the initial posttransplant period in addition to biopsies for symptoms. Protocol allograft biopsies are generally performed 2 or 3 times per week for the first month and 1 to 2 times per week for the following 2 months. On the basis of the endoscopic findings, if mucosal abnormalities are noted, biopsies are selected from both abnormal and normal-appearing regions of the allograft and also commonly from patient's native bowel. Multiple biopsies of the transplant bowel (minimum of 3 pieces) are often recommended since rejection may be focal. Biopsies should also be sufficiently deep to include adequate numbers of crypts for evaluation and ideally include the entire mucosa, the muscularis mucosae, and superficial submucosa. The endoscopist should not biopsy mucosa adjacent to the stoma, where nonspecific changes are often observed.

Biopsy specimens are collected in 10% buffered formalin, then processed for paraffin embedding by standard techniques. Routine hematoxylin-eosin sections (6–9 levels) are performed routinely. If there is a clinical concern for antibody-mediated rejection (AMR), a separate biopsy sample may be fixed in Zeus transport medium and analyzed by immunofluorescence. Immunostaining for C4d, cytomegalovirus, and adenovirus, and Epstein-Barr virus–encoded RNA in situ hybridization may be performed on paraffin-embedded sections. Trichrome staining may be performed to evaluate early fibrosis.

Preservation injury occurring in small-bowel transplants is often mild and quickly resolves. This is probably due to the intestine's rapid epithelial regeneration. Preservation injury changes are usually noted during the first few days following transplantation, and resolve within a week post transplant. There is mild superficial ischemic-type injury of the mucosa, with congestion and superficial injury or dropout of the surface epithelium, shortening of villi with regenerative changes in crypts, and minimal inflammation. Inflammation is usually neutrophilic and is associated with epithelial injury.15 

The intestine is rich in lymphoid tissue. In the first weeks and months post transplant, the donor graft's gut-associated lymphoid tissue (GALT) is infiltrated by recipient lymphocytes. The genotype of the epithelial cells of the intestine remains mainly that of the donor, making the allograft highly immunogenic. Small-bowel transplant recipients require heightened immunosuppression since the infiltrated GALT areas are sites of intense immune stimulation with the propensity for acute rejection. The region of the gut that has the highest concentration of lymphoid tissue is the ileum, which is thought to be the reason why the highest degree of acute rejection in the intestinal allograft preferentially involves this region.15 

Acute cellular rejection (ACR) is the leading cause of intestinal graft loss in the first 2 months after transplant.12 The clinical onset of ACR is usually between 1 week and 9 weeks, and most patients experience more than 1 episode. Clinical features include fever, nausea, vomiting, diarrhea, abdominal pain, and distension. The volume of stomal effluent increases. Endoscopic changes often parallel the severity of mucosal injury, ranging from edema and hyperemia to granularity with loss of delicate vascular pattern, with decreased peristalsis and focal erosions or ulcerations. Changes are not uniform; therefore, biopsies obtained from inflamed and noninflamed regions are recommended. Acute cellular rejection is also frequently a focal finding. There may be variability of rejection within a specific segment of intestine or rejection may differentially involve the proximal versus distal intestine.15,16 

In small-bowel transplants, several similar grading schemes have been published for evaluating rejection.17–20 In all these schemes the apoptotic body count (ABC), which is defined as the number of apoptotic bodies seen in 10 consecutive crypt cross-sections, is a key parameter that is useful in diagnosing rejection, particularly mild rejection or cases indeterminate for rejection. Crypt cell apoptosis is a process that is important for the physiologic regulation of the intestinal epithelium but is far more extensive in rejection. Apoptotic bodies are characterized as fragmented nuclear debris and cytoplasm that vary in morphology from well-developed “classical” exploding crypt cells to intraepithelial clusters of basophilic material. All biopsy criteria for rejection are based on morphologic evaluation on hematoxylin-eosin–stained slides. Various attempts to use other methods of recognizing apoptosis, (eg, cleaved caspase-3 immunostaining) have not been particularly contributory.21 

Increased apoptotic bodies may be noted in other inflammatory and immunologic processes such as viral enteritis or graft-versus-host disease. In the current transplant setting, patients receive a variety of medications and immunomodulatory agents, and medication-associated apoptosis also may enter the differential. Our current grading system for diagnosing rejection is based on a study by Lee and associates,15 who evaluated ABCs in a variety of conditions involving native and allograft small-intestinal biopsies. Apoptotic body counts greater than 5 were seldom seen in conditions other than allograft rejection. Important correlates for rejection were the presence of epithelial injury and mononuclear inflammation in the lamina propria. In distinguishing between nonspecific inflammatory changes and features suggestive of rejection, it is often helpful to review biopsy specimens of the allograft and compare with biopsy specimens of the native small bowel. Similar mild inflammatory changes involving both biopsy specimens, with an ABC inferior to 6, favor a diagnosis of “nonspecific enteritis.” Even with an ABC inferior to 6, if changes of focal, mild, increased inflammation and crypt injury are limited to the allograft, in conjunction with normal mucosa present in the native bowel, changes are more concerning for an emerging acute allograft rejection.

As a practical method of screening for apoptotic bodies, biopsy samples are scanned at medium power (×10 objective) and areas with the greatest concentration of apoptotic cells are searched for. Often there is adjacent focal increased inflammation or focal epithelial injury with increased mitotic activity. These areas are then evaluated under high power with formal evaluation of apoptotic counts by tallying the total number of apoptotic bodies in 10 consecutive crypts. The foci with an ABC  =  3, 4, or 5 are scrutinized in multiple deeper levels to determine if focally the threshold of ABC  =  6 is satisfied. Rejection of mild degree often is patchy, and examining multiple levels is critical in evaluating biopsy specimens. Although not formally stated in the criteria of mild rejection, in some institutions the finding of multiple apoptotic bodies within a single crypt, even with an ABC inferior to 6, may be an indication of mild acute rejection.

Once the diagnosis of acute rejection is made, the severity can be graded as indeterminate, mild, moderate, or severe on the basis of 3 parameters: (1) the extent of mucosal injury, (2) the degree of inflammatory infiltration, and (3) the crypt apoptotic body count. The criteria for different grades of rejection are outlined in the Table and described hereafter.

Negative for rejection (grade 0): Histologic features of ACR are not present. There is unremarkable mucosa with normal villous architecture, no increase in inflammation, and an ABC of 2 or less, or there are histopathologic abnormalities referable to a nonrejection etiology.

Indeterminate (grade IND): There is normal architecture or focal mild blunting of villi with slight increase in lamina propria inflammation (mixed mononuclear with focal activated lymphocytes, eosinophils, and scattered neutrophils); minor reactive glandular epithelial changes may be seen with an ABC inferior to 6. The diagnosis of indeterminate ACR implies features of minimal rejection that are insufficient for the diagnosis of mild rejection and should not be used when one is uncertain whether a biopsy specimen represents nonspecific inflammation, infection, or other nonrejection etiologies.

Mild ACR (grade 1): There is a patchy mild increase in lamina propria inflammation with increased crypt apoptosis with an ABC of 6 or greater (Figure 2, A and B). Reactive crypt epithelial changes are generally seen (mucin depletion, nuclear enlargement and hyperchromasia, cytoplasmic basophilia) but the surface epithelium is intact.

Moderate ACR (grade 2): Moderate increase in lamina propria inflammation with frequent crypt apoptotic bodies, including occasional confluent crypt apoptosis and focal crypt dropout (Figure 3). There is often moderate to marked architectural distortion and villous blunting with edema and congestion.

Severe ACR (grade 3): Moderate to marked increase in lamina propria inflammation and marked architectural distortion with crypt damage and crypt loss and erosion of the surface epithelium. Pronounced neutrophilic infiltrates may accompany mucosal breakdown. The most severe cases may progress to exfoliative rejection or widespread sloughing of the mucosa (Figure 4). In biopsy specimens that have sloughed mucosa and consist mainly of granulation tissue, it is important to find any areas that have some mucosa with residual crypts to assess for changes with morphologic features of rejection, although ABCs may be paradoxically inconspicuous.

It is essential that the biopsy findings be correlated with the endoscopic gross pathologic findings. For example, if an isolated ulcer near the stoma is biopsied and consists of granulation tissue, this is unlikely to represent severe rejection. If however the endoscopic appearance shows extensive areas with epithelial sloughing and granulation tissue is biopsied, a diagnosis of severe rejection is more probable.

It has been observed that the mononuclear inflammatory infiltrates in the lamina propria that typically accompany increased ABCs in ACR may be inapparent in biopsy specimens taken after the first few months post transplant.15 In these cases, it is even more important that interpretation of isolated increased ABCs be correlated with clinical-endoscopic data. With the use of new immunomodulatory induction regimens, such as thymoglobulin or alemtuzumab (Campath-1H, Genzyme Corporation, Cambridge, MA), variant histologic features may be associated with ACR (eg, scattered lamina propria neutrophilic or eosinophilic infiltrates and/or cryptitis associated with early ACR or absence of crypts with intact surface epithelium associated with severe ACR) (Figure 5).20 

In other solid organ transplants, such as kidney and heart allografts, antibody-mediated rejection (AMR) is well documented. Diagnostic criteria for AMR in kidney and heart transplants include positive donor-specific antibodies, characteristic histologic findings, and C4d deposition in allograft tissue.22 In the small intestine, there are limited data on humoral rejection, and the frequency and clinical significance of AMR is uncertain.

Several case reports have described hyperacute and acute humoral rejection in patients with preformed antibodies, most frequently immunoglobulin G (IgG) lymphocytotoxic antibodies.23 Wu and colleagues24 demonstrated that in a small group of ITx patients, 6 of 28 had preformed IgG lymphocytotoxic antibodies. In 5 of 6 with strong positive crossmatch grafts, cyanotic discoloration occurred immediately after reperfusion that resolved within 1 hour. Within the first 10 days after transplant, 3 of these 5 grafts demonstrated severe congestion, neutrophilic margination, and fibrin-platelet thrombi within the lamina propria microvasculature, but with no neutrophilic or necrotizing arteritis, and no deposition of C4d by indirect immunofluorescence. In contrast, the remaining ITx recipients with a weakly positive crossmatch, as well as the 22 crossmatch-negative recipients, exhibited none of these findings. With prompt augmentation of the immunosuppression therapy, histologic features of rejection for the 3 grafts with the characteristic clinicopathologic syndrome were successfully reversed with the monoclonal anti-lymphocyte preparation muromonab-CD3 (OKT3, Janssen Biotech, Inc. Horsham, PA).

As a working definition, AMR in the intestine22 is usually diagnosed during the first 2 weeks post transplant in the presence of circulating immunoglobulins (IgG). Clinical and endoscopic features strongly suggesting AMR include unexplained severe ischemic injury shortly after reperfusion, mucosal persistent diffuse congestion and hemorrhage, and a strong crossmatch for T- or B-cell lymphocytotoxic antibodies. Although C4d is used as a marker for AMR in kidney and lung allografts, its use in small-intestinal allografts has not yet been validated. In normal small intestine, C4d immunoreactivity is restricted to small arterial branches, but is negative in capillaries and venules.

The significance of diffuse C4d capillary staining in the lamina propria of intestinal allografts with respect to AMR or ACR is unclear. Two studies to date25,26 have shown lack of sensitivity and specificity of C4d immunostaining for diagnosing rejection or with regard to clinical outcome.

Chronic rejection is the main cause of late intestinal graft dysfunction and loss. Chronic rejection follows an insidious progressive course and lacks early specific clinical symptoms or mucosal findings. The common clinical presentation of chronic rejection is persistent diarrhea with nonhealing mucosal ulceration often preceded by repeated episodes of acute rejection.27 On endoscopy, mucosal folds are effaced and the bowel appears firm and fibrotic. Focal ulcers may be present. Mucosal biopsy specimens may show mild ischemic changes, low-grade apoptosis, and possibly mild fibrosis of the lamina propria. All too often, the mucosal biopsies are noncontributory, even in the setting of refractory intestinal dysfunction. In allograft biopsies with fibrosis of the lamina propria, often the etiology of fibrosis is not clear. Fibrosis may be secondary to diverse etiologies (previous episodes of rejection, ischemic injury, prior infections, medication-associated chronic injury, prior biopsy site). The diagnostic finding in chronic rejection is allograft vasculopathy in which there is marked intimal hyperplasia leading to impaired vascular perfusion of the graft.15,27 Unfortunately, this finding is seen in submucosal or mesenteric arteries, which are not normally sampled on an endoscopic biopsy. The diagnosis of chronic rejection is usually confirmed only in full-thickness biopsies or in failed grafts, which document the associated obliterative vasculopathy (Figure 6).

Immunosuppressive therapy presents a risk for opportunistic infections in transplant patients. Although similar infections occur in all solid transplant organs, infectious complications are higher in intestinal transplants, particularly for bacterial and fungal infections, likely owing to the higher level of immunosuppression required in these patients. In a study by Guaraldi et al,28 the overall rates for infections in intestinal transplant patients were 94% for bacterial, 67% for viral, and 28% for fungal infections. Bacterial infections were the most frequent and occurred the earliest (median, 11 days; range, 9–17 days) and involved primarily extraintestinal sites (vascular lines, abdominal cavity, wounds, urinary tract, respiratory tract). More than 80% of small-bowel transplant patients have a significant bacterial infection within 2 months after transplant. In a review of 124 intestinal transplant patients, bacterial infection was the direct cause of death in 18% of patients and was present in 76% of patients who died.29 Earlier studies reported higher rates of bacterial infections in patients with intestinal grafts that included colon as well as small intestine; however, more recent studies do not show this increased risk.30,31 

Viral infections are the second most frequently occurring infection and are seen most often within the first 6 months (median, 91 days; range, 65–101 days). The most common viral infections that directly involve the gastrointestinal tract include cytomegalovirus (CMV), Epstein-Barr virus (EBV), adenovirus, rotavirus, and calicivirus.32–36 

Fungal infections occur later (median, 181 days) compared to bacterial and viral infections. The data of Florescu and colleagues37 showed that fungal infections occurred in 25% of a total of 98 small-intestinal transplant recipients and Candida albicans was the most common cause. Intra-abdominal fungal infections occurred earlier (<1 month) than fungemia (>6 months) after transplant.

Since the clinical presentation due to allograft rejection or intestinal infection can be similar, yet the management of immunosuppression is diametrically opposed, it is critical to distinguish between the histopathologic changes of infection versus rejection. A brief discussion of the most common intestinal viral pathogens follows.

Adenovirus is a common cause of self-limited respiratory and gastrointestinal infections in immunocompetent hosts. In the immunocompromised small-intestinal transplant population, adenovirus infection can range from asymptomatic shedding to severe disseminated disease. Infection is more commonly seen in the pediatric population, where it has been reported in up to 24% of intestinal transplant recipients.38 Most cases affect the allograft and occur within the first 6 months after transplant. Risk factors for the development of invasive disease include younger age at transplant and more aggressive immunosuppressive therapy.38,39 Clinical signs of adenovirus enteritis include increased ostomy or stool output and unexplained fever.39 Adenovirus enteritis always involves the distal ileum but may also be seen in the jejunum and native colon.39 Histologically, there is often mild villous blunting with hyperplasia and disarray of the surface epithelium. The presence of scattered intranuclear glassy eosinophilic inclusions points to the diagnosis of adenoviral enteropathy (Figure 7). While the viral inclusions are typically present in the surface enterocytes, they also may involve the crypts if there is a high viral load. Rarely, inclusions are noted in stromal cells in cases with extensive epithelial denudation and acute and chronic inflammation resembling severe acute rejection.40 Adenovirus enteritis is associated with mild mixed inflammation in the lamina propria and slight increase of crypt apoptosis, resembling low-grade acute rejection.41 Morphologic confirmation of invasive adenovirus infection in tissue biopsy material can be performed by immunohistochemistry or electron microscopy that shows the typical crystalline lattice of 72- to 82-nm viral particles.39,41 Adenoviral enteritis is treated by a reduction in immunosuppression. It is important to recognize that increased crypt apoptosis can accompany intestinal adenoviral (and other viral) infections. A mistaken diagnosis of rejection and augmentation of immune suppression can lead to viral dissemination and potential fatality.32 

Cytomegalovirus enteritis currently occurs less frequently than adenovirus or EBV enteritis, mainly as the result of prophylactic antiviral treatment targeting CMV in the early stages post transplant.28,33,42 Cytomegalovirus enteritis may clinically mimic acute rejection with symptoms of diarrhea and epigastric pain. Multiple erosions and superficial ulcers are more commonly noted in the stomach than in the small intestine.

There is a spectrum of histologic changes with chronic active inflammation, associated with characteristic large CMV-infected cells with large amphophilic intranuclear inclusions surrounded by a clear halo and thickened nuclear membrane, and frequently, smaller intracytoplasmic eosinophilic inclusions (Figure 8). Inclusions are more commonly found in endothelial cells and stromal cells and less commonly in enterocytes. Immunohistochemical staining for CMV and polymerase chain reaction (PCR) for CMV can confirm the diagnosis of CMV enteritis. Cytomegalovirus PCR is currently routinely used to monitor infection and follow treatment.

These common pediatric intestinal pathogens may cause prolonged intestinal dysfunction in the immunosuppressed intestinal transplant recipient. Rotavirus causes seasonal infection in pediatric intestinal transplant patients. Infection can be confirmed by rapid stool antigen test.43 Caliciviral enteritis, which encompasses 2 groups of viruses (Norwalk-like virus and Sapporo virus), also can be diagnosed by reverse transcription PCR on stool and biopsy tissue.35 

Mucosal changes in enteritis effected by these intestinotropic viruses are nonspecific and include villous blunting and surface epithelial disarray with piling up of enterocytes in the absence of characteristic viral inclusions (Figure 9). Chronic nonspecific enteritis is usually present. Apoptosis is increased in the surface epithelium, which is not typically seen in ACR. However, ABC may also be increased in crypts and therefore be confused with changes of acute cellular rejection.35,44,45 In general, rotavirus- or calicivirus-associated enteritis is treated with either decrease in immunosuppression or no change in management.

Epstein-Barr virus infection commonly occurs in the transplant setting and ranges from viremia with self-limited mucosal lymphoid hyperplasia to the development of posttransplant lymphoproliferative disorders (PTLDs) with polymorphic or monomorphic features. Posttransplant lymphoproliferative disorders encompass a spectrum of lymphocytic and plasmacytic proliferations occurring in the posttransplant setting that are frequently, but not always, associated with EBV. The current WHO classification of PTLD is composed of 4 major categories: (1) early-type nondestructive PTLD, further divided into plasmacytic hyperplasia and infectious mononucleosis-like PTLD; (2) polymorphic PTLD composed of heterogeneous populations of lymphocytes and plasma cells, without a predominance of transformed cells, which does not meet criteria for lymphoma; (3) monomorphic PTLD resembling one of the non-Hodgkin lymphomas encountered in otherwise immunocompetent patients, such as diffuse large B-cell lymphoma, Burkitt lymphoma, plasma cell myeloma, or T-cell/natural killer–cell lymphoma; and (4) classical Hodgkin-type PTLD.46 Although these categories are useful for diagnosis and appropriate management, often there is a spectrum of changes ranging from early lesions to polymorphic to monomorphic lesions, making precise classification difficult.46 More detailed histopathologic assessment of PTLD is beyond the scope of this review. It is important to consider PTLD whenever there is a lymphoid infiltrate and to distinguish it from lymphoid proliferations that may accompany rejection, infection, or other conditions. Features that favor PTLD over reactive lymphoid infiltrates are the following: expansile nodules or a mass lesion, numerous transformed cells, lymphoid atypia, predominant B-cell or plasma cell infiltrate, serpiginous necrosis in the infiltrate, and/or the presence of numerous EBV-positive cells.46 Adequate evaluation includes morphologic and immunophenotypic characterization (Figure 10, A through E) including in situ hybridization for Epstein-Barr virus–encoded RNA, which labels EBV-encoded RNA transcripts in infected cells (Figures 10, F, and 11). Depending on these studies, further workup with B-cell or T-cell gene rearrangement studies to assess clonality, and/or other cytogenetic fluorescence in situ hybridization studies, may be performed.

In small-bowel transplant patients, the clinical presentation of PTLD is variable and nonspecific. Sites most commonly affected include allograft bowel, lymphatic tissue (lymph nodes, tonsils), and liver. Posttransplant lymphoproliferative disorder may present as gastrointestinal ulcers or strictures.47 Gastrointestinal ulcers may lead to bacteremia.48 

Risk factors for developing PTLD in the transplant setting include EBV seronegativity of the recipient with primary EBV infection, duration and type of immunosuppression, and type of allograft (higher risk in small intestine, heart, or lung than in liver).49 Epstein-Barr virus seropositivity in the recipient, however, is not protective against EBV infection or PTLD after intestinal transplant. It is recommended that serum EBV viral loads be monitored in both seronegative and seropositive patients. Low EBV viral loads are associated with a low risk of developing EBV-associated PTLD.50 The cumulative use of calcineurin inhibitors, or the use of muromonab-CD3 (OKT3) and antithymocyte globulin to treat refractory rejection increases PTLD risk.50,51 

Incidence, morbidity, and mortality of PTLD in intestinal transplants have significantly decreased over time. Data from the early 1990s showed high rates of PTLD in intestinal transplant patients, with PTLD occurring in 15% of adults and 40% of children.34 In an analysis of worldwide data from 1985 to 2007, 9.8% (168 of 1608) of patients developed PTLD. The incidence in children was 13.2% compared to 5.1% in adults.14 In a study of 119 pediatric intestinal transplant patients receiving intestinal transplants between 1994 and 2005, Quintini et al51 reported 12% incidence of PTLD. The incidence of PTLD is thought to have decreased in part due to EBV prophylaxis regimens and PCR monitoring for EBV viremia with preemptive therapy, and current avoidance of highly aggressive immunosuppressive management, when possible.

Treatment of PTLD in patients with a small-bowel transplant consists of reduction of immunosuppression by 25% to 50%; complete withdrawal of immunosuppression is not possible due to high rates of small-bowel rejection. Typically, steroids are maintained at lower baseline and tacrolimus levels are reduced while maintaining antiviral therapy with ganciclovir and intravenous immune globulin.52 

Endoscopic biopsy specimens are assessed weekly to monitor rejection and EBV PCR is performed weekly to monitor viral activity. In contrast to other organ transplants, it is not uncommon to have concomitant rejection while PTLD is resolving or active. In this setting, rituximab (anti-CD20 monoclonal antibody) and low-dose chemotherapy regimens are commonly used. This therapeutic approach is focused on targeting B-cell–derived PTLDs that express CD20 surface antigen. Under the current regimens using rituximab and low-dose chemotherapy options, mortality due to PTLD in intestinal transplant patients is 10% to 14%.51 

As for other-organ transplant recipients with PTLD, survival is dependent on the histologic type of PTLD (monomorphic versus polymorphic) and on the extent and location of organ involvement. Worse prognosis is associated with monomorphic PTLD, EBV-negative PTLD, and the presence of multiorgan involvement or advanced stage.

The clinical management of small-bowel transplant patients requires coordinated input from all members of the transplant team and involves the integration of clinical and laboratory parameters for early diagnosis and treatment of complications (infection, rejection, PTLD) for these patients. Although there is a need to develop sensitive and specific noninvasive methods for detecting rejection and other complications, the evaluation of histopathologic changes in the transplant allograft remains the gold standard in guiding patient management.

The authors would like to acknowledge the photographic assistance of Joseph Samet.