- Top of page
- Patients and Methods
Antibody-mediated rejection (AMR) consensus criteria are defined in kidney and heart transplantation by histological changes, circulating donor-specific antibody (DSA), and C4d deposition in affected tissue. AMR consensus criteria are not yet identified in small bowel transplantation (SBTx). We investigated those three criteria in 12 children undergoing SBTx, including one retransplantation and four combined liver-SBTx (SBTx), with a follow-up of 12 days to 2 years. All biopsies (91) were evaluated with a standardized grading scheme for acute rejection (AR), vascular lesions and C4d expression. Sera were obtained at day 0 and during the follow-up. C4d was expressed in 37% of biopsies with or without AR, but in 50% of biopsies with severe vascular lesions. In addition, vascular lesions were always associated with AR and a poor outcome. All children with AR (grade 2 or 3) observed before the third month died or lost the graft. DSA were never found in any studied sera. We found no evidence that C4d deposition was of any clinical relevance to the outcome of SBTx. However, the grading of vascular lesions may constitute a useful marker to identify AR that is potentially resistant to standard treatment, and for which an alternative therapy should be considered.
- Top of page
- Patients and Methods
Small bowel transplantation (SBTx) has now become a therapeutic option for patients with permanent intestinal failure (1,2). Factors influencing outcome after SBTx include the clinical status prior to the procedure, and the occurrence of acute and chronic intestinal allograft rejection (1,3).
The rejection of transplanted organs is induced by two immunologic pathways that are mediated by T cells and antibodies (4). T-cell-mediated rejection is morphologically well defined in all transplanted organs. The lesions caused by antibody-mediated rejection (AMR) are well recognized and have been characterized in some organs such as kidney and heart allograft, but they have not been described in the context of small bowel allografts (5–11).
Complement activation within the classical pathway can be triggered by antidonor antibodies during AMR (12). The activation of C4 generates C4b, which is cleaved into C4d and C4c. C4d binds covalently to target structures (endothelial cells or vascular basement membrane) and remains present for several days or weeks. In contrast, C4c is eliminated via the blood stream soon after its production. A diagnosis of AMR requires histological changes, and demonstration of the presence of C4d in tissue biopsies. AMR is then confirmed by the detection of donor-specific antibodies (DSA)(6,13).
Although the complement degradation product, C4d, has become an important marker of AMR in renal and heart allograft biopsies, its significance in liver and lung transplantation and in the setting of SBTx remains unclear (13–16). Only a small number of publications have focused on AMR in intestinal allograft recipients (10,11,16). During one study using C4d staining, the authors concluded there was a limited relevance of C4d (16). Another study suggested that vascular injury is partially associated with humoral pre-sensitization and could be a form of acute vascular rejection (11). In a third study, the development of DSA was shown to be significantly associated with acute rejection (AR) episodes (10).
As the lesions due to AMR in the kidney differ from those described in cellular rejection, we have hypothesized that humoral rejection after SBTx could appear as a specific, but not yet recognized, histological pattern associated with a characteristic C4d mucosal capillary deposition or vascular alterations (4,10,11).
Thus, the aim of this study was to prospectively investigate the significance of C4d deposition and DSA, and their relationships in a setting of AMR in a series of children following SBTx.
- Top of page
- Patients and Methods
A-C4d expression in normal control small bowel specimens (Figure 1B)
C4d expression was observed on small arterial branches of the lamina propria and submucosa, and on the germinal center of lymphoid aggregates. Small veins and mucosal capillaries were C4d-negative.
At 1 year, patient and graft survival rates following LSBTx were both 25%. Two patients died as a result of acute small bowel rejection with surgical complications and renal failure, and one because of liver insufficiency due to biliary lesions.
At 1 year, patient and graft survival rates following SBTx reached 88% and 86%, respectively. One patient died of surgical complications, and one graft was removed because of uncontrolled rejection:
Acute grade 2 or 3 rejection before the third month: patients 1, 2, 3, 6.
Three out of four children lost their grafts because of acute cellular rejection before the third month. In all these cases, acute grade 2 or 3 cellular rejection with grade 3 vascular lesions was observed, with or without C4d staining. Two patients died before the third month and the graft was removed from the third patient at D65.
During follow-up of the fourth patient, there were two grade 3 AR episodes with grade 2 vascular lesions, with or without C4d expression, which were associated with liver insufficiency due to biliary lesions. These pathologies led to a fatal outcome on D360:
AR grades 2 and 3 after the third month: patient 8.
At D270 and D275, grade 1 and 3 AR with grade 1 and 2 vascular lesions, without C4d expression, was observed. AR was brought under control with corticosteroids, and no lesions were observed 2 months later.
In this patient, all biopsies were C4d positive during the first month post-SBTx, with grade 0 rejection and grade 0 vascular lesions. Biopsies became C4d negative at D45 without any histological or clinical modifications:
Acute grade 1 or indeterminate rejection: patients 7, 9
In two children, grade 1 or indeterminate grade AR was diagnosed, with grade 0 vascular lesions and variable C4d expression. The 1-year follow-up period was uneventful:
No AR: patients 4, 5, 10, 11, 12
No lesions were noted in any of the biopsies examined from these five children.
However, extensive C4d expression was observed on all biopsies (4/4) from the child who died at D12 of surgical complications.
C4d was expressed focally or extensively on many biopsies from other children, but follow-up was uneventful in all cases:
Grades 1, 2, or 3 vascular lesions were always associated with AR,
Grade 2 or 3 vascular lesions were associated with grade 2 or 3 AR,
Grade 1 vascular lesions were observed with grade 1 or 3 AR,
No vascular lesions were noted in patients with indeterminate grade AR or no AR,
C4d expression was observed in approximately 35% of the biopsies examined.
Positive C4d staining of the mucosal capillaries was focally (<10%) or extensively (>50%) present, but never patchy.
In SBTx patients, C4d was expressed in mucosal capillaries in 22 out of 46 biopsies with no AR (48%), in three out of 10 biopsies with indeterminate grade AR (30%), and in four out of 8 biopsies with AR (50%).
In LSBTx patients, C4d expression was increased in parallel with the severity of AR. C4d was expressed in mucosal capillaries in five out of 25 biopsies (20%) associated with grade 2 or 3 AR:
C-Donor specific antibodies
One out of the 43 sera studied contained anti-HLA B antibodies, but no DSA. This serum was obtained from the child who underwent repeated transplantation surgeries (patient 8), and was associated with focal C4d expression on the biopsy with no signs of AR during the day of the biopsy.
All other serum samples were negative.
- Top of page
- Patients and Methods
Specific histological changes, including the detection of C4d deposition in tissue sections, and the presence of serum DSA are of value for the diagnosis of AMR in heart and kidney allografts. In this prospective study of 12 small bowel recipients, we aimed to assess the relevance of C4d staining on intestinal biopsies, in association with the detection of DSA and histological changes.
The different risks accompanying the rejection of transplanted organs are well recognized (23). Both clinical and experimental data have demonstrated that intestinal allografts elicit a strong immune response (24). The severity of this immune response contributes to the lowered graft and patient survival rates in patients undergoing SBTx, especially when compared with other organs such as the heart or kidney (23–26). However, because HLA compatibility is not a criterion for intestinal transplantation, a small bowel allograft appears to be fundamentally different from other organs. This was also our experience, as 30% of our patients died during the first year posttransplantation. Furthermore, in one of the surviving eight patients, the graft was removed less than 3 months after transplantation. Our results concerning patient and graft survival following SBTx (around 90% and 80%, respectively) are quite similar to those described in the literature, but were worse in LSBTx patients. During this study, three out of four patients died within the first year, compared to 20% in the literature (23). However, it would be necessary to study a larger number of LSBTx patients in order to be able to draw any clear conclusions, because follow-up was uneventful in the fourth patient after a period of 15 months.
In addition, the response to AMR differs in terms of the organs being transplanted. Hyper-AR can occur following kidney allografts in patients with preformed circulating antidonor alloantibodies, while liver allografts are resistant to this mechanism of injury (27). In our patients, specific antidonor alloantibodies and hyper-AR were never observed.
C4d tissue deposition was less frequent in LSBTx patients than in those undergoing SBTx (5/25 and 29/66, respectively), suggesting that the liver allograft could absorb the DSA. According to this hypothesis, C4d deposits would only be seen in the intestine when the absorptive capacity of the liver was saturated. Our future studies will adequately address this issue through the use of frozen liver biopsies.
In the setting of SBTx, it was observed that C4d could be expressed independent of the severity of lesions. C4d was expressed in approximately 50% of cases, with or without AR. In LSBTx, C4d was only expressed in mucosal capillaries of biopsies with acute grade 2 or 3 cellular rejection, while specimens with indeterminate or 0 grade AR were always C4d negative. However, the small number of patients in this study limited the interpretation of transplant outcomes. Nonetheless, in SBTx and in LSBTx, C4d staining was expressed more in biopsies with vascular lesions than in those where no vascular changes were observed.
Grade 3 vascular lesions with thrombosis were observed in biopsies collected from two patients with LSBTx around D20. These vascular lesions were associated with grade 2 acute cellular rejection, but not with C4d staining. For the third patient (SBTx), grade 3 vascular lesions were later observed on D45, with negative C4d staining and grade 2 acute cellular rejection. On D50, C4d expression became positive and was associated with grade 3 vascular lesions and grade 3 acute cellular rejection. At D65, the graft was removed for acute cellular rejection.
Many other factors, such as variations of vascular pressure, drug toxicity or infection could have indirectly contributed to vascular lesions. Ischemia-reperfusion injury was expected early. Vascular changes were always associated with severe acute cellular rejection of grade 2 or 3, and never associated with isolated or with infectious lesions. In addition, this phenomenon was localized in transplanted organs suggesting this is the first step of a local pathological process. Those facts suggested that the mechanism underlying the vascular lesions seems to be closely related, if not directly connected to AR. However, a humoral component was not clearly involved.
It is well known that prior transplantation constitutes a major risk factor for the development of complement-fixing sensitization (29). In our retransplanted child, antibodies against anti-HLA B, but not DSA, were detected at D5 posttransplantation. These antibodies were associated for 40 days with C4d expression, but without any clinical or histological modifications. This patient's presentation could have reflected sub-clinical acute AMR (as described in kidney transplantation), graft accommodation or ischemia/reperfusion alterations (30).
It has been proven that ischemia/reperfusion can locally activate the classical complement pathway (31). Moreover, C4d deposition is observed in ischemic/necrotic cardiac myocytes, but not in ischemic kidney (32). This suggests that, during the early posttransplant period, the local complement cascade may already be activated with induction of specific C4d deposition. This mechanism may have occurred in the case of one patient, in whom C4d expression was positive from the day of transplantation and persisted during subsequent days.
In conclusion, we found no evidence that C4d expression was of any clinical relevance in the setting of SBTx. Biopsies exhibiting grade 2 or 3 vascular lesions were more frequently C4d-positive and always associated with AR. In addition, all children for whom grade 2 or 3 AR was observed before the third month, died or lost their graft.
As pointed out by Ruiz et al., the grading of mucosal vascular lesions in SBTx during the first posttransplant month may constitute a useful marker in identifying patients at an increased risk of losing their grafts. But because vascular changes could be the result of humoral or cellular processes, the recognition of vascular lesions may be relevant in detecting an AR that is potentially resistant to standard treatment, and for which alternative therapy should be considered.