- Top of page
A 50-year-old recipient of an intestinal and coecal graft with sudden onset of abdominal distention and pain, lack of bowel movements, and vomiting after closure of the diagnostic ostomy 7 months after transplantation is reported. A plain abdominal radiograph revealed pneumatosis intestinalis. An angiography excluded obstruction of large vessels, however, with absent microcirculation of the intestine. Upper gastrointestinal endoscopy showed extensive ulcerative enteritis with several spontaneous perforations. The patient underwent exploration demonstrating a nonviable intestine. The entire necrotic intestine was removed. Vascular thrombosis was excluded. Clinical data, and macroscopic and histologic features of the intestinal graft were diagnostic for necrotizing enterocolitis (NEC). Though there has been evidence for the occurrence of NEC not only in premature infants but even in older infants, children and adolescents, the presented case is, to our knowledge, the first report of NEC as etiology of late graft loss after intestinal transplantation in an adult recipient.
- Top of page
Intestinal transplantation (ITx) has been gaining acceptance as a treatment option for patients with irreversible intestinal failure as a result of short bowel syndrome or functional intestinal failure.
Novel tacrolimus-based immunosuppressive regimens including monoclonal anti-IL-2 receptor antibodies and/or rapamycin have led to improved 1-year patient and graft survival rates reaching approximately 70–80% (1–3). Acute rejection with subsequent septic complications remains the most prominent risk factor following intestinal transplantation.
Necrotizing enterocolitis (NEC) is a gastrointestinal emergency with a mortality rate varying between 10 and 30% (4) that primarily affects premature infants weighing less than 1500 g at birth. The etiology remains unclear, and generally accepted prevention and treatment strategies are lacking. Increasing evidence suggests that the risk factors of NEC are namely bacterial colonization, intestinal ischemia/hypoxia, intestinal stasis, and formula feeding. They stimulate proinflammatory mediators, which in turn activate a series of events at the cellular level leading to necrosis of the bowel (4). Additional studies propose that the intestinal host defense in the premature infant is deficient. Following the activation of the local inflammatory cascade, the risk for the initiation of NEC is increased. However, there has been evidence for the occurrence of NEC even in older infants, children and adolescents (5). Khan et al. recently reported two cases of NEC secondary to multivisceral transplantation (6) in two adolescent patients aged 13 and 17 years with onset 2 weeks after transplantation.
We report on a 50-year-old patient with a sudden onset of severe abdominal complaints on postoperative day (POD) 2 after closure of the diagnostic ostomy 7 months after primary ITx. To our knowledge, this is the first report of NEC as etiology of late graft loss after intestinal transplantation in an adult recipient.
A 50-year-old-patient underwent combined transplantation of the intestine and the adjacent coecum in May 2001 due to a desmoid tumor and familial adenomatous polyposis (Gardner's disease) necessitating subtotal intestinal resection and colectomy. The postoperative course was complicated by an anastomotic leakage of the colo-anal junction leading to prolonged peritonitis, septic multisystem organ failure, and long-term intensive care treatment. Induction immunosuppression consisted of tacrolimus, rapamycin, prednisolone, ATG, and daclizumab. The patient received early enteral immunonutrition by postoperative day (POD) 0. The patient was entirely off parenteral nutrition by month 4, gained weight, and recovered completely. He was discharged from hospital 5 months after the transplantation.
He was then electively readmitted to our hospital in order to close the ostomy 7 months after the ITx. A limited surgical approach was applied including peristomal incision, preparation of 10 cm of the terminal portion of the intestine, and resection of the latter by a gastrointestinal stapler. Graft histology was regular. After an uneventful intraoperative course and postoperative phase, the patient complained of abdominal distention and abdominal pain on POD 2. Enteral nutrition was discontinued. The patient's abdomen was distended with palpable bowel loops and missing bowel sounds. Within a few hours, the patient became ill-appearing and started vomiting. A plain abdominal radiograph revealed pneumatosis intestinalis. An angiography excluded obstruction of large vessels, however, with absent microcirculation of the intestine. An upper gastrointestinal endoscopy revealed an ulcerative enteritic state of the graft with discolorization of the mucosa similar to mesenteric infarction. There were several perforations with acute peritonitis in the abdominal cavity visible by endoscopy. Histology displayed features of transmural necrosis with slight cryptitis but without typical signs of acute rejection. As the patient's state further declined accompanied by a dramatic worsening of pulmonary function because of aspiration, and clinical signs of a preseptic condition, he was intubated, stabilized hemodynamically, and transferred immediately to emergency exploration. The intestinal and coecal grafts were impressively distended and bluishly discolorized. Additionally, the recipient's own residual jejunum as well as the distal parts of his duodenum were affected and accompanied by peritonitis of the entire abdominal cavity. There were no signs of arterial or venous obstruction with regular arterial pulses. The intestinal graft as well as the residual recipient's jejunum and 10 cm of the duodenum were removed. Swabs taken during exploration identified Enterococcus fecium. The macroscopic appearance of the resected intestine tallied with pneumatosis intestinalis. Microscopically, an ulcerative, necrotizing enteritis of the entire transplanted intestine and recipient's jejunum and duodenum with accompanying vasculitis and fibrinous and purulent peritonitis were diagnosed (Figure 1). There were no signs of acute rejection and no signs of arterial or venous obstruction in the larger vessels.
Figure 1. HE-staining of the explanted graft (×10). Microscopically, an ulcerative, necrotizing enteritis with massive neutrophilic infiltration of the entire transplanted intestine and recipient's jejunum and duodenum with accompanying vasculitis and fibrinous and purulent peritonitis were diagnosed. In concordance with the angiographic findings, histologic assessment revealed microthrombi in submucosal vessels, confirming the picture of a nearly absent microcirculation.
Download figure to PowerPoint
Determination of mediators of inflammation showed an impressive increase from preoperative values [IL-8: 8 ng/L (ref. < 8 ng/mL); IL-6 6.7 pg/mL (ref. < 30 pg/mL); procalcitonin 149 pg/mL (ref. < 500 pg/mL)] to POD 2 [IL-8: 5938 ng/L (ref. < 8 ng/mL); IL-6 1000 pg/mL (ref. < 30 pg/mL); procalcitonin 10994 pg/mL (ref. < 500 pg/mL)].
The patient stabilized temporarily but died ultimately as a result of cerebral bleeding caused by sepsis-induced pancytopenia 2 months later.
- Top of page
Necrotizing enterocolitis is the most common gastrointestinal emergency occurring in neonates. This report, however, suggests that potent immunosuppression after ITx might induce a condition in which no classic age restrictions for the occurrence of NEC exist.
Because of its rarity, NEC after ITx has to be differentiated from the most relevant and prominent risk factors for graft failure, mainly acute rejection, viral enteritis, and vascular complications. These conditions could be ruled out in the presented case. Dynamics of onset and the composition of clinical signs and parameters, as well as macroscopic and histologic features of the intestinal graft were diagnostic for NEC. In contrast to the cases reported by Khan et al. (6), NEC occurred several months after ITx in an adult patient with normal oral food intake and off parenteral nutrition. Hence, different etiologies have to be considered.
Caplan et al. (4) recently reviewed the impact of known risk factors of NEC, namely bacterial colonization, intestinal ischemia/hypoxia, intestinal stasis, and formula feeding on the stimulation of proinflammatory mediators that activate a series of events leading to necrosis of the bowel. His group demonstrated an important role of platelet-activating factor (PAF) in an animal model of intestinal injury antagonized by PAF-acetylhydrolase naturally occurring in breast milk but not in formula feeding. The impact of oral feeding either in terms of the applied formula and feeding regime is still discussed controversially. A recent report has shown that careful oral feeding decreased the rate of NEC substantially (7). Other authors have found no difference in the incidence of NEC between groups with slow or fast feeding advancement (8).
Recent findings suggest that high-risk infants are at risk for overgrowth of specific pathogens that could then initiate the inflammatory cascade resulting in NEC (9). Additionally, several specific growth factors, such as erythropoietin, epidermal growth factor, heparin-binding epidermal growth factor, and hepatocyte growth factor have been demonstrated to exert important effects on developing intestine, and may protect against the final common pathway of NEC. Furthermore, fetal intestinal epithelial cells and tissue produce significantly more IL-8 in response to endotoxin and IL-1 than do mature cells and tissue (10), however, without being deficient for IL-1 receptor antagonist and IL-10 (11). A number of studies support the hypothesis that following stress, causing increased iNOS mRNA expression (12), activation of specific proinflammatory mediators with inhibition of counter-regulatory compounds, such as IL-11 and IL-12 (4), may promote the initiation of NEC. Deficiency of intestinal trefoil factor, a component of host defense that plays a key role in intestinal wound repair and protection of the mucosal barrier from microbial invasion (13,14), may further contribute. It remains speculative which of these models can be transferred into the setting of an adult ITx recipient developing NEC. Postoperative intestinal stasis or a selection of specific pathogens induced by perioperative antibiotic prophylaxis may be hypothesized as the leading cause in the presented case.
It is also unclear which preventive strategies should be proposed in ITx. IgA supplementation, enteral antibiotic prophylaxis, early postnatal dexamethasone, and egg phospholipid supplementation of neonatal formula have been reported to be protective in newborn infants. However, even more interesting are the approaches favoring probiotic supplementation (4). The efficacy of Bifidobacterium infantis supplementation in reducing the incidence of NEC was attributed to a protection against activation of endotoxin and the inflammatory cascade and a decreased fermentation of lactose to butyric acid (15) and was clinically proven in human newborns (16). This concept has already been implemented in our ITx program. Moreover, we believe that this concept is favorable compared with enteral antibiotic prophylaxis with respect to the side-effects and development of resistant bacteria species, as it has been proven to support the regeneration of ischemia/reperfusion injury (17) and to reduce infectious complications in patients undergoing liver transplantation (18).
In summary, despite the application of concepts considered at least partly protective in reducing the incidence of NEC, we report a case of NEC in an adult ITx recipient secondary to elective closure of the ostomy 7 months after intestinal transplantation. The exact cause of this deleterious complication after ITx has not yet been elucidated.