Evaluation of calprotectin level in intestinal content as an early marker for graft rejection

The diagnosis of rejection after intestinal transplantation is still performed by endoscopic biopsy monitoring. Less invasive diagnostic procedures are desirable, although they are not available so far. Calprotectin, a stable cytosolic granulocyte protein, which previously was used as a marker of inflammatory processes, has been proposed to be a biochemical marker for rejection. The aim of the present work was to analyze the concordance between calprotectin levels in intestinal content and the presence of graft rejection after small bowel transplantation.

Calprotectin level was measured using a commercial ELISA kit on 137 samples of intestinal content randomly collected during endoscopies performed on 11 intestinal transplantation patients during 2 years' follow-up. Calprotectin determinations were correlated with histological and clinical findings. The cut-off level was determined retrospectively by receiver-operator curve analysis.

Based on histological findings and clinical records, samples were discerned as rejection positive (37 of 137), versus negative (35 of 137) samples or those with no clinical, endoscopic, or histological findings (65 of 137 samples). A cut-off value of 65 microg of calprotectin/g of intestinal content provided the best assay parameter according to the clinical findings: a 76% sensitivity and a 47% specificity. False positive results corresponded to patients with gastrointestinal infections (13%), systemic infections (13%), ulcers (10%), or nonspecific histological alterations of the mucosa (15%). The other false positive cases corresponded to postsurgical samples (4%), or patients with concomitant gastrointestinal symptoms (2%). Most false negative results (78%) were observed during recovery from severe acute rejection episodes, among successfully treated patients. In these cases, epithelial reconstitution and no mucosal infiltration was observed. If the latter group were discarded, sensitivity rose to 93%, and specificity, to 50% with a 96% negative predictive value. Furthermore, a weak correlation was observed between calprotectin levels and the severity of rejection.

This study confirmed the results obtained by other groups: fecal calprotectin dosage showed a good sensitivity but low specificity for the diagnosis of intestinal rejection because high calprotectin levels can also be observed in other clinical conditions. Nevertheless, it might be used as a first line for continuous evaluation of intestinal transplantation status, like other biochemical parameters that are used in kidney or liver transplantation, before considering the need for a biopsy.