The Incidence and the Risk Factors of Bowel Perforation Due to Chemotherapy in Gastrointestinal Diffuse Large B Cell Lymphoma

BACKGROUND: Bowel perforation after chemotherapy is a serious life-threatening complication of diffuse large B cell lymphoma (DLBCL) involving gastrointestinal tract. R Vaidya et al reported that 9 % of patients with gastrointestinal lymphoma developed a perforation, of which 55% occurred after chemotherapy, and DLBCL was the most common lymphoma associated with perforation (R Vaidya et al Ann Oncol 2013). We assume that the dose reduction of chemotherapy may have advantages to reduce the risk of gastrointestinal perforation and shorten the duration of neutropenia. This study aimed to investigate the incidence of gastrointestinal perforation after chemotherapy, whether reduced dose intensity chemotherapy could reduce the risk of the bowel perforation and how therapeutic modalities influence on the prognosis.

PATIENTS AND METHODS: We retrospectively reviewed the medical records of 348 consecutive patients who were diagnosed as having DLBCL at the Gifu University Hospital between August 2004 and April 2015. Eighty-six patients (86/348, 24.7%) were identified to have gastrointestinal involvement of DLBCL by biopsy. The prognosis of DLBCL patients with GI involvement was retrospectively investigated regarding treatments.

RESULTS: The involved sites were gastric (51 patients, 48.1%), duodenal (7 patients, 6.6%), colorectal (6 patients, 5.7%), small intestine (36 patients, 34.0%) with duplicate counts permitted. Fourteen patients (14/86, 16.3%) were provided surgical treatment prior to chemotherapy since many of them had already suffered from serious complications such as ileus, perforation and bleeding. Sixty-five patients (65/86, 75.5%) received R-CHOP or CHOP like regimen ± radiation therapy as a first-line chemotherapy. As for remaining patients, two patients received R-hyperCVAD regimen because their histology were Burkitt like, and another patients received Rituximab ± other low dose chemotherapy for their frailness. The 45 patients (45/65, 69.2%) of them received reduced dose intensity (RDI) chemotherapy at the first course because increased risk of gastrointestinal perforation with chemotherapy was concerned about by endoscopic findings. The reduction rates of chemotherapy determined by the attending physician were approximately from 10% to 50%. Six patients (6/65, 9.23%) developed gastrointestinal perforation after chemotherapy (stomach 1, duodenum 0, small intestine 5, colon 0). Three of them received full dose chemotherapy at the time of small intestinal perforation including 2 patients who developed grade 4 neutropenia after the perforation. One of them died of infection after the perforation. The perforation rate was 15.0% in full dose chemotherapy group and 6.67% in the RDI chemotherapy group (p=0.361). The median day of perforation after initiation of chemotherapy was 21 days (range; 5-63 days). The small intestine was the most common site of perforation and 4 patients (4/5, 80%) had the ulcerative type lesions in their perforation site.

CONCLUSIONS: Prior to chemotherapy, risk of GI perforation should be evaluated. We consider that small intestinal ulcerative lesion is high risk of GI perforation. The RDI chemotherapy may be an optimal therapy for patients with primary GI DLBCL with high risk. Further and larger prospective study should be required to confirm this.