587 The Safety and Efficacy of Thalidomide for the Treatment of Gastrointestinal Bleeding Refractory to Endoscopic Intervention: A Meta-Analysis

Abstract

INTRODUCTION: Thalidomide's role in patients with gastrointestinal bleeding (GI) secondary to angiodysplasia, cirrhosis, Crohn's disease, portal hypertensive gastropathy, hereditary hemorrhagic telangiectasia and left ventricular assist device which are refractory to endoscopic interventions has not been extensively studied. We conducted a meta-analysis to evaluate the safety and efficacy of thalidomide in the current body of literature. METHODS: A systematic search of Medline was performed from 2000-present. Studies enrolling adults reporting on the use of thalidomide for the treatment of refractory GI bleeding were included. The primary endpoint was the response rate, defined by reduction in frequency or resolution of GI bleed among patients treated with thalidomide. A meta-analysis of proportions was carried out to calculate pooled response rates (with calculation of 95% confidence intervals (CI)). A logistic normal random effects model was employed to account for inter-study heterogeneity and for effect sizes close to 0 and 1. Descriptive statistics were employed to outline the characteristics of included studies. RESULTS: Screening of 182 studies yielded 8 studies, enrolling 171 patients, of whom 99 were females. The most common cause for refractory GI bleed was angiodysplasia followed by left ventricular assist devices. [Figure 1] All patients had GI bleeds refractory to endoscopic therapy and most of the patient population was transfusion dependent. Patients underwent treatment with thalidomide for a median of 4 months, with doses ranging from 50-300 mg daily. The pooled response rate to thalidomide therapy was 77% (95% CI 70-83%). The pooled adverse event rate was 61% (95% CI 42-77%). [Figure 2] Thalidomide was discontinued in 8% (95% CI 1-35%) of the patients due to adverse events. The most common adverse event was peripheral neuropathy, accounting for 12% (95% CI 4-31%) of the adverse events, followed by fatigue, which accounted for 8% (95% CI 1-61%) of adverse events. [Figure 3] No treatment associated mortality was reported. CONCLUSION: Thalidomide for GI bleeding refractory to endoscopic intervention is a potentially efficacious medical therapy which can be considered before any surgical intervention is planned. Most adverse events are mild and self-limiting, though close monitoring for development of neuropathy is advised. Randomized controlled trials to validate the safety and efficacy of thalidomide for the treatment of refractory GI bleeding are warranted.