Meta-analysis of somatostatin, octreotide and gabexate mesilate in the therapy of acute pancreatitis A. ANDRIULLI, G. LEANDRO*, R. CLEMENTE, V. FESTA, N. CARUSO, V. ANNESE, G. LEZZI, E. LICHINO*, F. BRUNO & F. PERRI Division of Gastroenterology, Ospedale `Casa Sollievo della Sofferenza', IRCCS, San Giovanni Rotondo, Italy; and *Division of Gastroenterology, Ospedale `De Bellis', IRCCS, Castellana Grotte, Italy Accepted for publication 24 October 1997 INTRODUCTION Autodigestion of the pancreas, secondary to activation of digestive enzymes, is the conventional mechanism usually advocated in the pathogenesis of acute pancrea- titis (AP). Attempts to ®nd a treatment for AP have been directed at these events. However, despite experimental and clinical studies, the role of antiproteases and inhibitors of pancreatic secretion in this clinical setting remains unclear. After reviewing the therapeutic studies on both animal and human pancreatitis, it appeared that most agents usually worked in the animal model but were ineffective in humans; the likely explanation for this was that a type II error may have occurred in most of the human studies. 1 At present, the use of these drugs in routine clinical practice for AP is not recom- mended. Larger studies on severely ill patients would be required to either prove or disprove the therapeutic value of these agents. In the meantime, a meta-analysis of controlled studies might provide useful information. Indeed, in the two previous meta-analyses, both somat- ostatin (SS) and gabexate mesilate (FOY) appeared to reduce, respectively, mortality 2 and complications 3 in a statistically signi®cant manner. However, these analys- es included only a few trials with a limited number of patients. A literature search was performed of English and non- English (Italian, Spanish and German) publications with the intention of performing a meta-analysis of all SUMMARY Background: Autodigestion of the pancreas, secondary to the activation of digestive enzymes, is the pathogenetic mechanism of acute pancreatitis (AP). Aim: Clinical trials in which somatostatin (SS), octreo- tide (OCT) and gabexate mesilate (FOY) were used to treat patients with AP, were submitted to a meta- analytical evaluation. Five end-points were evaluated: early and overall mortality, patients with complications, complication rate, and patients who needed surgery. Results: In mild AP, no agent proved of value. In severe AP, both SS and OCT were bene®cial in improving the overall mortality: the odds ratios (OR) were, respective- ly, 0.36 (95% CI: 0.20±0.64, P  0.001) and 0.57 (95% CI: 0.35±0.88, P  0.006). FOY had no effect on either early or overall mortality, but was effective in improving complication rate (OR  0.70, 95% CI: 0.56±0.88, P  0.02), number of patients with com- plications (OR  0.61, 95% CI: 0.41±0.91, P  0.01), and number of cases submitted to surgery (OR  0.60, 95% CI: 0.39±0.92, P  0.01). SS and OCT had no effect on these latter outcomes. Conclusions: Antisecretory agents, such as SS and OCT, are able to reduce mortality without affecting compli- cations, whereas antiproteases, such as FOY, have no effect on mortality but do reduce complications. A trial exploring the ef®cacy of combining antisecretory agents with antiproteases would be of great bene®t in patients with severe AP. Correspondence to: Dr A. Andriulli, Div. Gastroenterology, Ospedale `Casa Sollievo della Sofferenza', IRCCS, Viale Cappuccini, 71013 San Giovanni Rotondo, Italy. Aliment Pharmacol Ther 1998; 12: 237±245. Ó 1998 Blackwell Science Ltd 237