TRATTAMENTO DELLE VARICI ESOFAGEE. REVISIONE CRITICA DEI TRIALS RANDOMIZZATI E CONTROLLATI MEDIANTE META-ANALISI

All published randomized controlled trials of treatment of acute variceal hemorrhage in cirrhotics have been critically reviewed to verify the efficacy of various treatment modalities in controlling bleeding and in influencing short-term mortality. Data have been analyzed both by conventional and by cumulative meta-analysis. The latter allows the study of evolving trends in good and bad effects of treatments with time. Methods - From the Index Medicus and the Abstract books of meetings of national and international Gastroenterology and Hepatology Societies, 57 trials have been retrieved; 3 of these were discarded because they did not provide enough pertinent data. The major end-points considered were the efficacy of treatment (per cent control of active bleeding) and overall mortality. Results - As far as control of bleeding is concerned, vaso-active drugs were significantly better than placebo, H₂-blockers or no treatment (65% vs 47% success; p <0.02). Sclerotherapy was significantly better than conventional treatment (vaso-active drugs ± esophageal balloon tamponade; 87% vs 76% success; p <0.001). Sclerotherapy and variceal band ligation were equally effective in controlling bleeding. All other treatment comparisons showed no significant differences in efficacy between the treatment modalities under trial. As far as short-term mortality is concerned, sclerotherapy was the only form of treatment that showed a statistically significant benefit when compared with conventional treatment (overall mortality: conventional treatment = 32%; emergency sclerotherapy = 26%; p <0.02). Conclusions - a) Treatment with vaso-active drugs is significantly better than conservative measures in controlling bleeding. The main advantages of these drugs are their availability, safety and easy applicability. b) Sclerotherapy is more effective than conventional treatment in controlling bleeding, and is the only form of treatment that improves short-term survival significantly in comparison with conventional therapy. However, the latter conclusion should be accepted with caution, since the methodological quality of some of the sclerotherapy trials included in the meta-analysis is far from ideal.