Objective: Serious morbidity and mortality following snakebite injuries are common in tropical regions of the world. Although antivenom administration is clinically effective, it carries an important risk of early anaphylactic reactions, ranging from relatively benign nausea, vomiting, and urticaria to life-threatening angioedema, bronchospasm and hypotension. Currently, no adequately powered study has demonstrated significant benefit from the use of any prophylactic drug. A high rate of anaphylactic reactions observed during a trial of three different antivenoms in Ecuador prompted adoption of premedication with intravenous (IV) hydrocortisone and diphenhydramine together with dilution and slower administration of antivenom. Design: In a rural mission hospital in Eastern Ecuador, 53 consecutive snakebite victims received a new antivenom regimen in 2004-2006, comprising prophylactic drugs and IV infusion of diluted antivenom over 60 min. They were compared to an historical control cohort of 76 patients treated in 1997-2002 without prophylactic drugs and with IV "push" injection of undiluted antivenom over 10 min. All these patients had incoagulable blood on admission and all were treated with Brazilian Instituto Butantan polyspecific antivenom. Results: Baseline characteristics of the historical control and premedicated groups were broadly similar. In the historical group, early reaction rates were as follows: 51% of patients had no reaction; 35% had mild reactions; 6% moderate; and 6% severe. In the premedicated/slow IV group, 98% of patients had no reaction; 0 mild; 0 moderate; and 2% severe. The difference in reaction rates was statistically significant (p < 0.001). Conclusions: Premedication with intravenous hydrocortisone and diphenhydramine together with dilution of antivenom and its administration by IV infusion over 60 min appeared to reduce both the frequency and severity of anaphylactic reactions. A randomized blinded controlled trial is needed to confirm these encouraging preliminary findings. © 2009 Elsevier Ltd. All rights reserved.