Background Despite conflicting evidence, specific serotherapy is recommended for scorpion envenomation.

Methods A meta-analysis of prospective or observational controlled studies, comparing intravenous scorpion antivenin (SAV) with control, was performed. Binary outcomes are reported as risk difference for clinical improvement and mortality rates. Analysis was performed both for the whole number of included studies and for two subgroups (set up according to the geographic origin of scorpions).

Results Nine studies (four randomised controlled trials (RCTs), five observational) enrolling 687 patients were identified. Six dealt with Old World scorpions and three originated from Arizona. Overall, the rate of clinical improvement was similar in SAV treated and untreated patients (risk difference=0.22, 95% CI −0.35 to 0.79; p=0.45 for effect). Subgroup analysis showed favourable effects of SAV in the Arizona scorpion envenomation (risk difference=0.53; 95% CI 0.16 to 0.91; p<0.001), and non-significant unfavourable effects in Old World scorpion envenomation (risk difference=−0.05; 95% CI −0.28 to 0.18; p=0.65; p=0.003 for z-value, indicating a true heterogeneity of treatment effects). In Old World scorpion envenomation, there was no statistical difference in the risk of death in SAV treated and untreated scorpion envenomated patients (risk difference=0.007, 95% CI −0.02 to 0.03; p=0.6 for effect). Overall, administration of scorpion antivenin was associated with a reduction by 13 h in the mean time of symptom resolution (95% CI −17 to −9; p<0.0001). Serious adverse events were reported at a rate of 1–2% while minor adverse events occurred in up to 40% of patients.

Conclusions SAV should not be administered in Old World scorpion envenomation until its efficacy is established by an appropriately designed RCT. In the Arizona scorpion sting, SAV hastens the recovery process.