Permethrin is widely used as a first-line treatment for scabies. There is good evidence from one systematic review and a subsequent randomized controlled trial (RCT) that permethrin is highly effective for treating scabies. The evidence for malathion is less convincing.

• A Cochrane review (search date: March 2004) identified 44 RCTs that compared the use of an insecticide with placebo or another treatment. Only 13 trials (n = 1527) met the inclusion criteria and were included in the analysis [Walker and Johnstone, 2000].
• The studies were heterogeneous:
• A variety of patient groups were studied, including children and adults. Outcome measures were clinical cure, parasitic cure, cure of pruritus, and adverse effects. It was only possible to extract outcome data at 28–30 days for seven of the studies (evidence of cure requires follow-up for about 1 month, as this is the time it takes for lesions to heal and for any eggs and mites to reach maturity if treatment fails).
• Eight trials included all randomized patients (i.e. intention to treat analysis); the remaining five trials reported exclusions in their analyses. Seven of the 13 trials reported blinded assessments of outcome, and four of these were double-blind.
• Several insecticides were used as the intervention, including benzyl benzoate, crotamiton, lindane, ivermectin, permethrin, and sulphur.
• No RCTS including malathion were identified.
• Twelve of the 13 studies were conducted in developing countries.
• The main results were as follows:
• Permethrin versus crotamiton:
• Two trials (n = 194) compared these two topical treatments. Permethrin 5% was significantly more effective at achieving clinical cure in both studies (91/97 vs. 72/97; likelihood of clinical failure: OR 0.21, 95% CI 0.10 to 0.47). Parasitic cure was measured in one of these studies, and permethrin was found to be more effective (42/47 vs. 28/47; likelihood of parasitic failure: OR 0.21, 95% CI 0.08 to 0.53).
• Permethrin versus lindane:
• Four trials (n = 718) compared these two topical treatments. Overall, permethrin 5% was more effective than lindane 1% in achieving clinical cure (60/360 vs. 83/358; likelihood of clinical failure: OR 0.66, 95% CI 0.46 to 0.95). However, two trials (including the largest trial of 467 patients) found no difference between the two treatments.
• Crotamiton versus lindane:
• One trial (n = 100) compared these two treatments. No statistically significant difference in clinical cure rate was found (OR 0.41, 95% CI 0.15 to 1.10).
• Benzyl benzoate versus sulphur:
• One trial (n = 158) compared these two treatments. No statistically significant difference in clinical cure rate was found (OR 2.78, 95% CI 0.77 to 10.05).
• Ivermectin versus placebo:
• Only one trial (n = 55) compared oral ivermectin with placebo. The effectiveness of oral ivermectin over placebo was highly significant at day 7 (clinical cure 23/29 vs. 4/26 OR 0.08, 95% CI 0.03 to 0.23). The blinding was broken for those that had not improved at this time and control patients were given ivermectin.
• Ivermectin versus benzyl benzoate or lindane:
• Two trials (n = 97) compared oral ivermectin with topical benzyl benzoate or lindane. No differences in effectiveness were detected between the treatments in either trial, but the number of participants in the studies was small. No difference in effectiveness was found when these two trials were combined, i.e. oral ivermectin compared with a topical treatment (either benzyl benzoate or lindane; OR 0.58, 95% CI 0.25 to 1.34).
• Adverse effects: no serious adverse drug reactions were reported in any of the included or excluded trials. However, serious adverse drug reactions (including death and convulsions), most notably to lindane, permethrin, and ivermectin, have been reported elsewhere.
• The authors concluded that the evidence that permethrin is more effective than lindane is inconsistent. Lindane, permethrin, and ivermectin appear to be associated with rare but serious adverse drug reactions although this is not derived from trial data. More research is needed on the safety and effectiveness of ivermectin and malathion compared with permethrin, on community management, and on different regimens and vehicles for topical treatment.
• Subsequent to the Cochrane review:
• One RCT (n = 99) found that permethrin 5% significantly increased clinical cure rates compared with lindane at 14 days (failed clinical cure: 8/52 [15%] with permethrin vs. 24/47 [51%] with lindane p < 0.05) [Zargari et al, 2006].
• Another RCT (n = 85) compared a single application of topical permethrin 5% with a single dose of oral ivermectin. Treatment was repeated at 2 weeks if there was treatment failure. The outcome measured was the complete disappearance of clinical signs and symptoms and no appearance of new lesions 2 months after treatment. All patients were completely cured at 2 months [Usha and Gopalakrishnan Nair, 2000].
• Malathion has only been studied in uncontrolled trials. These have found that a single application of malathion 0.5% left on the skin for 24–48 hours cures 70–80% of people within 2–4 weeks [Hanna et al, 1978; Thianprasit and Schuetzenberger, 1984].