Twelve studies were included on antisepsis for blood culture collection (two systematic reviews and 10 primary studies, of which four were RCTs with 13,142 participants). Twenty studies were included on vascular catheter insertion (two systematic reviews and 18 primary studies, of which 15 were RCTs with over 5,700 participants). Fourteen studies were included on antisepsis before surgery (three systematic reviews and 11 primary studies, of which nine were RCTs with 3,832 participants). Sample size in RCT comparison groups ranged from 24 to 7,606.
Most studies were considered to be at low or moderate risk of bias.
The combination of chlorhexidine gluconate with alcohol was significantly better than povidone iodine alone in preventing blood culture contamination (RR 0.45, 95% CI 0.32 to 0.63, 2 RCTs). There was no significant difference between the combination of chlorhexidine gluconate with alcohol and sequential isopropanol and iodine tincture (2 RCTs). There was no significant heterogeneity. No evidence was identified on the use of chlorhexidine gluconate alone prior to blood culture collection.
Aqueous chlorhexidine gluconate was significantly better than aqueous povidone iodine at preventing catheter colonisation (RR 0.41, 95% CI 0.18 to 0.95, 3 RCTs), but there was no significant difference in blood stream infection (3 RCTs). Chlorhexidine gluconate plus alcohol was significantly better than aqueous povidone iodine at preventing catheter colonisation (RR 0.62, 95% CI 0.39 to 0.98, p=0.04, 8 RCTs) and at preventing blood stream infection (RR 0.44, 95% CI 0.26 to 0.73, 7 RCTs). Significant heterogeneity was detected for the outcome of catheter colonisation. In three other trials of chlorhexidine-alcohol combinations with various comparators, two trials showed a significant advantage for the chlorhexidine preparation in preventing catheter colonisation, but there was no significant effect on blood stream infection.
Chlorhexidine gluconate plus alcohol was significantly better than aqueous povidone iodine at preventing surgical site infection (RR 0.65, 95% CI 0.50 to 0.85, p=0.002, 4 RCTs, plus a fifth trial with no infection detected in either of the comparison groups). Four further trials comparing chlorhexidine-alcohol combinations with iodine-alcohol combinations were very heterogeneous and only one of the trials found significantly reduced surgical site infection in the chlorhexidine-alcohol group.