Background: Cord cleansing with chlorhexidine has been shown to reduce the risk of neonatal mortality in low-resource settings. The effect of whole body skin cleansing with chlorhexidine in neonates is, however, not clear.
Methods: We searched MEDLINE and other databases and included all studies that evaluated the effect of chlorhexidine cleansing on neonatal mortality rate (NMR) and/or the incidence of neonatal sepsis. We estimated the pooled relative risks by fixed effect and random-effects models and also explored the sources of heterogeneity by subgroup meta-analysis.
Results: Seven randomized trials and 2 before-and-after studies that evaluated single cleansing with chlorhexidine were included in the review. Pooled analysis showed no significant effect on NMR in either the fixed effect (5 studies; relative risk: 0.91; 95% confidence interval: 0.80 to 1.04) or random-effects model (0.83; 0.63 to 1.08). On subgroup analysis, the before-and-after study conducted in high-NMR setting showed significant reduction in mortality (0.68; 0.50 to 0.93) but the randomized controlled trials from low-NMR settings did not show any benefit (0.97; 0.84 to 1.13). Pooled analysis of 5 studies that reported the rates of sepsis revealed substantial heterogeneity (I2 = 80.2%). Pooled result by random-effects model as well as the sensitivity analysis including only the randomized trials found no significant effect (0.65; 0.40 to 1.05 and 0.97; 0.80 to 1.18, respectively).
Conclusions: There is no conclusive evidence for any beneficial effect after single skin cleansing with chlorhexidine. However, given the heterogeneity in the major outcomes it seems prudent to generate more evidence from randomized trials in high-NMR settings. Until such time, this intervention cannot be recommended in any settings.