Application of Antiseptic to Umbilical Stump Reduces Risk of Neonatal Infection and Mortality

First published online:

In resource-poor rural areas where women often give birth at home, application of the topical antiseptic chlorhexidine to an infant’s umbilical cord stump reduces the likelihood of infection and neonatal death, two recent clinical trials indicate.1,2 In a study conducted in Pakistan, the risks of cord stump infection (omphalitis) and neonatal death were lower among infants whose cord had been cleansed with chlorhexidine for up to 14 days than among infants not receiving such care (risk ratios, 0.4–0.6); in contrast, a parallel intervention that promoted hand washing among family members provided little or no benefit. The second study, done in Bangladesh, found a reduced risk of severe cord infection among infants who received daily chlorhexidine cleansings during their first week (0.4–0.6), but not among those who received a single cleansing after birth. However, although mortality was reduced in the single-cleansing group, it was not reduced in the multiple-cleansing group.

Worldwide, infections cause a third of neonatal deaths, in part because the umbilical stump is an entry point for pathogens. In some cultures, the problem is exacerbated by traditional practices of applying ash, oil or other substances to the cord. To minimize infection risk, the World Health Organization recommends that the cord stump be kept clean and dry, an approach known as dry cord care. However, a 2006 study in rural Nepal found that applying chlorhexidine solution to the cord soon after delivery can reduce rates of infection and neonatal death. To further examine this approach in resource-poor settings, two teams conducted separate randomized trials in Asia.

In the first study, researchers divided Pakistan’s rural Dadu district into 187 clusters (typically consisting of one or two villages) and randomly assigned to each cluster one of four cord care interventions. In all four, traditional birth attendants provided a birth kit to all pregnant women. In one intervention group, the kit included a supply of chlorhexidine solution and a bar of soap; after a woman gave birth, the attendant showed the family how to cleanse the infant’s cord stump with the antiseptic and instructed them to do the same once a day for 14 days. Family members were also advised to wash their hands with soap before handling the infant. In the second intervention group, the kits contained soap but no chlorhexidine; in the third, chlorhexidine but no soap. The final group served as controls; families were advised to practice dry cord care but were not taught to cleanse the cord or wash their hands.

Community health workers supervised the birth attendants and assessed infants’ health on days 1, 3, 5, 7, 14 and 28. Omphalitis was diagnosed if redness, swelling or pus was present on the cord or the surrounding skin; it was classified as mild if restricted to the stump, moderate if it extended no more than 2 cm from the base of the stump and severe if more extensive. Cases of neonatal mortality (death during first 28 days) were also noted. All birth attendants and community health workers received 3–5 days of training on relevant care practices and assessments.

Between January 2008 and June 2009, a total of 11,886 live births occurred in the study area. After exclusion of infants who were born in facilities, had obvious birth or cord abnormalities, or were delivered by providers not trained for the study, the final sample consisted of 9,741 infants. Mothers had a mean age of 30; only one in 10 was literate.

On average, infants in the two chlorhexidine groups received 2.4 applications per day for 11 days. Compared with infants in the control group, those whose cord had been cleansed were less likely to develop omphalitis, whether they were in the chlorhexidine-alone group (risk ratio, 0.4) or the chlorhexidine plus hand-washing group (0.5). No reduction in neonatal deaths was apparent, however. In another analysis, the researchers compared the two chlorhexidine groups with the other two groups; in this case, chlorhexidine cleansing was associated not only with a reduced risk of infection (0.6), but also with lower mortality (0.6).

Infants in the hand washing–alone group were less likely than those in the control group to develop omphalitis (relative risk, 0.7), though mortality rates did not differ. Moreover, neither infection risk nor mortality risk was reduced in analyses that compared the two hand-washing groups with the two other groups.

The researchers note that the study’s limitations include the use of community health workers, rather than physicians, to identify omphalitis (although this compromise reflects reality in most developing countries), and the inability of health workers to attend most births (and hence to supervise birth attendants’ cord cleansings and instructions to family members). They conclude that application of chlorhexidine can "reduce the incidence of neonatal omphalitis and neonatal mortality," but that promotion of hand washing alone, especially where water is scarce, "might not be enough" to prevent infection.

One issue not addressed by the study is whether a single application of chlorhexidine is sufficient to reduce infection risk. This question was examined in the second trial, conducted in 2007–2009 in rural areas of Bangladesh’s Sylhet district. The trial compared the effectiveness of three approaches: dry cord care, a single chlorhexidine cleansing as soon as possible after birth and daily cleansing during the first week. Researchers divided the study area into 133 clusters of 2,100–5,600 residents and randomly assigned to the clusters one of the three treatments. In each cluster, a community health worker and four village health workers provided birth kits and neonatal care counseling to all pregnant women and asked the family to notify them at the onset of labor. Village health workers visited the mother and infant daily during the week after birth, and the community health worker made visits on days 1, 3, 6, 9, and 15 to assess the infant’s health. Visits were identical in all three intervention groups, except that a village health worker applied chlorhexidine to the cords of infants in the two cleansing groups according to the relevant schedule. The community health worker made a final visit between days 28 and 35 to ascertain the infant’s survival; death within 28 days was the study’s primary outcome. Secondary outcomes were the level of redness at the cord stump, classified in a similar fashion to the Pakistan trial, and the presence of pus.

The analytic sample consisted of 28,308 live-born infants who received at least one visit during the first seven days and whose vital status at day 28 was known. Fewer than half of the mothers received any formal antenatal care. Only 9% of infants were delivered by a skilled birth attendant, though in 93% of cases a sterile tool was used to cut the cord. Most infants (84–85%) in the three groups received all seven daily visits during their first week; an additional 7–8% received five or six visits.

During follow-up, the incidence of moderate redness, or of pus without redness, did not differ among groups. However, compared with infants who received dry cord care, those who received multiple cleansings had a lower risk of developing severe redness or redness with pus (relative risk, 0.6), or of having both severe redness and pus (0.4). No reduction in infection risk occurred among infants who received just a single cleansing.

Mortality data yielded a different pattern: The risk of death among infants who received multiple cleansings did not differ from that of infants in the dry cord group, whereas infants who received a single cleansing had a reduced risk of mortality (relative risk, 0.8).

The researchers suggest that the lack of a reduction in mortality risk in the multiple-cleansing group may have been due to chance and insufficient statistical power. Despite the mixed results, they conclude that chlorhexidine cleansing "can save lives" and, given its low cost, is "especially attractive for countries with restricted resources and high neonatal mortality." They recommend that future studies seek to "confirm that [a] single cleansing is effective, because this intervention would simplify programme logistics and aid the achievement of high intervention coverage in real-life settings."—P. Doskoch


1. Arifeen SE et al., The effect of cord cleansing with chlorhexidine on neonatal mortality in rural Bangladesh: a community-based, cluster randomized trial, Lancet, 2012, 379(9820):1022–1028.

2. Soofi S et al., Topical application of chlorhexidine to neonatal umbilical cords for prevention of omphalitis and neonatal mortality, Lancet, 2012, 379(9820):1029–1036.