Description of the condition
The umbilical cord is a structure made of blood vessels and connective tissue that connects the baby and placenta in utero. Its outer surface is a membrane that is bathed in amniotic fluid. The umbilical cord is cut after birth, which separates the mother and her baby, both physically and symbolically. The cord stump then dries, falls off and the wound heals. The natural process of the umbilical cord falling off involves the formation of an area of separation between the drying cord and the abdominal wall in which polymorphonuclear leucocytes (a type of white blood cells) are present (Oudesluys-Murphy 1990). During this process, material may collect at this junction that sometimes looks like pus and is often wrongly identified as an infection. The cord usually separates between five and 15 days after birth (Oudesluys-Murphy 1987). Before the separation, the remaining stump can be considered to be a healing wound and thus a possible route for infection through the vessels into the baby’s blood stream.
Infection of an umbilical cord may be clinically obvious, but is also sometimes not apparent. In case of frank infections the cord may be swollen, the surrounding skin inflamed, or the cord may be ’smelly’ if infected with anaerobic bacteria (Mullany 2006a). Tracking of bacteria along the umbilical vessels is not obvious to the eye, but can cause septicaemia, or result in other focal infections as a result of blood-borne spread such as septic arthritis (Forshall 1957). In such cases, affected babies may also present with fever, lethargy or poor feeding, collectively called sepsis, in the neonatal period.
Description of the intervention
As described above, the umbilical cord stump can be the potential source of entry of pathogenic microorganisms causing morbidity and mortality. There are two important considerations in this regard; first, is the colonization of newborn skin and umbilical cord stump with potential pathogenic microorganisms and second is the application of harmful substances to the umbilical cord. It is well known that the skin of the newborn, including the umbilical stump, is colonized by microorganisms soon after birth (Mir 2011; Mullany 2012). These microorganisms include both pathogenic and non-pathogenic species. The profile of organisms colonizing the cord stump varies according to hygenic conditions at the time of birth and immediate postpartum period. In high-resource settings, the likely organisms are gram positive ones while in low resource, community settings, gram negative organisms seem more prevalent. A study from US showed that 210 of 211 (99.7%) infants studied were found positive for Staphylococcus aureus at least once in the first six days of life (Fairchild 1958). A recent study from community settings from Bangladesh showed that predominant flora that colonize the newborn umbilical stump were gram negative (Escherichia coli, Klebsiella pneumoniae, and Pseudomonas spp) (Mullany 2012). Pathogenic bacteria such as Escherichia coli, Klebsiella pneumoniae, and Pseudomonas spp and streptococci can track up the umbilical stump causing infection. It is therefore essential to keep the cord clean.
The practice of cord cutting at birth and care of the umbilical stump afterwards varies according to local practice and culture (Elhassani 1984; Mullany 2006; Mullany 2007). In many parts of the world, deliveries occur at home and the cord is cut with unsterile tools such as used razors or scissors after which various substances are applied including mustard oil, turmeric, charcoal, grease, cow dung or dried banana to speed up cord separation (Mir 2011; Mullany 2007; Mullany 2009; Smith 2009). This combination of unhygienic cutting of cord and application of potential harmful substances is an important sources of bacterial infection and neonatal tetanus (Bennett 1997; Mullany 2007; Mullany 2009). Up to this point, there is a general agreement about the ’clean’ technique for cutting the cord using a sterile cutting instrument (blade or scissors) and clean hands to avoid infection (Blencowe 2011), however, there is less agreement on what is the best care of the cord stump (Blencowe 2011; Zupan 2004).
The most frequent modern practice of umbilical cord care is applying antimicrobials to the cord stump. These include antiseptics (such as alcohol, silver sulphadiazine, iodine, chlorhexidine (CHX); and dyes such as triple dye, gentian violet, acriflavine and eosin) and/or topical application of antibiotics (for example, bacitracin, neomycin, nitrofurazone, or tetracycline, or moisture absorbing powders). These substances may be used as solutions in water, alcohol, detergent or ointments. Another approach is to keep the cord clean and dry without applying anything and this is recommended by the World Health Oraganization (WHO 1999). A previous approach was to bath the baby soon after birth with an antimicrobial solution such as hexachlorophene, however, hexachlorophene is no longer recommended in newborn babies as it is absorbed through the skin and is neurotoxic (WHO 1999). One potential side effect of topical antimicrobials is the delay in cord separation time, which can potentially increase the risk of bacterial entry (Novack 1988). However, it has been shown that it does not increase the risk of mortality (Mullany 2006b).
How the intervention might work
Studies from developed countries have shown that use of an antiseptic on umbilical cord stump in hospital nurseries significantly reduced umbilical colonization rates (Barrett 1979; Pezzati 2002; Speck 1977). Seventy per cent alcohol has been used since the 1900s for routine cord care along with antimicrobial solutions such as triple dye, tincture of iodine, iodophors, antibiotic ointments, silver sulphadiazine and CHX. Soaking the umbilical stump in 70% alcohol rapidly kills gram positive and gram negative bacteria. However, it has been shown in several hospital studies to be less effective in controlling umbilical colonization than other antimicrobials such as triple dye or CHX (Panyavudhikrai 2002; Pezzati 2002). Despite its limitations, its low cost and availability have led to widespread use, especially in low-resource settings (WHO 1999).
Recently, CHX has been studied for cord care in community settings (Arifeen 2012; Mullany 2006). CHX is a broad spectrum antiseptic that is extensively used in dental, obstetric and surgical scrub. It has also been used in obstetrics, peripartum, perineal and vaginal washes in concentrations as high as 4% (McClure 2007). Safety studies in newborn infants exposed to CHX washes in various concentrations found no evidence of toxicity even in babies in which percutaneous absorption may have taken place (Aggett 1981; Johnsson 1987). CHX is currently included in WHO’s Essential Drugs List (WHO 2011).
Why it is important to do this review
According to a recent estimate about 40.3% (3.1 million) of all deaths in children less than five years occurred in the neonatal period (Liu 2012). Most of these deaths occurred in developing countries and infections, along with complications of prematurity are the most important cause of mortality in the neonatal period (Liu 2012). In populations with high neonatal mortality rates, infections account for approximately half of all newborn deaths (Lawn 2005). Infection of the cord stump, called omphalitis, is a significant cause of mortality and morbidity in developing countries (Agrawal 2012; Lehmann 1999; Mir 2011; Mullany 2007; Mullany 2009; Sawardekar 2004; Thaver 2009). The infection typically presents as a superficial cellulitis that may progress to involve the abdominal wall and eventually to necrotizing fasciitis, myonecrosis, or systemic disease (Gallagher 2010). The risk is greatest in situations where deliveries take place at home, often with unskilled traditional birth attendants who do not employ clean delivery practices (Darmstadt 2009; Mullany 2009). Omphalitis is relatively rare in developed countries with an overall incidence rate which varies from 0.2% to 0.7% (McKenna 1977). Incidence of omphalitis in developing countries in community settings may range up to 21% (Mir 2011). In these settings, the mortality rate among all infants with omphalitis, including those who develop complications, is estimated at up to 46% (Mullany 2009). The mortality rate is significantly higher (about 71%) after the development of necrotizing fasciitis or myonecrosis (Sawin 1994). Suggested risk factors for poor prognosis include male sex, prematurity or being small-for-gestational age, and septic delivery (Faridi 1993; Gallagher 2010; Mullany 2007).
The WHO and American Academy of Pediatrics recommend good hygiene at delivery, and promote dry cord care practice after birth (AAP 2003; WHO 1999). These recommendations however, are based on insufficient evidence in favour of or against an antiseptic (McClure 2007; Zupan 2004). The aim of this review is to provide data useful for identifying good practice in both high- and low-income countries.
To determine the effect of application of antimicrobials on the umbilical cord of newborns versus routine care for prevention of morbidity and mortality in hospital and community settings.
Criteria for considering studies for this review
Types of studies
Randomized, cluster-randomized and quasi-randomized controlled trials.
Types of participants
Live newborns born to mothers with or without risk factors for the development of infection (for example, chorioamnionitis, preterm rupture of membranes, urinary tract infection), regardless of place of delivery (home, hospital, non-institutional birth, etc) and gestational age and birthweight.
Types of interventions
We evaluated the following interventions.
- Antiseptic versus no antiseptic or placebo/dry cord care.
- Antibiotics versus no antibiotic.
- Antiseptic versus antibiotic.
- Antiseptic versus antiseptic.
- Single versus multiple application.
- Washing umbilical cord with soap/water versus dry cord care.
Those studies were excluded where a combination of antiseptic and antibiotic was used. Studies that evaluated hexachlorophene were excluded as the antiseptic had been removed from the market because of central nervous toxicity.
Types of outcome measures
- All-cause mortality
- Confirmed or suspected sepsis
Confirmed sepsis is defined as clinical signs and symptoms consistent with infection and microbiologically proven with a positive blood culture, cerebrospinal fluid culture, urine culture or culture from a normally sterile site (e.g. pleural fluid, peritoneal fluid or autopsy specimens) for bacteria or fungi.
Suspected sepsis is defined as clinical signs and symptoms consistent with sepsis without isolation of a causative organism.
Tetanus is defined as trismus (spasm of the muscles involved in opening of the jaws) and severe generalized muscular spasms not attributable to other causes (i.e. hypocalcaemia, phenothiazine reaction, strychnine poisoning) (AAP 2003).
Omphalitis is defined as clinical signs and symptoms of umbilical stump infection which include the following.
- purulent or malodorous discharge from the umbilical stump;
- periumbilical erythema;
Three case definitions have been described by Mullany 2006a to describe the severity of omphalitis and had been used wherever data were available. The three definitions are described below.
Algorithm 1: Moderate or severe redness.
Algorithm 2: Moderate redness with pus, or severe redness (without regard to pus).
Algorithm 3: Severe redness with pus.
Extensive local disease that includes conditions such as necrotizing fasciitis or myonecrosis, which are typically found in a periumbilical location but may spread across the abdominal wall, onto the flanks and back, and into the scrotum. These signs may also suggest infection by both aerobic and anaerobic organisms and include the following:
- ecchymoses, violaceous discolorations;
- peau d'orange appearance (the skin looks like orange peel);
- progression of cellulitis despite antimicrobial therapy (Gallagher 2010).
- Bacterial colonization
- Time to cord separation
Search methods for identification of studies
We contacted the Trials Search Co-ordinator to search the Cochrane Pregnancy and Childbirth Group’s Trials Register (1 October 2012).
The Cochrane Pregnancy and Childbirth Group’s Trials Register is maintained by the Trials Search Co-ordinator and contains trials identified from:
- monthly searches of the Cochrane Central Register of Controlled Trials (CENTRAL);
- weekly searches of MEDLINE;
- weekly searches of EMBASE;
- handsearches of 30 journals and the proceedings of major conferences;
- weekly current awareness alerts for a further 44 journals plus monthly BioMed Central email alerts.
Details of the search strategies for CENTRAL, MEDLINE and EMBASE, the list of handsearched journals and conference proceedings, and the list of journals reviewed via the current awareness service can be found in the ‘Specialized Register’ section within the editorial information about the Cochrane Pregnancy and Childbirth Group.
Trials identified through the searching activities described above are each assigned to a review topic (or topics). The Trials Search Co-ordinator searches the register for each review using the topic list rather than keywords.
We did not apply any language restrictions.
Data collection and analysis
Selection of studies
Two review authors independently assessed for inclusion potential studies identified as a result of the search strategy. We resolved any disagreement through discussion or, if required, we consulted a third person.
Data extraction and management
For eligible studies, at least two review authors extracted the data. We resolved discrepancies through discussion or, if required, we consulted a third person. We entered data into Review Manager software (RevMan 2011) and checked for accuracy.
When information regarding any of the above was unclear, we attempted to contact authors of the original reports to provide further details.
Assessment of risk of bias in included studies
Two review authors independently assessed risk of bias for each study using the criteria outlined in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). We resolved any disagreement by discussion or by involving a third assessor.
(1) Random sequence generation (checking for possible selection bias)
We described for each included study the method used to generate the allocation sequence in sufficient detail to allow an assessment of whether it should produce comparable groups.
We assessed the method as:
- low risk of bias (any truly random process, e.g. random number table; computer random number generator);
- high risk of bias (any non-random process, e.g. odd or even date of birth; hospital or clinic record number);
- unclear risk of bias.
(2) Allocation concealment (checking for possible selection bias)
We described for each included study the method used to conceal the allocation sequence and determine whether intervention allocation could have been foreseen in advance of, or during recruitment, or changed after assignment.
We assessed the methods as:
- low risk of bias (e.g. telephone or central randomization; consecutively numbered sealed opaque envelopes);
- high risk of bias (open random allocation; unsealed or non-opaque envelopes, alternation; date of birth);
- unclear risk of bias.
(3) Blinding (checking for possible performance bias)
We described for each included study the methods used, if any, to blind study participants and personnel from knowledge of which intervention a participant received. We considered that studies were at low risk of bias if they were blinded, or if we judged that the lack of blinding could not have affected the results. We assessed blinding separately for different outcomes or classes of outcomes.
We assessed the methods as:
- low, high or unclear risk of bias for participants;
- low, high or unclear risk of bias for personnel;
- low, high or unclear risk of bias for outcome assessors.
(4) Incomplete outcome data (checking for possible attrition bias through withdrawals, dropouts, protocol deviations)
We described for each included study, and for each outcome or class of outcomes, the completeness of data including attrition and exclusions from the analysis. We stated whether attrition and exclusions were reported, the numbers included in the analysis at each stage (compared with the total randomized participants), reasons for attrition or exclusion where reported, and whether missing data were balanced across groups or were related to outcomes. Where sufficient information was reported, or could be supplied by the trial authors, we included the missing data in the analyses.
We assessed methods as:
- low risk of bias (e.g. less than 10% missing data);
- high risk of bias (e.g. numbers or reasons for missing data imbalanced across groups; ‘as treated’ analysis done with substantial departure of intervention received from that assigned at randomization);
- unclear risk of bias.
If missing data could not be supplied by the trial authors, we carried out subgroup analysis for groups with sufficient outcome data.
(5) Selective reporting (checking for reporting bias)
We described for each included study how we investigated the possibility of selective outcome reporting bias and what we found. We assessed the methods as:
- low risk of bias (where it was clear that all of the study’s pre-specified outcomes and all expected outcomes of interest to the review had been reported);
- high risk of bias (where not all the study’s pre-specified outcomes had been reported; one or more reported primary outcomes were not pre-specified; outcomes of interest were reported incompletely and so could not be used; study failed to include results of a key outcome that would have been expected to have been reported);
- unclear risk of bias.
(6) Other sources of bias
We described for each included study any important concerns we had about other possible sources of bias.
We assessed whether each study was free of other problems that could put it at risk of bias (e.g. study design, imbalance in baseline data):
- low risk of other bias;
- high risk of other bias;
- unclear whether there is risk of other bias.
(7) Overall risk of bias
We made explicit judgements about whether studies were at high risk of bias, according to the criteria given in the Cochrane Handbook (Higgins 2011). With reference to (1) to (6) above, we assessed the likely magnitude and direction of the bias and whether we considered it likely to impact on the findings. We explored the impact of the level of bias through sensitivity analyses - see Sensitivity analysis.
Measures of treatment effect
For dichotomous data, we presented results as summary risk ratio with 95% confidence intervals. For the three included cluster-randomized trials, data were entered using the generic inverse variance option in RevMan 2011 to allow adjustment of the variance for cluster effect (Higgins 2011).
For continuous data, we used the mean difference with 95% confidence intervals. For studies with multiple groups, comparison groups were combined into a single pair-wise comparison. Standard deviations were adjusted using the formula in the Cochrane Handbook (Higgins 2011).
Unit of analysis issues
We included cluster-randomized trials in the analyses along with individual-randomized trials. We used cluster-adjusted values from the trials, irrespective of the method used. In case a trial was not adjusted for cluster design, results were adjusting by inflating the standard error of the effect size by quare root of design effect given in the study.
Dealing with missing data
For included studies, we took note of levels of attrition. We explored the impact of including studies with high levels of missing data in the overall assessment of treatment effect by using sensitivity analysis.
For all outcomes, we carried out analyses, as far as possible, on an intention-to-treat basis, i.e. we included all participants randomized to each group in the analyses, and analyzed all participants in the group to which they were allocated, regardless of whether or not they received the allocated intervention.
Assessment of heterogeneity
We assessed statistical heterogeneity in each meta-analysis using the T², I² and Chi² statistics. We considered heterogeneity as substantial if an I² was greater than 50% and either the T² was greater than zero, or the P value was less than 0.10 in the Chi² test for heterogeneity.
Assessment of reporting biases
We planned that, if there were 10 or more studies in the meta-analysis, we would investigate reporting biases (such as publication bias) using funnel plots. We intended to assess funnel plot asymmetry visually. If asymmetry was suggested by a visual assessment, we planned to perform exploratory analyses to investigate it.
We performed statistical analysis using the Review Manager software (RevMan 2011). We used the random-effects analysis to produce an overall summary of the average treatment effect across trials considering the different interventions across different economic settings. We have presented the results as the average treatment effect with its 95% confidence interval, and the estimates of T² and I².
Subgroup analysis and investigation of heterogeneity
We planned to the following subgroup analyses:
- preterm (gestational age less than 37 weeks) versus term (gestational age 37 weeks or more);
- hospital setting; community-based studies; settings mixed or undefined;
- trials carried out in Europe-Americas; Western Pacific; Eastern Mediterranean; South-East Asia; Africa.
Gestational age and birthweight are relevant subgroups since the incidence of sepsis is high among low birthweight and preterm infants. Causative organisms, corresponding susceptibility to antiseptics and severity of disease, may differ in hospital versus community settings. We therefore analyzed hospital and community-based studies separately. Baseline neonatal mortality rate varies with region: Europe-America (11-12/1000 livebirths); Western Pacific (19/1000 livebirths); Eastern Mediterranean (40/1000 livebirths); Southeast Asia (38/1000 livebirths); Africa (44/1000 livebirths) (WHO 2006).
We included all primary outcomes in the subgroup analysis (mortality, confirmed or suspected sepsis, tetanus and omphalitis).
We carried out sensitivity analysis removing studies of low quality.
We also carried out sensitivity analyses to investigate the effect of missing data:
- less than 5% missing data;
- 5% to 10% missing data;
- 10% to 20% missing data;
- 20% or more missing data.
Description of studies
Results of the search
The search identified 77 trials; we included 34 studies in the review, involving 69,338 babies (See Characteristics of included studies). Thirty-six studies were excluded (see Characteristics of excluded studies). Five studies are awaiting classification (see Characteristics of studies awaiting classification) while there are two ongoing community trials (see Ongoing studies).
Thirty-four trials reported in 46 papers met the inclusion criteria. More than one report was available for six (17%) trials (Arifeen 2012; Evens 2004; Mullany 2006; Speck 1977; Soofi 2012; Suliman 2010). When results of an included trial were reported in more than one publication, we extracted data from all reports but counted it as one trial. All included trials reported data that could be included in a meta-analysis.
There were three large, cluster-randomized trials conducted in community settings (Arifeen 2012; Mullany 2006; Soofi 2012). These three trials contributed 78% of the total number of children included in this review. The two ongoing studies are also community trials that assess effect of topical application of chlorhexidine to umbilical cord. One study is being conducted in Zambia (Hamer 2010) and other in Pemba (Sazawal 2012).
Of the trials conducted in hospital settings, the majority had small sample sizes. Sample size of these 31 (91%) trials ranged between 71 and 2241.
Seventeen (50%) of the included studies had a 'dry cord care' group and 16 (47%) studies comprised more than two study groups.
The studies were conducted in both developing and developed countries. All three community studies were conducted in developing countries and most of the hospital-based studies were conducted in developed countries. There were eight studies conducted in USA (Barrett 1979; Evens 2004; Gladstone 1988; Golombek 2002; Rosenfeld 1989; Schuman 1985; Speck 1977; Suliman 2010), four in Canada (Dore 1998; Janssen 2003; Medves 1997; Rush 1986), three in Taiwan (Hsu 1999; Hsu 2010; Huang 2001), two each in Iran (Ahmadpour-Kacho 2006; Nourian 2009), Norway (Meberg 1985; Meberg 1990), Italy (Pezzati 2002; Pezzati 2003), and Pakistan (Shafique 2006; Soofi 2012) and one each in Saudi Arabia (Al-Binali 2006), Palestine (Arad 1981), Bangladesh (Arifeen 2012), Scotland (Bain 1994), Peru (Davila 2007), Germany (Kapellen 2009), England (Mugford 1986), Nepal (Mullany 2006), Japan (Oishi 2004), Thailand (Panyavudhikrai 2002), and Spain (Perapoch 1993).
There were 18 (52%) studies that included full-term babies, five (15%) included preterm babies (Bain 1994; Gladstone 1988; Pezzati 2002; Pezzati 2003; Rosenfeld 1989), seven (20%) included both term and preterm babies (Ahmadpour-Kacho 2006; Arifeen 2012; Davila 2007; Janssen 2003; Mullany 2006; Oishi 2004; Soofi 2012) and gestational age was not mentioned in four (12%) studies (Al-Binali 2006; Barrett 1979; Mugford 1986; Schuman 1985).
There were 22 different interventions studied across the included trials. These included topical antiseptics, antibiotics and measures such as washing the cord with soap and water. There were 20 studies (58%) that evaluated topical alcohol application, 10 (29%) triple dye, nine (26%) chlorhexidine, five (14%) Silver Sulfadiazine, three (8%) Povidine and two (5%) salicylic powder. There was one study each for Beniktol, neomycin, bismuth, breastmilk, bacitracin, benzine, hydrophobic gauze, mercurochrome, green clay powder, katoxin, fuschine, and citrane.
Of the primary outcomes considered, three trials (Arifeen 2012; Mullany 2006; Soofi 2012) evaluated the effect of cord care on mortality. Thirteen studies reported data on omphalitis. One study reported data on sepsis (Pezzati 2003) and one study reported sepsis-related mortality (Mullany 2006). No study reported tetanus.
Of the secondary outcomes, 26 studies reported cord separation time, 12 reported bacterial colonization.
Thirty-six studies were excluded (See Characteristics of excluded studies). Reasons for exclusion were: 1) not a randomized controlled trial, 2) not a cord care study or, 3) effect of individual intervention was not established due to co-interventions.
Risk of bias in included studies
See Figure 1 for summary of assessment of risk of bias.
|Figure 1. 'Risk of bias' summary: review authors' judgements about each risk of bias item for each included study.|
Twelve (35%) studies had adequate methods of sequence generation, eight (24%) had inadequate methods or were quasi-randomized trials while in 14 (41%) studies, methods were inadequately described to make a complete assessment.
Allocation was adequately concealed in seven (21%) studies while in 22 (65%) studies it was not mentioned or unclear. Five (15%) studies had inadequate methods of allocation concealment.
Method of blinding was not done, not feasible, or not mentioned in 30 (88%) studies. Three studies had "adequate" methods of blinding. The study of Mullany 2006 mentioned that investigators, field workers and participants were masked to chlorhexidine and soap and water but not to the dry cord care.
Incomplete outcome data
Seventeen (50%) studies reported data on attrition and exclusion while 14 studies (41%) did not describe enough details on follow-up. There was insufficient information in three (9%) studies to make a complete assessment.
It is difficult to assess true selective reporting bias when protocols of the included studies are not available. Review authors had access to the protocols of three studies to make an assessment (Arifeen 2012; Mullany 2006; Soofi 2012) and they were considered to be adequate. In the rest of the studies, the judgment was made based on variables mentioned in the methods section compared to those reported in the results section of the study. Based on this assessment, 13 studies had adequate descriptions of outcomes in the results section for the outcomes mentioned in the methods section (Ahmadpour-Kacho 2006; Arifeen 2012; Bain 1994; Barrett 1979; Dore 1998; Hsu 2010; Medves 1997; Mugford 1986; Mullany 2006; Nourian 2009; Rosenfeld 1989; Soofi 2012; Suliman 2010).
Other potential sources of bias
In the study of (Ahmadpour-Kacho 2006), baseline characteristics showed more vaginal deliveries in the dry care, alcohol and silver groups compared with the breastmilk group. Arad 1981 included the outcome of sepsis but no bacterial cultures were obtained to prove the condition. The community trial from Bangladesh (Arifeen 2012), showed that there was no effect of multiple cleansing with chlorhexidine on neonatal mortality. This may be due to the fact that the study was not powered enough to detect a significant difference in this arm, as pointed out by authors (Arifeen 2012). In the study of Bain 1994, there was a higher rate of rupture of membranes in the dry care group. In Davila 2007, there was no table of baseline characteristics. In the study of Evens 2004, there was a higher rate of vaginal delivery in the alcohol group. The studies by Golombek 2002 and Shafique 2006 gathered data on the incidence of omphalitis by telephone calls and no home visits were made to make an assessment. This could have resulted in false positives or negatives as symptoms were reported by mothers. Kapellen 2009 had no table of baseline characteristics but the text mentioned that the groups were comparable in terms of sex, ethnicity, birthweight, birth length, gestational age, model of delivery and Apgar score. No description of baseline characteristics of study participants was given in five studies (Al-Binali 2006; Meberg 1985; Meberg 1990; Perapoch 1993; Speck 1977). The study of Medves 1997 had no table of baseline characteristics. Mugford 1986 mentioned in the text that the groups were comparable in terms of sex, birthweight and mode of delivery. The study of Oishi 2004 states that the babies in the alcohol group had low birthweights but the proportion was not mentioned. The desired sample size in Soofi 2012was not achieved due to a security situation in the study area.
Effects of interventions
Antiseptics versus dry cord care/placebo
There were 18 studies that had a comparison group between 'dry cord care' and 'an antiseptic'. Fifteen of these studies were conducted in hospital settings while three were conducted in community settings (Arifeen 2012; Mullany 2006; Soofi 2012).
The antiseptics included in this comparison were: alcohol, triple dye, chlorhexidine, salicylic sugar powder, green clay powder, silver sulphadiazine, benzine, katoxin powder, fuschine, zinc powder and breastmilk.
Only one antiseptic, i.e. chlorhexidine was studied in community settings for umbilical cord care. Three community trials reported data on all-cause mortality (Arifeen 2012; Mullany 2006; Soofi 2012). There were 1325 deaths in 54,624 participants of three studies and combined results showed a reduction of 23% in the intervention group compared with control (average risk ratio (RR) 0.77, 95% confidence interval (CI) 0.63 to 0.94; random-effects, T² = 0.02, I² = 50% ( Analysis 1.1)).
No community-based study reported data on the incidence of sepsis, however, one trial (Mullany 2006) from Nepal reported data on sepsis-specific mortality and showed a 31% reduction in the chlorhexidine group compared with dry cord care but the results were not statistically significant (RR 0.69, 95% CI 0.40 to 1.18).
Antiseptics were associated with a significant reduction in omphalitis ranging from 27% to 56% depending on the severity of infection: redness extending to skin: average RR 0.73, 95% CI 0.64 to 0.83; three studies, random-effects, T² = 0.00, I² = 34% ( Analysis 1.2); redness with pus or severe redness: average RR 0.69, 95% CI 0.60 to 0.79; three studies, random-effects, T² = 0.00, I² = 0% ( Analysis 1.3); severe redness with pus: average RR 0.44, 95% CI 0.28 to 0.69; three studies, random-effects, T² = 0.03, I² = 19% ( Analysis 1.4), with the most significant reduction in severe cases.
One study (Arifeen 2012) reported data on bacterial colonization and showed a significant reduction in bacterial colonization of Staphylococcus aureus (RR 0.27, 95% CI 0.23 to 0.31 ( Analysis 1.5)); Enterococcus coli (RR 0.50, 95% CI 0.46 to 0.54 ( Analysis 1.6)) and streptococci (RR 0.28, 95% CI 0.22 to 0.37 ( Analysis 1.7)) with topical chlorhexidine application compared with control.
Cord separation time
Data for cord separation time was available from two studies (Arifeen 2012; Mullany 2006), carried out in Bangladesh and Nepal respectively. In the study by Mullany 2006, cord separation time was longer in the chlorhexidine group (5.32 ± 2.4 days) compared with the dry cord care group (4.24 ± 1.6 days). There was no increased risk of mortality in children who had increased cord separation time (Mullany 2006). In Arifeen 2012, separation time in the combined chlorhexidine group (7.20 ± 3.0 days) was 2.40 (95% CI 2.17 to 2.64) days longer than among babies not exposed to chlorhexidine (4.79 ± 1.8 days) (Unpublished data). The combined data for both of these studies showed that cord separation time was 1.7 days longer in the chlorhexidine group compared with the non-chlorhexidine group (mean difference (MD) 1.75, 95% CI 0.44, to 3.05, random-effects, T² = 0.88, I² = 100% ( Analysis 1.8)).
Among studies conducted in hospital settings, no study reported data on mortality, sepsis or tetanus for the comparison of antiseptics versus dry cord care/placebo.
Five studies reported data for incidence of omphalitis (Bain 1994; Janssen 2003; Kapellen 2009; Nourian 2009; Pezzati 2002). There was no significant difference for the incidence of omphalitis when alcohol (three studies) (average RR 0.92, 95% CI 0.62 to 1.39), triple dye (two studies) (average RR 0.71, 95% CI 0.13 to 3.73), chlorhexidine (one study) (RR 0.28, 95% CI 0.06 to 1.35), salicylic sugar powder (one study) (RR 0.21, 95% CI 0.01 to 4.38) and green clay powder (one study) (RR 0.48, 95% CI 0.04 to 5.26) were compared with dry cord care/placebo ( Analysis 2.1).
Compared with dry cord care/placebo, bacterial colonization of umbilical cord with Staphylococcus aureus was significantly reduced with application of triple dye (four studies) (average RR 0.15, 95% CI 0.10 to 0.22, random-effects, T² = 0.04, I² = 24%), silver sulphadiazine (two studies) (average RR 0.72, 95% CI 0.60 to 0.87, random-effects, T² = 0.00, I² = 0%), chlorhexidine (one study) (RR 0.65, 95% CI 0.55 to 0.77), salicylic sugar powder (one study) (RR 0.32, 95% CI 0.17 to 0.58), green clay powder (one study) (RR 0.51, 95% CI 0.31 to 0.82) and fuschine (one study) (RR 0.52, 95% CI 0.32 to 0.84). For the same control group, there was no significant difference for alcohol (two studies) (average RR 0.61, 95% CI 0.11 to 3.36), benzine (one study) (RR 0.99, 95% CI 0.90 to 1.09). Compared with dry cord care/placebo, there was a significantly increased risk of bacterial colonization of umbilical cord with Staphylococcus aureus associated with use of katoxin (one study) (RR 1.43, 95% CI 1.02 to 2.00). See Analysis 2.2.
For the outcome of colonization with streptococci, there was a significant reduction with application of silver sulphadiazine (two studies) (average RR 0.62, 95% CI 0.43 to 0.89, random-effects, T² = 0.00, I² = 0%) and fuschine (one study) (RR 0.19, 95% CI 0.04 to 0.85) compared with dry cord care/placebo. For the same comparison group, there was no difference in colonization of streptococci with the application of alcohol (two studies) (RR 0.42, 95% CI 0.15 to 1.19, random-effects, T² = 0.33, I² = 50%), triple dye (three studies) (average RR 0.57, 95% CI 0.28 to 1.18, random-effects, T² = 0.31, I² = 79%), chlorhexidine (one study) (RR 0.53, 95% CI 0.27 to 1.04), salicylic sugar powder (one study) (RR 0.74, 95% CI 0.29 to 1.90). However, there was an increased risk of streptococcal bacterial colonization associated with the use of green clay powder (one study) (RR 4.62, 95% CI 2.41 to 8.84) and katoxin powder (one study) (RR 5.87, 95% CI 3.12 to 11.05). See Analysis 2.3.
Compared with dry cord care/placebo, there was a significant reduction in the colonization of Enterococcus coli with topical application of alcohol (two studies) (average RR 0.73, 95% CI 0.58 to 0.92, random-effects, T² = 0.00, I² = 0%), silver sulphadiazine (one study) (RR 0.70, 95% CI 0.53 to 0.93) and chlorhexidine (one study) (RR 0.59, 95% CI 0.39 to 0.90). In contrast, there was no significant difference with application of triple dye (two studies) (average RR 0.79, 95% CI 0.53 to 1.17, random-effects, T² = 0.06, I² = 79%), salicylic sugar powder (one study) (RR 0.59, 95% CI 0.32 to 1.10), green clay powder (one study) (RR 1.27, 95% CI 0.79 to 2.05) and katoxin powder (one study) (RR 1.12, 95% CI 0.70 to 1.81). There was increased risk of Enterococcus coli colonization associated with application of fuschine (one study) (RR 2.04, 95% CI 1.33 to 3.13). See Analysis 2.4.
Cord separation time
There was a significant increase in cord separation time when alcohol was applied to umbilical cord compared with dry cord care/placebo (nine studies) (MD 1.76 days, 95% CI 0.03 to 3.48, random-effects, T² = 6.54, I² = 97%). There was one study each for triple dye (MD 4.10 days, 95% CI 3.07 to 5.13), katoxin powder (MD 0.80 days, 95% CI 0.18 to 1.42), fuschine (MD 2.80 days, 95% CI 2.01 to 3.59) and silver sulphadiazine (MD 3.60 days, 95% CI 2.66 to 4.54) and these antiseptics were associated with a significant increase in cord separation time compared with dry cord care/placebo. One study each for zinc powder (MD -1.82 days, 95% CI -2.23 to -1.41), salicylic sugar powder (MD -1.90 days, 95% CI -2.47 to -1.33), green clay powder (MD -0.80 days, 95% CI -1.36 to -0.24), breastmilk (MD -1.69 days, 95% CI -2.31 to -1.07) and chlorhexidine (MD -0.80 days, 95% CI -1.21 to -0.39) were associated with decreased cord separation time compared with dry cord care. See Analysis 2.6.
Antibiotics versus no antibiotic
There were no studies in community or hospital settings that investigated this comparison.
Antiseptics versus antibiotics
There was no study that was conducted in community settings for this comparison.
Among studies conducted in hospital settings, no data were reported on mortality, sepsis, tetanus or omphalitis for this comparison.
One study compared triple dye (RR 0.65, 95% CI 0.03 to 12.87), silver sulphadiazine (RR 2.18, 95% CI 0.42 to 11.33) and povidone (RR 2.18, 95% CI 0.42 to 11.33) with bacitracin and there were no differences in bacterial colonization rates for Staphylococcus aureus ( Analysis 3.1). No data were reported for colonization with streptococcus and enterococcus.
Cord separation time
Cord separation time was significantly reduced when umbilical cord was treated with triple dye compared with bacitracin (one study) (MD -5.60 days, 95% CI -9.36 to -1.84) and neomycin (one study) (MD -4.30, 95% CI -6.27 to -2.33). There was no difference in cord separation time when silver sulphadiazine was compared with neomycin (one study) (MD -1.40 days, 95% CI -3.65 to 0.85). Cord separation time was significantly reduced when umbilical cord was treated with povidone versus bacitracin (one study) (MD -2.00 days, 95% CI -3.67 to -0.33). Similarly, alcohol was associated with a significantly reduced cord separation time compared with beniktol (MD -2.33 days, 95% CI -3.77 to -0.89). See Analysis 3.2.
Antiseptics versus antiseptics
There was no study that was conducted in community settings for this comparison.
Sepsis and mortality
Among studies conducted in hospital settings, no study reported data on mortality or tetanus. One study compared chlorhexidine with salicylic acid powder and reported no difference in the incidence of sepsis between the two groups (RR 1.11, 95% CI 0.07 to 17.50 ( Analysis 4.1)).
Three studies evaluated triple dye versus alcohol and reported no significant reduction in the incidence of omphalitis in the triple dye group compared with the alcohol-treated group (average RR 0.48, 95% CI 0.14 to 1.63, random-effects, T² = 0.54, I² = 41%). There was no difference when triple dye was compared with salicylic sugar powder (one study) (RR 4.29, 95% CI 0.21 to 88.65) and katoxin (one study) (RR 2.13, 95% CI 0.19 to 23.34).
Similarly, there was no difference in the incidence of omphalitis for the comparisons of alcohol versus povidone (RR 0.84, 95% CI 0.62 to 1.16, one study), alcohol versus salicylic powder (RR 4.69, 95% CI 0.23 to 97.03, one study), alcohol versus green clay powder (RR 2.07, 95% CI 0.19 to 22.60, one study), alcohol versus katoxin (RR 2.10, 95% CI 0.19 to 22.97, one study), alcohol versus fuschine (RR 2.10, 95% CI 0.19 to 22.97, one study) and alcohol versus chlorhexidine (RR 2.77, 95% CI 0.12 to 66.49, one study). One study also compared chlorhexidine with hydrophobic gauze and reported no significant difference in the incidence of omphalitis (RR 1.36, 95% CI 0.55 to 3.36).
In contrast, triple dye was associated with a significant reduction in the incidence of omphalitis compared with povidone-iodine (RR 0.15, 95% CI 0.07 to 0.32, one study). See Analysis 4.2.
Bacterial colonization with Staphylococcus aureus was significantly reduced when umbilical cord was treated with triple dye compared with alcohol (two studies) (average RR 0.45, 95% CI 0.25 to 0.80, random-effects, T² = 0.00, I² = 0%), silver sulphadiazine (three studies) (average RR 0.36, 95% CI 0.25 to 0.50, random-effects, T² = 0.00, I² = 0%), green clay powder (one study) (RR 0.27, 95% CI 0.11 to 0.65) and katoxin powder (one study) (RR 0.26, 95% CI 0.11 to 0.63) ( Analysis 4.3). There was no significant difference in Staphylococcus aureus colonization rates for comparisons of triple dye versus povidone (RR 0.65, 95% CI 0.03 to 12.87), triple dye versus salicylic sugar powder (one study) (RR 0.43, 95% CI 0.16 to 1.12) or triple dye versus fuschine (one study) (RR 1.92, 95% CI 0.18 to 20.97). One trial (Pezzati 2002) studied alcohol in comparison with green clay powder (RR 0.59, 95% CI 0.30 to 1.16), fuschine (RR 0.57, 95% CI 0.29 to 1.12), salicylic powder (RR 0.94, 95% CI 0.43 to 2.03) and povidone (RR 0.84, 95% CI 0.62 to 1.16) and found no significant difference in the colonization of umbilical cord with Staphylococcus aureus.
However, treating the umbilical cord with alcohol was associated with a significant reduction in bacterial colonization with Staphylococcus aureus compared with the use of katoxin powder (RR 0.21, 95% CI 0.12 to 0.37, one study). One study each evaluated chlorhexidine in comparison with hydrophobic gauze (RR 0.90, 95% CI 0.70 to 1.15) and mercurochrome (RR 0.11, 95% CI 0.01 to 2.04) and found no significant difference in colonization rates for Staphylococcus aureus. There was however, a significant difference when chlorhexidine was compared with alcohol (two studies) (RR 0.41, 95% CI 0.24 to 0.71). See ( Analysis 4.3).
One study each compared triple dye with alcohol (RR 0.46, 95% CI 0.04 to 4.94), silver sulphadiazine (RR 1.26, 95% CI 0.71 to 2.25), and fuschine (RR 0.48, 95% CI 0.04 to 5.24) and reported no significant difference between the groups for colonization of streptococcus ( Analysis 4.4). However, for the same outcome, there was a significant effect of triple dye compared with green clay powder (one study) (RR 0.02, 95% CI 0.00 to 0.14), salicylic sugar powder (one study) (RR 0.12, 95% CI 0.02 to 0.98) and katoxin powder (one study) (RR 0.02, 95% CI 0.00 to 0.11). Application of alcohol to the umbilical cord was also associated with a reduction in colonization of streptococcus compared with green clay powder (one study) (RR 0.04, 95% CI 0.01 to 0.17) and katoxin powder (one study) (RR 0.03, 95% CI 0.01 to 0.14) but there was no difference when alcohol was compared with fuschine (one study) (RR 1.05, 95% CI 0.15 to 7.38) or salicylic powder (one study) (RR 0.94, 95% CI 0.43 to 2.03). See Analysis 4.4.
Triple dye led to an increased risk of colonization of Enterococcus coli compared with alcohol (one study) (RR 3.44, 95% CI 2.10 to 5.64), silver sulphadiazine (one study) (RR 1.36, 95% CI 1.02 to 1.81), green clay powder (one study) (RR 1.83, 95% CI 1.27 to 2.65) and salicylic sugar powder (one study) (RR 3.92, 95% CI 2.28 to 6.72) and katoxin powder (one study) (RR 2.07, 95% CI 1.43 to 3.00) ( Analysis 4.5). For the same outcomes, alcohol had a significant preventive effect compared with green clay powder (RR 0.53, 95% CI 0.31 to 0.92) and fuschine (RR 0.33, 95% CI 0.20 to 0.55). However, there was no difference in colonization rates of Enterococcus coli for the comparison of alcohol with salicylic powder (one study) (RR 1.14, 95% CI 0.58 to 2.4) and katoxin powder (one study) (RR 0.60, 95% CI 0.35 to 1.04). Similarly, there were no differences in colonization rates when comparing chlorhexidine with hydrophobic gauze (one study) (RR 0.79, 95% CI 0.31 to 2.00) or mercurochrome (one study) (RR 0.21, 95% CI 0.01 to 4.13). See Analysis 4.5.
Cord separation time
Cord separation time was not significantly different when triple dye was compared with alcohol (two studies) (MD 0.43 days, 95% CI -8.49 to 9.35, random-effects, T² = 40.92, I² = 99%) or when triple dye was compared with silver sulphadiazine (two studies) (MD 0.15 days, 95% CI -6.20 to 6.51, random-effects, T² = 18.82, I² = 89%). Cord separation time was increased when triple dye was compared with salicylic sugar powder (one study) (MD 6.00 days, 95% CI 5.01 to 6.99), povidone-iodine (one study) (MD 7.60 days, 95% CI 3.96 to 11.24), green clay powder (one study) (MD 4.90 days, 95% CI 3.92 to 5.88), katoxin powder (one study) (MD 3.30 days, 95% CI 2.28 to 4.32) and fuschine (one study) (MD 1.30 days, 95% CI 0.17 to 2.43) ( Analysis 4.6). There was also a significant increase in cord separation time when the umbilical cord was treated with alcohol compared with green clay powder (one study) (MD 10.20 days, 95% CI 9.05 to 11.35), katoxin powder (one study) (MD 8.60 days, 95% CI 7.42 to 9.78), salicylic powder (one study) (MD 11.30 days, 95% CI 10.14 to 12.46) or fuschine (one study) (MD 6.60 days, 95% CI 5.32 to 7.88) ( Analysis 4.6). Chlorhexidine decreased cord separation time compared with hydrophobic gauze (one study) (MD -0.40 days, 95% CI -0.57 to -0.23), however, chlorhexidine significantly increased cord separation time compared with either salicylic powder (one study) (MD 3.00 days, 95% CI 2.46 to 3.54) or mercurochrome (one study) (MD 6.40 days, 95% CI 5.25 to 7.55) ( Analysis 4.6). One study showed that application of povidone resulted in a shorter cord separation time compared with silver sulphadiazine (MD -4.00 days, 95% CI -5.53 to -2.47). However, there was no significant difference in cord separation time when a combination of triple dye and alcohol was compared with triple dye alone (MD 1.00 day, 95% CI -0.45 to 2.45). See Analysis 4.6.
Single versus multiple application of antiseptic
There was one study from community settings that compared single versus multiple application of chlorhexidine and reported all-cause mortality, incidence of omphalitis and bacterial colonization (Arifeen 2012). No study reported sepsis and tetanus for this comparison.
The effect on the incidence of omphalitis varied according to the severity of omphalitis. Single application was associated with an increased incidence of moderate (RR 1.53, 95% CI 1.22 to 1.92) and severe episodes (RR 2.12, 95% CI 1.10 to 4.11) of omphalitis ( Analysis 5.3; Analysis 5.4) compared with multiple applications. There was no difference between single application and multiple application groups for mild omphalitis (RR 1.14, 95% CI 0.97 to 1.34 ( Analysis 5.2)).
Compared with multiple applications, single application of chlorhexidine increased the risk of bacterial colonization with Staphylococcus aureus in one study (RR 3.63, 95% CI 2.74 to 4.82 ( Analysis 5.5)) and Enterococcus coli (RR 1.17, 95% CI 1.03 to 1.32 ( Analysis 5.7)). For bacterial colonization with streptococcus, there was no difference between single and multiple application of chlorhexidine groups in one study (RR 1.01, 95% CI 0.67 to 1.53 ( Analysis 5.6)).
Among studies conducted in hospitals settings, no study reported data for mortality, sepsis, tetanus, omphalitis or bacterial colonization.
Cord separation time
Four studies reported data for cord separation time. Among these studies, three studied triple dye and reported a significant reduction with single application compared with multiple application (MD -4.24 days, 95% CI -4.41 to -4.07, random-effects, T² = 0.00, I² = 0% ( Analysis 6.1). One study evaluated zinc powder and showed no difference in cord separation time for single application compared with multiple application (MD -0.02 days, 95% CI -0.31 to 0.27 ( Analysis 6.1)).
Washing umbilical cord versus dry cord care
One large cluster-randomized community trial (Mullany 2006) evaluated washing the umbilical cord with soap and water compared with dry cord care. No study reported data on the incidence of sepsis or tetanus.
There was no significant difference in all-cause mortality between the group washing the umbilical cord with soap and water and the dry cord care group (RR 1.00, 95% CI 0.76 to 1.32 ( Analysis 7.1)).
There was no differential effect of cord washing with soap/water compared with dry cord care for the incidence of mild (RR 1.03, 95% CI 0.87 to 1.22 ( Analysis 7.2)), moderate (RR 0.88, 95% CI 0.69 to 1.12 ( Analysis 7.3)) and severe omphalitis (RR 1.01, 95% CI 0.58 to 1.76 ( Analysis 7.4)).
Cord separation time
Cord separation times were similar between the group washing the umbilical cord with soap and water and the dry cord care group (MD 0.01 day, 95% CI -0.05 to 0.07 ( Analysis 7.5)).
No study reported data for mortality, sepsis, tetanus, omphalitis and cord separation time.
One study (Rush 1986), was conducted in hospital settings and reported no significant difference in colonization rates of Staphylococcus aureus between the cord washing and dry cord care groups (RR 0.93, 95% CI 0.65 to 1.34 ( Analysis 8.1)).
Summary of main results
This review included 34 studies, involving 69,338 babies. Five further studies are awaiting classification and there are two ongoing community trials. Included studies were conducted in both developed and developing countries. Among the 34 included trials, three were large, cluster-randomized trials conducted in community settings in developing countries and 31 studies were conducted in hospital settings mostly in developed countries. Data for community and hospital studies were analyzed separately. There were 22 different interventions studied across the included trials; 70% alcohol, triple dye and chlorhexidine were the most commonly studied antiseptics in included studies.
Studies conducted in community settings
Combined results of three large, community-based, cluster-randomized trials showed that topical application of chlorhexidine to umbilical cord stump reduced neonatal mortality and incidence of omphalitis (Arifeen 2012; Mullany 2006; Soofi 2012). One study carried out in community-based settings reported microbiologic data and showed that topical application of chlorhexidine reduces colonization of common pathologic bacteria, which correlates with a reduction in mortality and omphalitis (Arifeen 2012). There was no difference in mortality reduction with single versus multiple application of chlorhexidine to umbilical cord, however, multiple application seems advantageous for the reduction of omphalitis and bacterial colonization compared with single application.Topical application of chlorhexidine may increase cord separation time by about 1.7 days based on the combined results of two trials from community settings (Arifeen 2012; Mullany 2006). There was no beneficial effect of washing the umbilical cord with soap and water compared with dry cord care for reduction of mortality, omphalitis and cord separation time (Mullany 2006). Results of one trial showed that there was no advantage of the promotion of handwashing among caretakers for prevention of omphalitis and mortality in community settings compared with control (Soofi 2012).
Studies conducted in hospital settings
Studies conducted in hospital setting were small and had limitations. No study reported data for mortality or tetanus. Triple dye and alcohol were the most commonly studied antiseptics in hospital settings. When compared with dry cord care, no antiseptic was convincingly advantageous to reduce the incidence of omphalitis. Topical triple dye application reduced bacterial colonization with Staphylococcus aureus compared with dry cord care and that of alcohol application (two separate analyses). There was no advantage of application of alcohol and triple dye for reduction of colonization with streptococcus. Topical alcohol application was advantageous for the reduction of colonization with Enterococcus coli compared with dry cord care and triple dye application (two separate analyses). Topical application of alcohol and triple dye increased cord separation time compared with dry cord care in hospital settings. When triple dye was compared with alcohol, no significant difference was noticed in the incidence of omphalitis and cord separation time. Single application of triple dye was associated with a decrease in cord separation time compared with multiple application. Washing the cord was not advantageous compared with dry cord care. The number of studies was insufficient to make an inference about the efficacy of other antiseptics.
Overall completeness and applicability of evidence
Studies conducted in community settings
Three large, well-conducted, cluster-randomized, community trials reported data on the effectiveness of topical application of chlorhexidine for prevention of mortality and omphalitis. The aggregated sample size of these trials involved 78% of the 69,338 babies included in this review. Pooled results for all-cause mortality showed a significant reduction of 23% in the intervention group compared with control. The statistical heterogeneity for this comparison was significant (I
Studies conducted in hospital settings
Most of the studies conducted in hospital settings were small and had limitations. The sample size ranged between 71 and 2241. There were no studies conducted in hospital settings that reported mortality or tetanus. Compared with dry cord care, no antiseptic was advantageous to prevent omphalitis. The most commonly studied antiseptics were alcohol and triple dye. Both of them reduced bacterial colonization but there are no data to show that this decrease in bacterial colonization converts into prevention of clinical outcomes of mortality, sepsis, tetanus or omphalitis. Cord separation time may increase with application of alcohol and triple dye. In summary, there is no convincing evidence to recommend an antiseptic in hospital settings compared with dry cord care. Most of the hospital-based studies were conducted in developed countries. The lack of protective effect of antiseptics may be correlated with better hygiene at the time of birth and later care of babies in newborn nurseries. It is also likely that most of the hospital-based deliveries were planned and mothers were getting regular prenatal care. The chances of getting therapeutic care for perinatal infections are also greater with hospital-based deliveries. It may thus be correlated that the overall risk of infection is low in hospital-based deliveries in developed countries compared with community-based deliveries in developing countries and thus the apparent protective effect of chlorhexidine in community studies.
Quality of the evidence
|Figure 2. 'Risk of bias' graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.|
Studies conducted in community settings were large and were at low risk of bias. Pooled data showed a moderate amount of statistical heterogeneity that could be expected because of the baseline difference in study settings, topical application of other substances and baseline mortality.
Studies conducted in hospital settings were small and had outcomes that did not have clinical significance. No study reported data on mortality or tetanus.
Potential biases in the review process
This review used clearly specified inclusion and exclusion criteria and a comprehensive search strategy for the identification of relevant studies. The post-hoc decision to include studies on all antiseptics for umbilical cord care is noted in Differences between protocol and review. This decision was made to include all available evidence on umbilical cord care. Studies conducted in hospital and community studies were analysed separately because the risk factors for umbilical cord infection may differ between hospital and community settings. The comprehensive search strategy was devised to minimize publication bias by searching for both published and unpublished studies. While studies with positive results are more likely to be published than studies with negative results, studies large enough to make a difference in this review are very likely to be published. Pre-specified subgroup analyses for gestational age and geographic location were not performed because there were insufficient studies to make a conclusive comparison.
The three community trials involved 10 individual study groups that tested different frequencies and durations of chlorhexidine application along with other interventions such as washing the cord with soap and water (Mullany 2006) and promotion of handwashing among caretakers (Soofi 2012). There was no differential effect of interventions other than topical chlorhexidine application. In order to examine whether chlorhexidine has any protective effect, all the chlorhexidine groups were combined and compared with non-chlorhexidine groups. This combination of study groups is not expected to bias the results as the soap/water group in Mullany 2006 had an effect size very similar to control, i.e. dry cord care. Soofi 2012 was a factorial design trial and we included the factorial analyses in which handwashing groups were balanced between the two study groups (chlorhexidine + handwashing plus chlorhexidine only versus handwashing only plus dry cord care). Arifeen 2012 had two chlorhexidine groups, i.e. seven-day and one-day application. These were combined to include chlorhexidine groups in one arm and compared it with dry cord care.
Random-effects models were used for all meta-analyses. There are no comprehensive rules on when to use random-effects or fixed-effect models for meta-analysis (Higgins 2011). The difference between two models is that a fixed-effect model assumes that observed differences between results of trials is due to sampling variation of individual studies only whereas a random-effects model assumes that outcomes of trials might differ both because of sampling variation of individual studies and true diversity in effects. Both models can be appropriately applied for pooling data but a random-effects model is usually preferred with heterogeneity. We used random-effects models because there was substantial heterogeneity across studies in study design, settings, and package of interventions and/or intensity of delivery of those interventions.
Agreements and disagreements with other studies or reviews
The only comprehensive review on umbilical cord care was by Zupan et al that was first published in 1998. The most updated version (Zupan 2004) came to conclusion that there are not enough data to recommend in favour of or against an antiseptic for umbilical cord care. Most of the included studies were small and conducted in hospital settings. Results of our review are in agreement with those of Zupan 2004 for studies conducted in hospital settings with the addition of the effectiveness of chlorhexidine to reduce mortality and morbidity in community settings. We have not only included studies from community settings but have also included more studies from hospital settings. The updated analyses did not change the results for hospital-based studies.
Implications for practice
There is significant evidence to suggest that topical application of chlorhexidine to umbilical cord reduces neonatal mortality and omphalitis in community and primary care settings in developing countries. It may increase cord separation time however there is no evidence that it increases subsequent risk of mortality or infection.
There is not sufficient evidence to support the application of an antiseptic to umbilical cord in hospital settings compared with dry cord care in developed countries.
Implications for research
More studies are needed to explore the effects of antiseptics in hospital settings with clinically important outcomes of mortality, sepsis, tetanus and omphalitis. Studies should be undertaken in a range of settings and should account for differences in outcomes among different gestational ages.
Three community trials were conducted in South Asia. Further trials should be conducted in other parts of the world to replicate the results. Further evaluation of the effectiveness of chlorhexidine should be performed in the context of interventions packaged to facilitate their delivery by health systems. Such packages may include a wide range of interventions to reduce perinatal and neonatal mortality, such as clean delivery practices, breastfeeding, mother/infant skin-to-skin care, and delayed bathing.
We would like to acknowledge Professor Lourdes Amarillo and Professor Cynthia Cordero for their statistical contributions.
We also thank Sara Roden-Scott for her translation of Davila 2007.
We would like to give special thanks to Sonja Henderson from Pregnancy and Childbirth Cochrane Review Group for her continuous support and co-ordination during the conduct of this review.
The National Institute for Health Research (NIHR) is the largest single funder of the Cochrane Pregnancy and Childbirth Group. The views and opinions expressed therein are those of the authors and do not necessarily reflect those of the NIHR, NHS or the Department of Health.
Data and analyses
- Top of page
- Authors' conclusions
- Data and analyses
- What's new
- Contributions of authors
- Declarations of interest
- Sources of support
- Differences between protocol and review
- Index terms
Appendix 1. LILACS search strategy
((Pt randomized controlled trial OR Pt controlled clinical trial OR Mh r andomized controlled trials OR Mh random allocation OR Mh double-blind method OR Mh single-blind method) AND NOT (Ct animal AND NOT (Ct human and Ct animal)) OR (Pt clinical trial OR Ex E05.318.760.535$ OR (Tw clin$ AND (Tw trial$ OR Tw ensa$ OR Tw estud$ OR Tw experim$ OR Tw investiga$)) OR ((Tw singl$ OR Tw simple$ OR Tw doubl$ OR Tw doble$ OR Tw duplo$ OR Tw trebl$ OR Tw trip$) AND (Tw blind$ OR Tw cego$ OR Tw ciego$ OR Tw mask$ OR Tw mascar$)) OR Mh placebos OR Tw placebo$ OR (Tw random$ OR Tw randon$ OR Tw casual$ OR Tw acaso$ OR Tw azar OR Tw aleator$) OR Mh research design) AND NOT (Ct animal AND NOT (Ct human and Ct animal)) OR (Ct comparative study OR Ex E05.337$ OR Mh follow-up studies OR Mh prospective studies OR Tw control$ OR Tw prospectiv$ OR Tw volunt$ OR Tw volunteer$) AND NOT (Ct animal AND NOT (Ct human and Ct animal))) AND (umbilical OR umbilicus) AND (cord OR cordón OR care OR hygiene OR higiene OR separate OR separation OR chlorhexidine OR clorhexidina OR iodin$ OR alcohol OR clean OR cleanse OR cleansing OR Mh Anti-Infective Agents OR regime OR regimen OR topical OR topically OR “triple dye” OR antiseptic$ OR anti-septic$ or antiséptic$ OR antimicrob$ OR omphalitis)
Aydin, 27 July 2013
Chlorhexidine gluconate is a widely used topical antiseptic that is recommended by the Center for Disease Control and Prevention for skin cleansing before central venous catheter insertion in adults and children. But, because of limited safety data, infection prevention guidelines do not recommend its use in babies who are less than two months old. There are many reports on the potential toxicity of chlorhexidine gluconate, for example genotoxic and cytotoxic effects (1). Chlorhexidine gluconate contact may cause corneal damage, ototoxicity, and neurotoxicity (2). In one study chlorhexidine gluconate was detected in the blood of preterm infants receiving chlorhexidine gluconate skin antisepsis for peripherally central catheter insertion (3). Further investigation is required to determine the safety of this use for newborn infants, especially preterm infants.
1. Arabaci T, et al. Assessment of cytogenetic and cytotoxic effects of chlorhexidine digluconate on cultured human lymphocytes. Acta Odontol Scand 2013 71: 1255-60.
2. Zinn J, Jenkins JB,et al. Intraoperative patient skin prep agents: is there a difference. AORN J 2012 92: 662-74.
3. Chapman AK, et al. Absorption and tolerability of aqueous chlorhexidine gluconate used for skin antisepsis prior to catheter insertion in preterm neonates. J Perinatol 2013 33):768-71. doi: 10.1038/jp.2013.61. Epub 2013 May 23.
Comment submitted by Mustafa Aydin, July 2013
Last assessed as up-to-date: 25 March 2013.
Contributions of authors
Aamer Imdad contributed to the background. Aamer Imdad and Jacinto Blas V Mantaring were primarily responsible for the methods. Jacinto Blas V Mantaring carried out the additional searches listed in the methods section. Aamer Imdad, Jacinto Blas V Mantaring, Resti Ma M Bautista, Ma Esterlita V Uy and Kathlynne Anne Abat-Senen reviewed citations for inclusion, extracted data and formulated the 'Characteristics of included study' table along with 'Risk of bias' table. Aamer Imdad, Jacinto Blas V Mantaring, Resti Ma M Bautista, Ma Esterlita V Uy and Kathlynne Anne Abat-Senen entered outcome data into RevMan, analyzed the data and wrote the results. Jacinto Blas V Mantaring, Resti Ma M Bautista, Ma Esterlita V Uy and Kathlynne Anne Abat-Senen drafted the discussion text, which Aamer Imdad and Zulfiqar Ahmed Bhutta subsequently revised. Zulfiqar Ahmed Bhutta provided supervision and is the guarantor for the review.
Declarations of interest
Sources of support
- Effective Health Care Research Programme Consortium, Philippines.
- SEA-ORCHID Project (South-East Asia - Optimising Reproductive and Child Health Outcomes in Developing Countries), Australia.
Differences between protocol and review
- It was decided post-hoc to include studies on all antiseptics rather than just alcohol. This was to include all possible evidence on umbilical cord care.
- Comparison of dry cord care with other antiseptics was not performed as there was too much clinical heterogeneity in the comparison group.
- It was decided not to pool community and hospital-based studies together because of significant clinical heterogeneity of population of patients in hospitals and community settings.
- Subgroup analyses based on gestational age and that of geographical regions were not performed as the number of studies were not sufficient.
Medical Subject Headings (MeSH)
*Umbilical Cord [microbiology]; Anti-Infective Agents, Local [*administration & dosage]; Chlorhexidine [administration & dosage]; Infant, Newborn; Inflammation [mortality; prevention & control]; Randomized Controlled Trials as Topic; Sepsis [mortality; *prevention & control]
MeSH check words
* Indicates the major publication for the study