Oral Mucosal Decontamination With Chlorhexidine for the Prevention of Ventilator-Associated Pneumonia in Children

A Randomized, Controlled Trial

Meghna Raju Sebastian, MD; Rakesh Lodha, MD; Arti Kapil, MD; Sushil K. Kabra, MD


Pediatr Crit Care Med. 2012;13(5):E305-310. 

In This Article


From November 2007 to April 2009 (excluding the 2-month span during the safety review of July and August 2008), 283 patients were admitted to the PICU. All these patients were assessed for eligibility. Of the patients assessed, 197 were excluded. Among them, 56 were never intubated, 59 were under the age of 3 months, 47 had been mechanically ventilated for over 24 hrs prior to PICU admission, 14 died within 24 hrs of PICU admission, 13 were immune deficient (these patients were excluded before April 2008), four were tracheostomized, two refused consent, and one each were beyond the age of 15 yrs and extubated prior to inclusion. Of the 86 patients who were enrolled, stratified, and randomized, 45 received placebo and 41 chlorhexidine. Two patients in each group were considered as protocol deviations as these children were taken away from the hospital against medical advice. However, all the enrolled patients were analyzed in an intention to treat manner (Fig. 1).

Figure 1.

Flow of participants. PICU, pediatric intensive care unit.

Baseline data collected included demographic variables such as age and sex as well as other relevant variables including underlying diagnosis and its severity, underlying infection, clinical evidence of pneumonia, baseline BAL cultures, and antibiotic use at admission, and were found to be comparable in the chlorhexidine and placebo groups (Table 1).

The mean amount of gel application received by children in the chlorhexidine and placebo groups was 10.9 ± 8.3 g and 10.4 ± 8.6 g, respectively. The children were mechanically ventilated for 7.89 ± 5.7 days and 8.2 ± 10.3 days, respectively.

The primary outcome analyzed was the occurrence of VAP in patients who received chlorhexidine vs. placebo. A final diagnosis of VAP, based on satisfaction of the Centers for Disease Control and Prevention criteria, was made in 12 of the 41 children who received chlorhexidine and 14 of the 45 who received placebo (p = .853); the absolute risk reduction was 1.8% (95% confidence interval −17.5% to 21%) while the relative risk reduction was 5.9%. The rates of VAP were found to be nearly the same, i.e., 39.6/1000 ventilator days for chlorhexidine and 38.14/1000 ventilator days for placebo (Table 2). The lack of a reduction of risk of VAP with chlorhexidine did not change whether or not the child had preexisting pneumonia. In children with underlying pneumonia, 27.6% (8 of 29) of those who were given placebo developed VAP as opposed to 19.2% (5 of 26) of those who were given chlorhexidine (p = .467). In those without underlying pneumonia, 37.5% (6 of 16) and 46.6% (7 of 15) of those who received placebo and chlorhexidine developed VAP (p = .605), respectively. These differences were insignificant.

The mean day of diagnosis of VAP in chlorhexidine group was 8.8 ± 5.6 days compared with 8.1 ± 4.4 days in the placebo group.

There was no difference in risk of VAP with age, gender, admission diagnosis, or Pediatric Index of Mortality 2 score. The patients who developed VAP had significantly longer durations of ventilation (p = .001), ICU stay (p = .0001), and hospital stay (p = .0008) as compared with those who did not develop it. The median duration of ventilation in these two groups were 10.5 and 3.5 days, respectively. These children stayed in ICU for nearly twice (12 days) as long as those who were pneumonia free (5 days). Also VAP led to an increased duration of hospital stay, and the time to discharge was about half as long if the patient did not develop this complication.

Despite leading to a significantly longer period of ventilation, ICU, and hospital stay, the occurrence of VAP in a child did not significantly alter the final outcome in the studied patients. The mortality, though higher with VAP (53.8% vs. 40%), was not statistically significant.

No significant differences were observed in any of the secondary outcomes (Table 3 and Table 4). None of the patients were found to have any local or systemic adverse reactions to the chlorhexidine or placebo gels used during the course of the trial.