Medical-grade honey does not reduce skin colonization at central venous catheter-insertion sites of critically ill patients: a randomized controlled trial

INTRODUCTION Catheter-related bloodstream infections (CRBSIs) associated with short-term central venous catheters (CVCs) in intensive care unit (ICU) patients are a major clinical problem. Bacterial colonization of the skin at the CVC insertion site is an important etiologic factor for CRBSI. The aim of this study was to assess the efficacy of medical-grade honey in reducing bacterial skin colonization at insertion sites. METHODS A prospective, single-center, open-label randomized controlled trial was performed at the ICU of a university hospital in The Netherlands to assess the efficacy of medical-grade honey to reduce skin colonization of insertion sites. Medical-grade honey was applied in addition to standard CVC-site dressing and disinfection with 0.5% chlorhexidine in 70% alcohol. Skin colonization was assessed on a daily basis before CVC-site disinfection. The primary end point was colonization of insertion sites with >100 colony-forming units at the last sampling before removal of the CVC or transfer of the patient from the ICU. Secondary end points were quantitative levels of colonization of the insertion sites and colonization of insertion sites stratified for CVC location. RESULTS Colonization of insertion sites was not affected by the use of medical-grade honey, as 44 (34%) of 129 and 36 (34%) of 106 patients in the honey and standard care groups, respectively, had a positive skin culture (P = 0.98). Median levels of skin colonization at the last sampling were 1 (0 to 2.84) and 1 (0 to 2.70) log colony-forming units (CFUs)/swab for the honey and control groups, respectively (P = 0.94). Gender, days of CVC placement, CVC location, and CVC type were predictive for a positive skin culture. Correction for these variables did not change the effect of honey on skin-culture positivity. CONCLUSIONS Medical-grade honey does not affect colonization of the skin at CVC insertion sites in ICU patients when applied in addition to standard disinfection with 0.5% chlorhexidine in 70% alcohol. TRIAL REGISTRATION Netherlands Trial Registry, NTR1652.