Bacterial sensitivity to chlorhexidine and povidone-iodine antiseptics over time: a systematic review and meta-analysis of human-derived data

Abstract

Surgical site infection (SSI) is the most common complication of surgery, increasing healthcare costs and hospital stay. Chlorhexidine (CHX) and povidone-iodine (PVI) are used for skin antisepsis, minimising SSIs. There is concern that resistance to topical biocides may be emergeing, although the potential clinical implications remain unclear. The objective of this systematic review was to determine whether the minimum bactericidal concentration (MBC) of topical preparations of CHX or PVI have changed over time, in microbes relevant to SSI. We included studies reporting the MBC of laboratory and clinical isolates of common microbes to CHX and PVI. We excluded studies using non-human samples and antimicrobial solvents or mixtures with other active substances. MBC was pooled in random effects meta-analyses and the change in MBC over time was explored using meta-regression. Seventy-nine studies were included, analysing 6218 microbes over 45 years. Most studies investigated CHX (93%), with insufficient data for meta-analysis of PVI. There was no change in the MBC of CHX to Staphylococci or Streptococci over time. Overall, we find no evidence of reduced susceptibility of common SSI-causing microbes to CHX over time. This provides reassurance and confidence in the worldwide guidance that CHX should remain the first-choice agent for surgical skin antisepsis.

Figure 1

PRISMA flowchart—the following studies were excluded from the systematic review: (1) no MBC calculated, (2) use of biocide and alcohol mixtures, (3) use of non-human bacteria, (4) use of dental preparations of biocides, (5) no data on the strength of biocide used.

Figure 2

Forest plot of the mean MBC for different species and families of bacteria.

Figure 3

A scatterplot of study-level estimates of mean MBC over time, for Staphylococci. The size of the points corresponds to the precision (inverse variance) of the study, whereby larger bubbles are more precise (bigger and so, more accurate) studies.

Figure 4

A scatterplot of study-level estimates of mean MBC over time, for Streptococci. The size of the points corresponds to the precision (inverse variance) of the study, whereby larger bubbles are more precise (bigger and so, more accurate) studies.