Chlorhexidine Versus Povidone-Iodine As Cutaneous Antisepsis for Prevention of PEG Site Infection. A Prospective Randomized Pilot Study

Abstract

Introduction: Percutaneous endoscopic gastrostomy (PEG) tube site infection is the most common complication following PEG tube placement. To reduce infection, antibiotic prophylaxis is given in addition to cutaneous antisepsis using an iodine skin preparation during placement of a PEG tube. Chlorhexidine skin preparation was shown in several studies to be superior to iodine in reducing catheter related infections and wound infections in clean abdominal surgery. However, there is no PEG study to date which addresses this question. This pilot randomized trial compares chlorhexidine (C) versus povidone-iodine (P-I) as cutaneous antisepsis for the prevention of PEG site infections. Method: Patients were prospectively randomized to C or P-I. The groups were similar with respect to gender, BMI, race and hospital status (inpatient vs. outpatient). All patients received intravenous prophylactic antibiotics before the procedure. PEG tube placement was done in a standard fashion at a single tertiary care center using the “pull- through” technique. Patients were followed by a blinded physician (the same physician in all cases) on post procedure day number 5 and 10. Primary endpoints included the presence of PEG tube site infection at any stage within 10 days of placement. Infection was measured based on a standard scoring system that includes erythema, induration and exudate. Results: Sixty-eight patients were randomized with 36 (53%) receiving C and 32 (47%) P-I. Two patients in the C group were excluded due to non-endoscopic related pre-procedure complications. Nine patients developed a PEG site infection (14%), three patients in the C group (9%) and six patients in P-I group (19%) (p = 0.30 Fisher's exact test). There was a 53% reduction of PEG site infections with the use of C compared with P-I. Seven out of the nine (78%) infections occurred in patients who were in a hospital inpatient setting. Conclusions: Based on the results of this pilot study, infections were 53% less common with chlorhexidine than with povidone-iodine. However, the difference was not statistically significant possibly due to the relatively small sample size. A larger randomized trial is warranted to find out whether chlorhexidine is superior to povidone iodine before any recommendations are made to switch to chlorhexidine as primary cutaneous antisepsis during PEG placement.