A Randomized, Double-Blind, Placebo-Controlled Trial of Selective Digestive Decontamination Using Oral Gentamicin and Oral Polymyxin E for Eradication of Carbapenem-Resistant Klebsiella pneumoniae Carriage

OBJECTIVE. To assess the effectiveness of selective digestive decontamination (SDD) for eradicating carbapenem-resistant Klebsiella pneumoniae (CRKP) oropharyngeal and gastrointestinal carriage. DESIGN. A randomized, double-blind, placebo-controlled trial with 7 weeks of follow-up per patient. SETTING. A 1,000-bed tertiary-care university hospital. PATIENTS. Adults with CRKP-positive rectal swab cultures. METHODS. Patients were blindly randomized (1 : 1) over a 20-month period. The SDD arm received oral gentamicin and polymyxin E gel (0.5 g 4 times per day) and oral solutions of gentamicin (80 mg 4 times per day) and polymyxin E (1 x 10(6) units 4 times per day for 7 days). The placebo arm received oral placebo gel 4 times per day and 2 placebo oral solutions 4 times per day for 7 days. Strict contact precautions were applied. Samples obtained from the throat, groin, and urine were also cultured. RESULTS. Forty patients (mean age +/- standard deviation, 71 +/- 16 years; 65% male) were included. At screening, greater than or equal to 30% of oropharyngeal, greater than or equal to 60% of skin, and greater than or equal to 35% of urine cultures yielded CRKP isolates. All throat cultures became negative in the SDD arm after 3 days (P < .0001). The percentages of rectal cultures that were positive for CRKP were significantly reduced at 2 weeks. At that time, 16.1% of rectal cultures in the placebo arm and 61.1% in the SDD arm were negative (odds ratio, 0.13; 95% confidence interval, 0.02-0.74; P < .0016). A difference between the percentages in the 2 arms was still maintained at 6 weeks (33.3% vs 58.5%). Groin colonization prevalence did not change in either arm, and the prevalence of urine colonization increased in the placebo arm. CONCLUSIONS. This SDD regimen could be a suitable decolonization therapy for selected patients colonized with CRKP, such as transplant recipients or immunocompromised patients pending chemotherapy and patients who require major intestinal or oropharyngeal surgery. Moreover, in outbreaks caused by CRKP infections that are uncontrolled by routine infection control measures, SDD could provide additional infection containment. Infect Control Hosp Epidemiol 2012; 33(1):14-19