The Present Clinical Trial Is a Pilot Non-Blinded, Prospective Study Performed in a Single

Methods

The present clinical trial is a pilot non-blinded, prospective study performed in a single ICU of the University Hospital of Pisa, Italy, to assess the feasibility and the efficacy of administering oral gentamicin for KPC-Kp gut prophylaxis.

The study was conducted in the post-surgery ICU where the first outbreak of KPC-Kp occurred in the Pisa Hospital and remained thereafter a reservoir of KPC-Kp colonization/infection. The study was approved by the Institutional Review Board of the University Hospital of Pisa, Italy, and all patients signed a written informed consent.

Patients: In a twelve-month period, 31 consecutive adult patients admitted to this ICU for monitoring after hepatectomy were administered oral gentamicin (solution of 80 mg four times daily administered in the gastric tube) starting at the time of surgery and continued until discharge from the ICU, via a nasogastric tube during sedation. Oral gentamicin was combined with a surgical antibiotic prophylaxis consisting of ampicillin 2 g / sulbactam 1 g every 6 hours only for the first 24 hours. In addition to the prophylaxis, ICU staff strictly followed the same procedure described in the CDC guidelines, consisting of patient isolation or cohorting with strict contact precautions and dedicated caregiver and equipment, for the entire period of the study.

The control group is represented by 31 consecutive patients admitted to the same ICU after surgery performed by the same team in the year before the prophylaxis study. Furthermore, in the year of the study and in the year before, colonization rates were recorded in all patients admitted to the same ICU.

Laboratory analysis: Rectal swab cultures were performed at the time of hospital admission and every 3 days thereafter since to hospital discharge. A validated direct screening method who allowed a rapid detection of intestinal carriage of KPC-Kp on direct plating of rectal swabs on to MacConkey agar with a meropenem disc and a meropenem disc plus 600 mg of 3-aminophenylboronic acid was used to identify patients with KPC-Kp gut colonization. KPC-Kp stool isolates were identified with the mini-API system (bioMérieux, Mercy L'Etoile, France) or by matrix-assisted laser desorption ionization–time of flight (MALDI-TOF) mass spectrometry (Vitek MS, bioMérieux).

Antimicrobial susceptibility testing was performed using a disk diffusion method according to EUCAST version 3.1 guidelines, and MICs of gentamicin were determined with E-test, according to the manufacturer’s instruction (AB Biodisk, Solna, Sweden). Interpretation of susceptibility data was according to EUCAST breakpoints (version 3.1). The gentamicin plasma concentration was determined with a fluorescence immunoassay method (Abbott, Abbott Park, IL) on day 2 of treatment on samples obtained 2 h after drug administration (detection limit 0,05 mg/L).

Adverse drug reaction to gentamicin were identified and scored on the basis of the NCI Common Toxicity Criteria, vers. 4.0.

Study endpoints: The primary study endpoint was the rate of KPC-Kp gut colonization, defined by the development of a positive rectal swab cultures. The data obtained in the 31 consecutive patients who received the antibiotic therapy to prevent KPC-Kp gut colonization were compared with controls.

Statistical analysis: data are expressed as mean ± standard deviation, median and interquartile range or proportion, as appropriate. Comparisons between groups were performed with the unpaired t test, Mann-Whitney test, or chi-square test with continuity correction, as appropriate. Factors related to gut colonization events during oral gentamicin prophylaxis were identified with multivariate logistic regression; each variable that showed p < 0.10 at univariate logistic regression for colonization events, was included in a multivariate logistic regression model; to address the problem constituted by the small number of patients with positive endpoints, backward stepwise variable selection was performed. The stability of the obtained model was assessed by bootstrap analysis. P-values less than 0.05 was considered statistically significant. All computation were done with R statistical software (R, version 2.11.1,2010; R Development Core Team 2006; R Foundation for Statistical Computing, Vienna, Austria).