The Effect of Chlorhexidine Bathing on MRSA and VRE Colonization Among Haematology-Oncology ICU Patients

The study was designed as an experimental, 2-arm, cross-over clinical trial, involving patients admitted to ICU of a university oncology hospital between September 2018 and July 2019.

This study was conducted at one medical intensive care unit of an oncology hospital, which is one of the four hospitals of a state university located at the capital city of Turkey. The University is the first public institution in Turkey accredited by the Joint Commission International (JCI) for its quality of health care services. The ICU of the oncology hospital provides care for patients with a variety of medical conditions, but particularly specialized for critical patients with oncological and haematological diseases. The ICU includes eight beds and two isolation rooms each of which has one patient bed, and the average daily admission of patients to ICU was 7.

Using the NCSS-PASS 2007 statistical package program (Blackwelder, 1998), the minimum sample size with 80.0% power and type 1 error of 5.0% was calculated as 58 patients for an inclusion period of 1 year; however 61 patients who met the inclusion criteria composed the study sample.

All patients admitted to the ICU were assessed for eligibility and included to the study sample if they aged over 18 years, diagnosed with a hematologic-oncologic disease and admitted within the first 24 hours to the ICU. Exclusion criteria for participation were; age < 18 years, burns to >20% of the total skin surface, pregnancy, previous MRSA and/or VRE infection history or antibiotic use for these infections, receiving radiation therapy, admitted before 24 hours to ICU, re-admission to ICU, diagnosed with severe septic shock, massive pulmonary thromboembolism, massive haemoptysis, and status epilepticus etc., general condition disorder.

A total of 139 patients with a hematologic-oncologic disease were assessed for eligibility of whom 61 excluded due to discharge (n=11) or death (n=6) before 24 hours of hospitalization, admission to ICU before the eligibility assessment of researcher (n=8), repeated admission to ICU (n=6), severely impaired medical condition (n=6), age under 18 years (n=2), massive pulmonary thromboembolism (n=2), previous MRSA infection (n=2), and severe septic shock (n=1). Seventeen patients did not provide consent for participation and their relatives were not available. After eligibility assessment, 78 patients were randomised into two arms according to the order of admission to the ICU. Arm 1 (39 patients) included first the control period, followed by one-day wash-out, and then the intervention period. Arm 2 (39 patients) included first the intervention period, followed by one-day wash-out, and then the control period. The first patient was assigned to the second arm after the draw, and the patients continued to take the first arm and second arm respectively. Each period lasted three consecutive days with a total of 7 days for each patient. During the period of the study, 8 patients from Arm1 and 9 patients from Arm 2 could not complete the study due to several reasons. Thus, the final sample consisted of 31 patients in Arm 1 and 30 patients in Arm 2.

Following the ethical approval of the study protocol, official permissions, and informed consents, the researcher started to work at the ICU every day from 8.00 am to 8.00 pm to collect data and perform the procedures. A patient information and evaluation form was used to collect data on patients' demographic and clinical characteristics, including the Glasgow coma scale, medications ordered for each patient, patient risk factors, skin reactions, and bathing protocols. This form was filled out every morning for each patient except the demographic characteristics. The MRSA and/or VRE colonization results of swab samples taken before and after the bathing procedures were also recorded in this form. Medical conditions and the APACHE II scores of patients were evaluated and recorded to patient electronic files.

Procedures for control period included providing daily bed bathing with soap and water over three consecutive days, while intervention period included daily bed bathing with 2% CHG solution over three consecutive days. Patients were bathed from the neck down, avoiding contact with face, mucous membranes and wounds by wiping with soap or CHG for each period. The patient's skin bathed in the order of clean area to dirty area. Additionally, if patients in both arms became soiled after the daily baths, contaminated body areas were wiped using water and disposable washcloths. Patients were assessed daily for localised or body-wide skin reactions. During control period, patients were washed with soap and then rinsed with water, and dried with disposable towels. During intervention period, 4% CHG solution was diluted with water. In order to dilute the 4% CHG solution, 120 ml (~4 ounce) of 4% CHG solution was added to 120 ml (~4 ounce) of water in a disposable box each time (to a final concentration of 2%) (Swan et al., 2016) and this solution was applied to the patient body directly. The lowest concentration of the extract inhibited growth of bacteria was selected as minimum inhibitory concentration (MIC) (Knapp, 2014; Vali, Dashti, El-Shazly, \& Jadaon, 2015) and diluting CHG to a concentration as low as 0.4% does not affect CHG antiseptic efficacy (Bajaj, Loh, \& Borgstrom, 2014).

The swab samples were collected from nares, groins, and perirectal area of each patient within 24 hours of admission to the ICU and on days 3, 5, and 7; immediately before and 4-6 hours after the bathing by the researcher. Totally, 24 swab samples (3 area x 2 times x 4 days) were taken from each patient at the end of the periods. In order to obtain the sample, a sterile cotton swab was moistened with sterile buffered transport medium and each swap specimen was placed in a vial containing transport media and transported to the clinical microbiology laboratory of the study institution by the researcher (first author).

The swab samples were cultured on 5% sheep blood agar and incubated for 16-18 hours at 37⁰C. Typical colonies were examined by Gram staining, catalase and plasma coagulase tests and were identified as S. aureus .In order to determine the resistance to methicillin, disc diffusion test was performed by using cefoxitin (30 μg, Oxoid, UK) disc on Mueller-Hinton agar (Oxoid, UK) in accordance with EUCAST recommendations (Matuschek, Brown, \& Kahlmeter, 2014). Resistance was confirmed after the determination of mecA gene by a commercially-available polymerase chain reaction (PCR) assay; (BD MaxTM MRSA XT, Canada) (Baby et al., 2017).

In order to detect VRE colonization, chromID® VRE (bioMerieux, France) selective media was used and typical colonies formed after 24 hours of incubation at 37⁰Cwere identified as Enterococcus faecium by API-ID Strep (bioMerieux, France). Vancomycin resistance was determined by vancomycin E-test (Oxoid, UK) and vanA and vanB, genes were examined by by PCR method (BD GeneOhmTM, Ireland (He et al., 2019).

For detection of both bacteria, non-growth media were re-evaluated after incubating for another 48 hours under the same conditions. The ICU physicians in the study unit were informed for MRSA and/or VRE-positive patients by the researcher. and all identified MRSA and / or VRE colonized patients were placed on contact precautions.