The study was conducted within Nanjing Drum Tower Hospital, a 3,325-bed general tertiary care and university-affiliated teaching hospital in Nanjing, Jiangsu province, China. Ethical approval was approved by the Ethics Committee of Nanjing Drum Tower hospital (Number: 2013-042).
The study design was shown in Figure 1. Briefly, the samples of patients hospitalized into in a 27-bed intensive care unit (ICU) in our hospital were taken to monitor MDR-AB from 2013 to 2014. The HTCS were marked by fluorescence labeling and continuously monitored for clearance level of fluorescence labeling and for contamination rate of MDR-AB, which was resistant against at least one agent in three or more of tested antimicrobial categories (18).
Investigation on the hand hygiene compliance
Observational survey of compliance with hand hygiene were conducted. Healthcare personnel was required to perform hand hygiene in accordance with the guidelines recommended by Centers for Disease Control and Prevention (CDC) (19). An alcohol-based hand rub or wash with soap and water should been used according to following the special indications for hand hygiene (20). The potential opportunities for hand hygiene were recorded according to recommended guidelines (21-23), and the actual number of episodes of handwashes and hand rubs were also noted. 20-min observations were conducted at optional time periods throughout the week. Healthcare personnel did not know the schedule of observation periods. The observers were as unobtrusive as possible, but were not hidden.
Labeling of fluorescence marks and determination of the clearance level
The fluorescence marks were drawn by a fluorescent pen (RUHOF), which uses a special nontoxic target solution. When exposed to black light, the marks emit fluorescence brightly. Noteworthily, the fluorescence marks are inconspicuous, dry rapidly on surfaces, remain environmentally stable for several weeks, resist dry abrasion, but could be easily removed by minimal abrasion with moistened cloth (17, 24). Fluorescence labeling was performed twice a day before cleaning (at 10 a.m. and 5 p.m. respectively), one mark every square centimeter. The targets were evaluated after the twice daily routine cleaning. Terminal cleaning was performed after inpatients were transferred from the ICUs. Clearance level of fluorescence labeling was calculated by comparing the number of fluorescence marks before and after cleaning. Environment cleanliness was divided into cleaning (labeling clearance rate >80%) and contamination (labeling clearance rate < 80%).
Monitoring contamination of MDR-AB on the HTCS
To monitor contamination of MDR-AB on the HTCS of the ICUs, before the daily cleaning was performed, samples were accordingly taken for bacterial culture from each site of fluorescence labeling by a cotton swab moistened with saline, according to the Technical Specification for Disinfection of Hospital Disinfection Hygiene Standard issued by the Ministry of Health of China (http://www.biaozhun8.cn/biaozhun108760). Once A. baumannii was detected, antimicrobial susceptibility was further implemented to detect MDR-AB.
The colonization and infection rates of MDR-AB among inpatient in the ICU
Clinical samples including sputum, urine, blood, etc., from patients within the ICU during 2013-2014 were routinely taken and sent to the clinical microbiology laboratory for bacterial culture and susceptibility testing once infections were suspected. The diagnostic criteria for colonization and infection referred to the criteria issued by the US CDC in 2008 (25). According to the international epidemiological quantitative statistical methods, the newly isolated multidrug-resistant bacteria per thousand bed days was adopted as the quantitative statistical standard, that is, the detection or infection density of multidrug-resistant bacteria in a specific time range (Number of newly isolated multidrug-resistant bacteria infected or colonized new patients in a period/number of hospital days in a period).
Bacterial identification and antimicrobial susceptibility testing
Strains isolated were identified by ATB32E or Vitek-2 technology (BioMerieux, France). The susceptibility was determined by Kirby-Bauer method. The tested antimicrobial agents were as follows: amikacin, ceftazidime, cefoperazone/sulbactam, imipenem, meropenem, piperacillin-tazobactam, cefepime, ticarcillin/clavulanate, ciprofloxacin, levofloxacin, sulfamethoxazole, minocycline and tigecycline. Escherichia coli American Type Culture Collection (ATCC) 25922 and Pseudomonas aeruginosa ATCC27853 were used as the quality controls in parallel. The results were interpreted according to guidelines of Clinical Laboratory Standard Institute (CLSI) 2015(26). However, the interpretation of tigecycline was referred to the guidelines of the current European Committee on Antimicrobial Susceptibility Testing (EUCAST) (www.eucast.org), cutoff MICs of ≤1 μg/ml and >2 μg/ml were used for tigecycline as the susceptibility and resistance breakpoints, respectively.
Pulsed field gel electrophoresis
When 14 MDR-AB isolates were detected simultaneously during Jan-Mar, 2013, the genetic relatedness among those MDR-AB strains collected from the patients and the HTCS during the same period were further analyzed through pulsed field gel electrophoresis (PFGE) according to the protocol (27). Briefly. Fresh and pure bacterial cultures were embedded in agarose plugs and digested with proteinase K (20 mg/mL), followed by paI restriction endonuclease (TaKaRa, Dalian, Beijing, China). The standard strain Salmonella enterica serotype Braenderup H9812 digested with XbaI was used as a marker. The electrophoresis was performed in 0.5 × TBE buffer in a pulsed-field electrophoresis system (Chef Mapper; Bio-Rad Laboratories, Hercules, CA, USA), and the conditions were as follows: 14°C, 6 V/cm, switch angle 120°, switch ramp 5–20 s for 19 h. BioNumerics software version 7.6 (Applied Maths, Sint-Martens-Latem, Belgium) was used to analyze the PFGE banding patterns. A cut off of 85% was used to judge the relatedness of strains analyzed based on the tree constructed by the unweighted pair group method of averages and a position tolerance of 1.5%.
IBM SPSS Statistics 20.0 software was used to perform statistical analysis. To determine whether there are statistical outliers among the fluorescence label clearance rates, we performed multivariate linear regression analysis to check their Mahalanobis distance. Different marker numbers in the 8 quarters was tested for Normal distribution. The correlation between the removal level of fluorescence labeling and colonization rates of MDR-AB, and the relationship between the colonization rates and infection rates of MDR-AB were analyzed by the Spearman correlation analysis. P < 0.05 was taken as statistically significant.