[Risk factors and antibiotic use in methicillin-resistant Staphylococcus aureus bacteremia in hospitalized patients at Hacettepe University Adult and Oncology Hospitals (2004-2011) and antimicrobial susceptibilities of the isolates: a nested case-control study].

The aim of this study was to evaluate methicillin-resistant Staphylococcus aureus (MRSA) bacteremia cases who were followed at the Infectious Diseases Unit of Internal Medicine Department, at Hacettepe University Adult and Oncology Hospitals between January 2004-December 2011. A total of 198 patients, of them 99 had positive MRSA blood cultures (case group), and 99 without MRSA bacteremia (control group) who were selected randomly among patients at the same wards during the same time period, were included in the study. Demographic data, risk factors for MRSA bacteremia and antibiotic use of case (60 male, 39 female; mean age: 59.37 ± 16.96 yrs) and control (60 male, 39 female; mean age: 59.11 ± 17.60 yrs) groups were obtained from the patient files and the hospital data system and were compared. Methicillin susceptibility was determined by the cefoxitin (30 µg, BD, USA) disc diffusion method and confirmed by mecA PCR test. Antimicrobial susceptibilities were also determined by disc diffusion and Etest (BioMerieux, France) methods according to CLSI guidelines. There was no statistically significant difference between the two groups according to age, gender, presence of an underlying chronic disease, burn, hemodialysis, malignancy or immunosupression (p> 0.05). The results of the univariate analysis revealed that antibiotic use and parameters most likely to be associated with MRSA bacteremia (obesity, cerebrovascular event, hospitalization history, central/arterial catheter, presence of tracheostomy, invasive/non-invasive mechanical ventilation, use of proton pump inhibitors, H2 receptor blockers, sucralfate, nasogastric or urinary tubes, gastrostomia, total parenteral nutrition, acute organ failure and surgical operation) were found to be statistically higher in the case group (p< 0.05). Median length of hospital stay was also higher in the case group (59 days versus 8 days; p< 0.001). Multivariate regression analysis indicated that obesity (OR= 7.98; p= 0.013), central venous catheterization (OR= 6.65; p= 0.005), nasogastric tube (OR= 16.58; p< 0.001) and use of H2 receptor blockers (OR= 4.41; p= 0.010) were independent risk factors. The number of patient given at least one antibiotic (92 in case group, 51 in control group) was statistically higher than those who were not (48 in case group, 7 in control group) (OR= 14.86; p< 0.001). Use of antibiotics [ampicillin-sulbactam and/or amoxicillin-clavulanate, fluoroquinolones, aminoglycosides, piperacillin-tazobactam (TZP), meropenem (MEM), imipenem (IPM), vancomycin (VAN), cephalosporins and teicoplanin (TEC)] were found to be statistically significantly higher in the case group by univariate analysis (p< 0.05). In multivariate analysis, it was determined that TZP (OR= 6.82; p< 0.001), IPM (OR= 3.97; p= 0.023) and VAN (OR= 8.46; p= 0.001) use were independent risk factors in MRSA bacteremia. The duration of MEM (p= 0.037) and cephalosporin use (p< 0.001) were significantly longer in the case group, however there was no statistically significant difference between the duration of use of other antibiotics (p> 0.05). All MRSA isolates were mecA gene positive (n= 99), the resistance rates for ciprofloxacin, rifampin, gentamicin, tetracyclin, cefoxitin, erythromycin and clindamycin were 95%, 95%, 94%, 96%, 98%, 71% and 36%, respectively. All of the isolates were found to be susceptible to trimethoprim-sulfamethoxazole, VAN, TEC, tigecycline, linezolid and daptomycin. Mortality rates in patients who were infected with MRSA strains exhibiting vancomycin MIC value of ≤ 1.0 µg/ml (n= 49) and with MRSA strains exhibiting MIC > 1.0 µg/ml (n= 50) were 34.6% (17/49) and 60% (30/50), respectively. This difference was found to be statistically significant (p= 0.012). Thus it was concluded that the mortality rate increased in patients infected with MRSA with high (> 1.0 µg/ml) vancomycin MIC value. The results of this study indicated that obesity, presence of central venous catheter and nasogastric tube, and the use of H2 receptor blockers, IPM, TZP and VAN were independent risk factors for MRSA bacteremia. This was the first study showing the relationship between increasing mortality and high vancomycin MIC values in MRSA bacteremia in Turkey.