Predicting Factors of Methicillin-resistant Staphylococcus aureus Bacteremia among Hospitalized Patients in a Tertiary-care Hospital : A Case-control Study
T. Chuamuangphan,P. Chongtrakool,S. Sungkanuparph
Abstract:Background: Methicillin-resistant Staphylococcus aureus (MRSA) has become a major nosocomial pathogen in community hospitals, long-term-care facilities, and tertiary care hospitals. The data regarding predicting factors of MRSA in Thailand is limited. We aimed to determine the rate and predicting factors of MRSA bacteremia among hospitalized patients with S. aureus in Ramathibodi Hospital where there is a tertiarycare setting. Methods: A case-control study was conducted among adult patients with S. aureus bacteremia who were admitted at the Department of Medicine, Ramathibodi Hospital between January 2008 and December 2012. Clinical characteristics and culture results of S. aureus isolates and antimicrobial-susceptibility testing were studied. MRSA was defined as S. aureus isolates that were resistant to methicillin. The factors associated with MRSA bacteremia was assessed by multivariate logistic regression analysis. Results: The prevalence of MRSA bacteremia among patients with S. aureus bacteremia during the study period was 42.3% (267 of 632 patients). A total of 156 patients with S. aureus bacteremia were studied. Fifty-two patients had MRSA bacteremia and 104 patients had methicillin-susceptible S. aureus (MSSA) bacteremia. The mean age of patients was 63.5 years and 56.4% of patients were males. There were no differences of demographics and underlying diseases between the two groups (p > 0.05). Length of hospital stay and period from admission until positive blood cultures result for S. aureus were significantly longer among MRSA bacteremia patients (p < 0.05). In multivariate logistic regression analysis, hospitalization within the past 6 months (Odds ratio (OR) 3.514; 95% confidence interval (CI), 1.319-9.363, p = 0.012) and duration of admission prior to S. aureus bacteremia (OR 1.044; 95% CI, 1.015-1.075, p = 0.003) were significantly associated with MRSA bacteremia. MRSA susceptibilities were as follows: 98.1% to fusidic acid, 71.2% to trimethoprim-sulfamethoxazole and to vancomycin. Of all patients with MSSA, 43.3% were treated 80 J INFECT DIS ANTIMICROB AGENTS May-August 2014 with cloxacillin. Of all patients with MRSA, 82.7% were treated with vancomycin. Of all, 61.4% of patients were discharged with improvement. The overall mortality rate was 17.3% and was significantly higher in patients with MRSA bacteremia (26.9%) when compared with patients with MSSA bacteremia (12.5%) (p < 0.05). Conclusion: The prevalence of MRSA among patients with S. aureus bacteremia at a tertiary-care hospital is high. Prior hospitalization and prolonged duration of admission were significant predicting factors for MRSA bacteremia. (J Infect Dis Antimicrob Agents 2014;31:79-89.) INTRODUCTION Methicillin-resistant Staphylococcus aureus (MRSA) has been recognized as a common cause of both hospital-acquired infections worldwide and recently reported as community-acquired infection in Western countries.1 In addition, it has emerged as an important pathogen among hospitalized patients. According to recent estimates from the National Nosocomial Infection Surveillance System of the Centers for Disease Control and Prevention (CDC), the rate of MRSA causing nosocomial infections in intensive care unit patients has approached 50% in the United States health care facilities.2 Healthcareassociated (HA)-MRSA is associated with severe, invasive disease in hospitalized patients.3-7 HAMRSA strains tend to have multidrug resistance and carry staphylococcal cassette chromosome type II (SCCmec type II).8 They are typified by a USA100 or USA200 pulsed field gel electrophoresis (PFGE) pattern.9 The prevalence of antimicrobial resistance in hospital-acquired S. aureus infections has been progressively increasing.3,8,10 In a surveillance report of over 24,000 cases of nosocomial S. aureus bacteremia in the United States, isolates with methicillin resistance increased from 22% to 57% between 1995 and 2001.3 Worldwide, HA-MRSA prevalence varies considerably, from < 1 percent in Scandinavia to up to 40% in Japan, Israel, and elsewhere in Europe.11,12 In Thailand, a total of 1,783 (25%) isolations of MRSA were reported from 23 hospitals to the National Institute of Health, Thailand in 1998. The increased proportion was much more evenly distributed around the country than before. Between 1988 and 1998 the MRSA proportion rose from 11% to 23.4% in the northeast region, from 16% to 30.5% in the central region, and from 21% to 30.3% in the southern region.13,14 The manifestations of HA-MRSA consisted of skin and soft tissue infections, including wound infection, bacteremia, and pneumonia.15 Patients with MRSA infection have higher mortality, longer hospital stays, and higher health care costs than patients with methicillin-susceptible S. aureus (MSSA) infection.6,16-20 In a retrospective analysis of 85 cases with S. aureus bacteremia in critically ill patients (47 with MRSA and 38 with MSSA), patients with MRSA bacteremia had a higher 30-day mortality rate than those with MSSA (53% versus 18%).17 They also had higher rates of acute renal failure, hemodynamic instability, and prolonged ventilator dependency. Although there is a substantial amount of data regarding prevalence and bacteriologic characteristic of MRSA elsewhere, it is limited in Thailand. Therefore, we herein evaluated the rate and characteristics of MRSA bacteremia among hospitalized patients in Ramathibodi Hospital where there is a tertiary-care setting. Vol. 31 No. 2 81 Methicillin-resistant Staphylococcus aureus bacteremia:Chuamuangphan T, et al. 81 The primary objective of this study was to determine the rate of MRSA among hospitalized patients with S. aureus bacteremia in Ramathibodi Hospital. The secondary objectives were to identify the risk factors for MRSA among hospitalized patients with S. aureus bacteremia in Ramathibodi Hospital and describe the characteristics of MRSA infection susceptibility patterns of S. aureus among hospitalized patients with S. aureus bacteremia in Ramathibodi Hospital. MATERIALS AND METHODS Case Control study was conducted among adult patients with S. aureus bacteremia who were admitted at the Department of Medicine, Ramathibodi Hospital between January 2008 and December 2012. Patients were categorized into MSSA and MRSA bacteremia and were randomized from the hospital database at a ratio of 2:1. Data was extracted from medical records and database of laboratory results. Information obtained included demographic characteristics, underlying disease, clinical presentation, potential risk factors of MRSA bacteremia (i.e. long term dialysis, permanent indwelling catheter/medical device, previous surgery in past 6 months, resident in long-term care facility, taken any antibiotic in past 6 months, hospitalized in past 6 months, intravenous drug use), diagnosis, date of admission, date of discharge, source of specimen, susceptibility, treatment, and outcome. Specimens were obtained from blood with other sources such as wound, sputum, urine and central catheter and were processed and cultured at the hospital laboratories according to standard techniques. Clinical characteristics and culture results of S. aureus isolates and antimicrobialsusceptibility test were studied. The susceptibility of S. aureus was determined using CLSI protocol. Categorical data is presented as frequencies and percentages. Continuous data is presented as means and standard deviations (SD), or medians and ranges, for data with and without normal distribution, respectively. The prevalence of MRSA among hospitalized patients with S. aureus bacteremia in Ramathibodi Hospital was determined. Patients with S. aureus bacteremia were categorized into two groups, MRSA and MSSA. Baseline characteristics were compared using Chi-Square or Fisher’s exact test for categorical variables and Student T-test or Mann-Whitney U test for continuous variables. The association of factors with MRSA bacteremia was assessed by multivariate logistic regression analysis. Data was collected and analyzed using SPSS (IBM SPSS statistics 21). A p-value of < 0.05 was considered statistically significant. The study had been approved by the Research Ethic Committee of the Faculty of Medicine, Ramathibodi Hospital, Mahidol University. RESULTS During the study period, 632 patients had positive blood culture for S. aureus. Of these, 267 (42.2%) had MRSA bacteremia and 365 (57.8%) had MSSA bacteremia. A total of 156 patients with S. aureus bacteremia that had available medical record and susceptibility data were studied. Fiftytwo (33.3%) patients had MRSA bacteremia and 104 (66.7%) patients had MSSA bacteremia. Of all, the mean age (SD) was 63.5 (16.6) and 56.4% of patients were males. Table 1 shows the baseline characteristics of the patients with MRSA and MSSA bacteremia. There were no differences on demographics or underlying disease, between the two groups. Having taken any antibiotic in past 6 months, hospitalized in past 6 months and previous surgery in past 6 months were significantly 82 J INFECT DIS ANTIMICROB AGENTS May-August 2014 Table 1. Baseline characteristics of 156 study patients. *hemodialysis or chronic ambulatory peritoneal dialysis s c i t s i r e t c a r a h C l l A ) 6 5 1 = n ( A S R M ) 2 5 = n ( A S S M ) 4 0 1 = n ( e u l a v P D S n a e m , s r a e y , e g A 6 . 6 1 5 . 3 6 7 . 5 1 9 . 4 6 1 . 7 1 8 . 2 6 9 7 4 . 0 ) % ( r e b m u n , r e d n e G 0 1 6 . 0 e l a M ) 4 . 6 5 ( 8 8 ) 6 . 9 5 ( 1 3 ) 8 . 4 5 ( 7 5 e l a m e F ) 6 . 3 4 ( 8 6 ) 4 . 0 4 ( 1 2 ) 2 . 5 4 ( 7 4 ) % ( r e b m u n , e c n i v o r P 1 3 7 . 0 k o k g n a B ) 3 . 8 5 ( 1 9 ) 8 . 5 5 ( 9 2 ) 6 . 9 5 ( 2 6 s r e h t O ) 7 . 1 4 ( 5 6 ) 2 . 4 4 ( 3 2 ) 4 . 0 4 ( 2 4 ) % ( r e b m u n , s e s a e s i d g n i y l r e d n U s u t i l l e m s e t e b a i D ) 4 . 1 3 ( 9 4 ) 8 . 8 2 ( 5 1 ) 7 . 2 3 ( 4 3 6 1 7 . 0 n o i t c e f n i V I H ) 6 . 2 ( 4 ) 8 . 3 ( 2 ) 9 . 1 ( 2 1 0 6 . 0 s e s a e s i d y e n d i k c i n o r h C ) 1