A Risk-Scoring System for Predicting Methicillin Resistance in Community-Onset Staphylococcus aureus Bacteremia in Korea

Abstract

Aims:

We aimed to develop a simple scoring system to predict risk for methicillin resistance in community-onset Staphylococcus aureus bacteremia (CO-SAB) by identifying the clinical and epidemiological risk factors for community-onset methicillin-resistant S. aureus (MRSA).

Methods:

We retrospectively analyzed data from three multicenter cohort studies in Korea in which patient information was prospectively collected and risk factors for methicillin resistance in CO-SAB were identified. We then developed and validated a risk-scoring system.

Results:

To analyze the 1,802 cases of CO-SAB, we included the four most powerful predictors of methicillin resistance that we identified in the scoring system: underlying hematologic disease (−1 point), endovascular infection as the primary site of infection (−1 point), history of hospitalization or surgery in ≤1 year (+0.5 points), and previous isolation of MRSA in ≤6 months (+1.5 points). With this scoring system, cases were classified into low (less than −0.5), intermediate (−0.5–1.5), and high (≥1.5) risk groups. The proportions of MRSA cases in each group were 24.7% (22/89), 39.0% (607/1,557), and 78.8% (123/156), respectively, and 16.7% (1/6), 33.8% (112/331), and 76.9% (10/13) in a validation set.

Conclusions:

This risk-scoring system for methicillin resistance in CO-SAB may help physicians select appropriate empirical antibiotics more quickly.

Introduction

Staphylococcus aureus imposes a significant burden on healthcare systems, and cases of methicillin-resistant S. aureus (MRSA) infection cause 64% more deaths than cases of methicillin-susceptible S. aureus (MSSA) infection.^1–3 In Korea, MRSA accounts for some 45% of community-onset (CO) invasive S. aureus infections.^4,5 However, when CO invasive staphylococcal infections are suspected, it is difficult for physicians to determine whether MRSA-targeted antibiotics should be empirically selected. Since it takes >3 days to get blood culture and antibiotic susceptibility test results, appropriate antibiotic treatment for patients with MRSA infection may be delayed, with serious consequences.^6–8 Although new technologies for identifying antimicrobial susceptibility such as matrix-assisted laser desorption ionization-time of flight mass spectrometry and Accelerate Pheno^™ systems have emerged,^9,10 these are unavailable in many hospitals due to limited resources or lack of strong evidence of their clinical implications. In contrast, overuse of MRSA-targeted antibiotics such as vancomycin can lead to antibiotic resistance and avoidable antibiotic-related complications such as acute kidney injury.¹¹ Therefore, it is important to predict methicillin resistance when physicians clinically suspect invasive S. aureus infections and choose an empiric antimicrobial agent.

CO-MRSA infections can be classified into healthcare-associated (HA) and community-associated (CA) infections according to the presence of risk factors for MRSA acquisition: history of hospitalization, surgery, dialysis, or residence in a long-term care facility within 1 year of the MRSA culture date; presence of an indwelling catheter or percutaneous device at the time of culture; and previous isolation of MRSA.^12,13 CA-MRSA has emerged in North America and Europe, and distinctive clinical factors of CA-MRSA have also been identified in previous studies.^14–16 There have been attempts to develop tools for predicting methicillin resistance in CO invasive S. aureus infections based on these risk factors, but the tools had low sensitivity and specificity.^17–21

CA-MRSA infections are less common in Korea than in North America, and HA-MRSA clones circulate more frequently in the community.⁵ Because of this difference in molecular epidemiology, the risk factors for methicillin resistance in CO S. aureus infections in Korea will not be same as those in North America. We, therefore, investigated the clinical and epidemiological factors associated with methicillin resistance in CO S. aureus bacteremia (CO-SAB), and developed and validated a simple clinical scoring system for predicting methicillin resistance.

Materials and Methods

Study design and patients

This retrospective cohort study used data from three multicenter cohorts of invasive S. aureus infections in Korea obtained using a similar study design: (1) a prospective cohort from July 2009 to June 2011 at 9 hospitals,⁴ (2) a prospective cohort from May to December 2012 at 16 hospitals,⁵ and (3) a prospective cohort from September 2013 to March 2015 at 10 hospitals.²² For validation, we collected additional patient data from April 2015 to March 2017, which was included as a last cohort. All studies were conducted with a standardized protocol and the institutions of the Korea INfectious Diseases (KIND) study group participated as much as possible. The individual authors and affiliations within this group are listed in the acknowledgments section.

CO-SAB was defined when S. aureus was isolated from blood cultures within 48 hr of hospitalization or in the emergency room.¹³ Infectious diseases specialists at each hospital checked the results of blood cultures routinely. When repeated blood cultures were positive in the same patient during the study period, only the first case was included. Patients who were discharged or died before the results of blood culture tests were available were excluded because we were unable to obtain the relevant information. Clinical and epidemiological information was collected by trained research nurses or infectious diseases specialists, and assessed for the following potential risk factors for MRSA infection: age, sex, underlying illness, primary site of infection, history of recent hospitalization or surgery, long-term care facility, indwelling device, dialysis, previous isolation of MRSA, acupuncture, and steroid or immunosuppressant use. To take into account comorbid conditions, Charlson's weighted index of comorbidity (WIC) was calculated.²³ The primary sites of SAB were established according to the best judgment of an infectious diseases specialist at each hospital as follows: catheter-related infection, pneumonia, skin and soft tissue infection (SSTI), bone and joint infection (BJI), endovascular infection (endocarditis, mycotic aneurysm, septic thrombophlebitis and infected implanted vascular device such as graft infection or arteriovenous fistula), intra-abdominal infection, urinary tract infection (UTI), central nervous system infection, other, and bacteremia of unknown primary focus. All the data were additionally reviewed by two infectious diseases specialists in Seoul National University Bundang Hospital.

This study was approved by the Institutional Review Board of Seoul National University Bundang Hospital.

Statistical analysis

Univariate analysis was performed to test for differences between groups. We used Student's t-test and the chi-square test or Fisher's exact test, depending on the type of variable. For multivariate analysis, potential covariates with p-values <0.10 in the univariate analyses were included.

Multivariate logistic regression models were tested with adjustment for covariates. Adjusted odds ratios (aORs) and their 95% confidence intervals (CIs) were calculated, and p-values <0.05 were considered statistically significant. Charlson's WIC was excluded from multivariate analyses because of multicollinearity with other underlying illnesses. The risk-scoring system was based on the beta coefficients in the logistic regression model. All statistical analyses were performed with IBM SPSS Statistics, version 22.0 (IBM Corp., Armonk, NY).

Results

A total of 1,802 cases of CO-SAB (716, 516, and 570 cases, respectively) were included, 752 (41.7%) of whom were MRSA infections. The mean age was 62 years, and 53.6% (968/1,802) of the patients were male. Bone and joints were the most common identifiable primary sites of infection. The mean Charlson's WICs of the three cohorts (see Study design and patients section) were 3.0 ± 2.7, 2.4 ± 2.3, 2.1 ± 2.2, and 4.6 ± 2.9, respectively, with relatively higher value of WIC in the third cohort. A total of 39.1% (691/1,766) of the patients had histories of hospitalization or surgery within the previous year, and 24.6% (425/1,727) of the patients had history of residence in a long-term care facility within the previous year.

The clinical and epidemiological factors associated with methicillin resistance in CO-SAB were identified by univariate analysis (Tables 1 and 2). Older age (≥65 years), presence of renal disease, solid tumor, cerebrovascular disease, chronic pulmonary diseases, peptic ulcer diseases as underlying illnesses, Charlson's WIC ≥3, catheter-related infection, and UTI were risk factors for methicillin resistance in univariate analysis (Table 1). In contrast, presence of hematologic disease as underlying illnesses, SSTIs, and BJIs were significantly more frequent among MSSA infections. The risk of MRSA infections was increased in patients with HA risk factors such as history of hospitalization, surgery, dialysis, or residence in a long-term care facility within the previous year; presence of an indwelling catheter or percutaneous device; and a history of isolation of MRSA within the previous 6 months (Table 2).

Table 1.

Clinical Factors of 1,802 Patients with Community-Onset Staphylococcus aureus Bacteremia

	MSSA (n = 1,050)	MRSA (n = 752)
	No. of cases (%)	No. of cases (%)	p
Age ≥65	512/1,050 (48.8)	465/752 (61.8)	<0.001
Male gender	564/1,050 (53.7)	404/752 (53.7)	0.997
Underlying illness
Diabetes mellitus^a	348/1,047 (33.2)	269/751 (35.8)	0.256
Renal disease^a	179/1,043 (17.2)	180/751 (24.0)	<0.001
Solid tumor	165/1,050 (15.7)	156/752 (20.7)	0.006
Cerebrovascular disease^a	110/1,041 (10.6)	128/746 (17.2)	<0.001
Cardiovascular disease	98/1,050 (9.3)	76/752 (10.1)	0.584
Liver disease^a	88/1,038 (8.5)	56/743 (7.5)	0.473
Chronic pulmonary disease^a	42/1,042 (4.0)	55/747 (7.4)	0.002
Peptic ulcer disease^a	31/1,042 (3.0)	46/746 (6.2)	0.001
Dementia^a	40/1,043 (3.8)	41/747 (5.5)	0.097
Connective tissue disease^a	26/1,042 (2.5)	20/748 (2.7)	0.814
Hematologic disease	42/1,050 (4.0)	16/752 (2.1)	0.026
AIDS^a	1/1,045 (0.1)	3/747 (0.4)	0.314
Charlson's WIC ≥3^a	464/1,035 (44.8)	385/734 (52.5)	0.002
Primary site of infection
BJI	205/1,050 (19.5)	114/752 (15.2)	0.017
Catheter-related infection	77/1,050 (7.3)	110/752 (14.6)	<0.001
Pneumonia	113/1,050 (10.8)	108/752 (14.4)	0.22
SSTI	171/1,050 (16.3)	84/752 (11.2)	0.002
IAI	41/1,050 (3.9)	36/752 (4.8)	0.361
Endovascular infection	65/1,050 (6.2)	32/752 (4.3)	0.073
UTI	26/1,050 (2.5)	32/752 (4.3)	0.035
CNS infection	3/1,050 (0.3)	5/752 (0.7)	0.29
Others^b	24/1,050 (2.3)	22/752 (2.9)	0.396
Unknown	325/1,050 (31.0)	209/752 (27.8)	0.147

There are some missing data.

Others include ocular infection, mediastinal infection, and ostitis.

AIDS, acquired immune deficiency syndrome; BJI, bone and joint infection; CNS, central nervous system; IAI, intra-abdominal infection; MRSA, methicillin-resistant S. aureus; MSSA, methicillin-susceptible S. aureus; SSI, surgical site infection; SSTI, skin and soft tissue infection; UTI, urinary tract infection; WIC, weighted index of comorbidity.

Table 2.

Epidemiological Characteristics of 1,802 Patients with Community-Onset Staphylococcus aureus Bacteremia

	MSSA (n = 1,050)	MRSA (n = 752)
	No. of cases (%)	No. of cases (%)	p
Hospitalization or surgery^{a b}	317/1,024 (31.0)	374/742 (50.4)	<0.001
Long-term care facility^{a b}	185/1,006 (18.4)	240/721 (33.3)	<0.001
Indwelling device^b	198/1,032 (19.2)	240/737 (32.6)	<0.001
Dialysis^{a b}	125/1,038 (12.0)	149/744 (20.0)	<0.001
Previous isolation of MRSA^{b c}	41/945 (4.3)	129/680 (19.0)	<0.001
Acupuncture <1 month^b	157/961 (16.3)	101/678 (14.9)	0.430
Steroid use^b	55/768 (7.2)	37/521 (7.1)	0.967
Immunosuppressant^b	54/1,045 (5.2)	30/748 (4.0)	0.253

Within past 1 year.

Some data were missing.

Within 6 months.

In a multivariate regression model, the risk factors for MRSA infection in CO-SAB were older age ((aOR, 1.70; 95% CI, 1.41–2.06), history of hospitalization or surgery (aOR 2.05; 95% CI 1.64–2.56), residence in a long-term care facility (aOR 1.68; 95% CI 1.30–2.17), and dialysis (aOR 1.48; 95% CI 1.08–2.02) within ≤1 year, and history of previous isolation of MRSA (aOR 4.46; 95% CI 2.97–6.68) within ≤6 months. In contrast, MRSA bacteremia patients had less hematologic disease (aOR 0.45; 95% CI 0.23–0.85), and their primary infections were less frequently SSTIs (aOR 0.57; 95% CI 0.41–0.79), BJIs (aOR 0.65; 95% CI 0.48–0.88), and endovascular infections (aOR 0.46; 95% CI 0.26–0.79).

To develop the scoring system, we chose the four most powerful predictors and allocated them points: underlying hematologic disease (−1 point), endovascular infection as primary infection (−1 point), hospitalization or surgery within ≤1 year (+0.5 points), and history of previous isolation of MRSA within ≤6 months (+1.5 points) (Table 3). We classified the risk of MRSA bacteremia according to the total score, which ranged from −2 to +2 points, into low (score < −0.5), intermediate (−0.5 ≤ score <1.5), and high (score ≥1.5). The proportions of MRSA in each group were 24.7% (22/89), 39.0% (607/1,557), and 78.8% (123/156), respectively (Fig. 1).

FIG. 1.

The proportion of MRSA among community-onset Staphylococcus aureus bacteremia patients in each risk group. *The score ranged from −2 to +2 points. MRSA, methicillin-resistant S. aureus.

Table 3.

The Methicillin-Resistant Staphylococcus aureus -Predictive Scoring System for Community-Onset Staphylococcus aureus Bacteremia

	Beta	aOR (95% CI)	p	Points
Age ≥65 years	0.456	1.577 (1.263–1.970)	<0.001
Underlying illness
Hematologic disease	−0.809	0.445 (0.234–0.847)	0.014	√	−1.0
Primary site of infection
BJI	−0.431	0.650 (0.481–0.878)	0.005
SSTI	−0.562	0.570 (0.410–0.793)	0.001
Endovascular	−0.786	0.456 (0.263–0.790)	0.005	√	−1.0
Epidemiological characteristics
Hospitalization or surgery^a	0.718	2.050 (1.640–2.563)	<0.001	√	+0.5
Long-term facility^a	0.518	1.679 (1.297–2.173)	<0.001
Dialysis^a	0.389	1.475 (1.075–2.024)	0.016
Previous isolation of MRSA^b	1.494	4.456 (2.974–6.677)	<0.001	√	+1.5

Within past 1 year.

Within 6 months.

√, factors included in the scoring system; aOR, adjusted odds ratio; CI, confidence interval.

In addition, a total of 350 patients were enrolled in the validation set, 35.1% (123/350) of which were MRSA infections. In this set, the proportions of MRSA in each group were 16.7% (1/6), 33.8% (112/331), and 76.9% (10/13), respectively (Fig. 1).

Discussion

Of the 1,802 patients with CO-SAB, 752 (41.7%) had MRSA infections. In the three cohorts of CO-SAB cases enrolled at different times between 2009 and 2015, the proportions of CO-MRSA infections were 42.3% (303/716) in the cohort from 2009 to 2011, 45.3% (234/516) in the cohort in 2012, and 37.7% (215/570) in the cohort from 2013 to 2015. The proportions of CA-MRSA infections among the CO-SABs were 11.3% (81/716), 10.1% (52/516), and 8.8% (50/570), respectively, with no significant difference between them. These proportions were higher than in our survey of CA-MRSA infections in 2005,²⁴ and there are other studies that suggest that CA-MRSA infections are increasing in Korea.^5,24–26 However, the proportion of CA-MRSA infections has not been increasing rapidly according to these large scale studies.

Currently, we do not recommend routine use of MRSA-targeting antibiotics as initial empirical therapy for patients with CO infections in Korea.²⁷ However, there are no local practical guidelines concerning the possibility of methicillin resistance among CO-SABs. In multilocus sequence typing analysis of CA-MRSA, the most common sequence type in Korea was ST72 (USA700), but ST5 (USA100) and ST239, which are generally found among HA MRSAs, were also circulating widely in the community.^24,28 This suggested that the boundary between community and healthcare settings is sometimes unclear. Hence, by analyzing a large number of prospectively and systematically collected multicenter CO-SAB cases, we hoped to develop a simple clinical and epidemiological risk factor-based system for predicting methicillin resistance in cases of CO-SAB in Korea.

In this study, we identified several risk factors for methicillin resistance that were different from those identified in studies in other countries. In our case, the risk of MRSA was higher in the elderly, whereas the corresponding results in previous studies were inconsistent: some reported results similar to this study,²⁹ others suggested that the risk was not related to age^14,30,31 or even that lower age was associated with increased risk of MRSA.^18,19 These disparities may reflect cultural differences. For example, there is a unique caregiver culture involving family members in Korea, and it is common for the elderly to be main caregivers. Hematologic diseases were also more likely to be non-MRSA in another study, but this effect was not statistically significant.¹⁸ In addition, SSTIs, BJIs, and endovascular infections were more likely to be due to MSSA in our study, whereas other studies found that these factors had less impact or were unrelated to MSSA infections.^19,30,31 In particular, SSTI was found to be associated with CO-MRSA in North America.³² Because of these diverse results, the relationships between specific underlying illnesses or primary sites of infection and methicillin resistance observed in this study should not be generalized, and further studies are needed.

We classified the risk of MRSA in the CO-SABs as low, intermediate, or high using a simple scoring system. As a result, in both the derivation set and the validation set, the frequency of MRSA infections was <25% in the low-risk group and >75% in the high-risk group. This scoring system could be helpful for determining the initial choice of MRSA-targeted antibiotic when SAB is suspected in the early course of hospitalization of patients. If there is a high score (≥1.5 points) in this scoring system, administration of empirical MRSA-targeted antibiotics may be considered, whereas they can be deferred in cases with low probabilities of MRSA infection. Although empirical MRSA-targeted antibiotics can be permitted in severe cases, this policy should allow an adequate choice of empirical antibiotic in CO-SAB cases, in terms of both appropriateness of antibiotic spectrum and avoidance of unnecessary antibiotic use.

This study has several limitations. First, the data were analyzed retrospectively; they were collected from three different studies, and there might be some heterogeneity including comorbidities among the patients participating in those studies. Although each study was prospective with a similar design, and ∼40% of the hospitals contributed to at least two of the cohorts, this heterogeneity could bias the results. Second, since we used three cohorts for 8 years, epidemiological changes may have affected our results. Third, since we targeted SAB, the results could be different from those obtained for invasive S. aureus infections. SSTIs, especially, are difficult to grow in blood culture, so caution is needed in interpreting the results. In addition, not all patients were evaluated for endovascular infections using transthoracic or transesophageal echocardiography, so endovascular infections could be underestimated. Fourth, although we tried to construct an effective system for analyzing risk, substantial number of patients fell into the intermediate-risk group. Hence, additional early detection techniques such as polymerase chain reaction, which can detect antibiotic susceptibility, are needed. Fifth, since only Korean hospitals participated in this study, the results may not be applicable to other countries with different clinical and epidemiological situation. Finally, there were some missing values for several factors, which reduced statistical significance. However, we included a large number of cases, and most of the factors with large number of missing values did not have statistically significant effects on risk in the univariate analysis and were not included in the multivariate analysis.

In conclusion, we have developed a simple clinical risk-scoring system to predict methicillin resistance in CO-SAB, and have stratified cases into low-, intermediate-, and high-risk groups. This stratification according to a scoring system based on clinical and epidemiological risk factors may help physicians identify appropriate empirical antibiotics early in the course of CO-SAB.

Footnotes

Acknowledgments

We thank the members of the KIND study group and associated staff for their cooperation in this study. The collaborators in the KIND Study Group were Taek Soo Kim, Sue Shin, Kyoung Un Park, and Eui-Chong Kim, Department of Laboratory Medicine, Seoul National University College of Medicine, Seoul, Republic of Korea; Shinhye Cheon, Nak-Hyun Kim, and Wan Beom Park, Department of Internal Medicine, Seoul National University Hospital, Seoul, Republic of Korea; Sun Hee Lee, Department of Internal Medicine, Pusan National University Hospital, Busan, Republic of Korea; Ki Tae Kwon, Department of Internal Medicine, Daegu Fatima Hospital, Daegu, Republic of Korea; Chisook Moon, Department of Internal Medicine, Inje University College of Medicine, Busan, Republic of Korea; Ji Hwan Bang and Sang Won Park, Division of Infectious Diseases, Seoul Metropolitan Government—Seoul National University Boramae Medical Center, Seoul, Republic of Korea; Hee Jung Choi, Department of Internal Medicine, Ewha Womans University, School of Medicine, Seoul, Republic of Korea; Eun Jung Lee, Department of Internal Medicine, Soonchunhyang University Seoul Hospital, Seoul, Republic of Korea; Eun Ju Choo, Department of Internal Medicine, Soonchunhyang University Bucheon Hospital, Bucheon, Republic of Korea; Yeong Keun Kim, Department of Internal Medicine, Yonsei University Wonju College of Medicine, Wonju, Republic of Korea; Min Hyok Jeon, Department of Internal Medicine, Soonchunhyang University Cheonan Hospital, Cheonan, Republic of Korea; Jeong-Hwan Hwang and Chang-Seop Lee, Department of Internal Medicine, Chonbuk National University Medical School, Jeonju, Republic of Korea; and Hee-Chang Jang, Department of Internal Medicine, Chonnam National University Medical School, Gwangju, Republic Korea.

This study was performed using data from three multicenter cohorts. The first cohort from July 2009 to June 2011 was supported by grant No. 2009-E00609-00 from the Korea Centers for Disease Control and Prevention. The second cohort from May to December 2012 was supported by grant No. 2012-E44003-00 from the Korea Centers for Disease Control and Prevention. The last cohort from September 2013 to March 2015 was the prospective cohort study, “Korea INfectious Diseases Study Group/Staphylococcus aureus Bacteremia 2013 (KIND-SAB 2013) cohort” with project name, “Establishment of Network for Clinical Research of Staphylococcus aureus Infection”. The KIND-SAB 2013 cohort was supported by Grant No. HI10C2020 from the National Strategic Coordinating Center for Clinical Research, which is run by the Ministry of Health and Welfare, Government of Korea. We have registered the protocol of the last KIND-SAB 2013 cohort study (KCT0001070) and the present comparative study (KCT0001088) with the Clinical Research Information Service of the Republic of Korea () in cooperation with the WHO International Clinical Trials Registry Platform.

Disclosure Statement

No competing financial interests exist.

References

Cosgrove

S.E.

, Sakoulas

, Perencevich

E.N.

, Schwaber

M.J.

, Karchmer

A.W.

, and Carmeli

Y..

2003. Comparison of mortality associated with methicillin-resistant and methicillin-susceptible Staphylococcus aureus bacteremia: a meta-analysis. Clin. Infect. Dis., 36:53–59.

Pastagia

, Kleinman

L.C.

, Lacerda de la Cruz

E.G.

, and Jenkins

S.G..

2012. Predicting risk for death from MRSA bacteremia. Emerg. Infect. Dis., 18:1072–1080.

World Health Organization. 2014. WHO's first global report on antibiotic resistance reveals serious, worldwide threat to public health. Antimicrobial Resistance–Global Surveillance Report Virtual Press Conference (Vol. 30). Geneva, Switzerland.

Song

K.H.

, Kim

E.S.

, Sin

H.Y.

, Park

K.H.

, Jung

S.I.

, Yoon

, Kim

D.M.

, Lee

C.S.

, Jang

H.C.

, and Park

Y..

2013. Characteristics of invasive Staphylococcus aureus infections in three regions of Korea, 2009–2011: a multi-center cohort study. BMC Infect. Dis., 13:1.

Kim

E.S.

, Kim

H.B.

, Kim

K.H.

, Park

K.H.

, Lee

, Choi

Y.H.

, Yi

, Kim

C.J.

, and Song

K.H..

2014. Clinical and epidemiological factors associated with methicillin resistance in community-onset invasive Staphylococcus aureus infections: prospective multicenter cross-sectional study in Korea. PLoS One, 9:e114127.

Khatib

, Saeed

, Sharma

, Riederer

, Fakih

M.G.

, and Johnson

L.B..

2006. Impact of initial antibiotic choice and delayed appropriate treatment on the outcome of Staphylococcus aureus bacteremia. Eur. J. Clin. Microbiol. Infect. Dis., 25:181–185.

Gómez

, García-Vázquez

, Baños

, Canteras

, Ruiz

, Baños

, Herrero

J.A.

, and Valdés

M..

2007. Predictors of mortality in patients with methicillin-resistant Staphylococcus aureus (MRSA) bacteraemia: the role of empiric antibiotic therapy. Eur. J. Clin. Microbiol. Infect. Dis., 26:239–245.

Kim

S.H.

, Park

W.B.

, Lee

K.D.

, Kang

C.I.

, Bang

J.W.

, Kim

H.B.

, Kim

E.C.

, Oh

M.D.

, and Choe

K.W..

2004. Outcome of inappropriate initial antimicrobial treatment in patients with methicillin-resistant Staphylococcus aureus bacteraemia. J. Antimicrob. Chemother., 54:489–497.

Marschal

, Bachmaier

, Autenrieth

, Oberhettinger

, Willmann

, and Peter

S..

2017. Evaluation of the Accelerate Pheno™ system for fast identification and antimicrobial susceptibility testing from positive blood culture in gram-negative bloodstream infection. J. Clin. Microbiol., ;55:2116–2126.

10.

Singhal

, Kumar

, Kanaujia

P.K.

, and Virdi

J.S..

2015. MALDI-TOF mass spectrometry: an emerging technology for microbial identification and diagnosis. Front. Microbiol., 6:791.

11.

Minejima

, Choi

, Beringer

, Lou

, Tse

, and Wong-Beringer

A..

2011. Applying new diagnostic criteria for acute kidney injury to facilitate early identification of nephrotoxicity in vancomycin-treated patients. Antimicrob. Agents Chemother., 55:3278–3283.

12.

David

M.Z.

, Glikman

, Crawford

S.E.

, Peng

, King

K.J.

, Hostetler

M.A.

, Boyle-Vavra

, and Daum

R.S..

2008. What is community-associated methicillin-resistant Staphylococcus aureus? J. Infect. Dis., 197:1235–1243.

13.

Klevens

R.M.

, Morrison

M.A.

, Fridkin

S.K.

, Reingold

, Petit

, Gershman

, Ray

, Harrison

L.H.

, Lynfield

, and Dumyati

G..

2006. Community-associated methicillin-resistant Staphylococcus aureus and healthcare risk factors. Emerg. Infect. Dis., 12:1991–1993.

14.

Böcher

, Gervelmeyer

, Monnet

, Mølbak

, and Skov

R..

2008. Methicillin-resistant Staphylococcus aureus: risk factors associated with community-onset infections in Denmark. Clin. Microbiol. Infect., 14:942–948.

15.

Salgado

C.D.

, Farr

B.M.

, and Calfee

D.P..

2003. Community-acquired methicillin-resistant Staphylococcus aureus: a meta-analysis of prevalence and risk factors. Clin. Infect. Dis., 36:131–139.

16.

Warshawsky

, Hussain

, Gregson

D.B.

, Alder

, Austin

, Bruckschwaiger

, Chagla

A.H.

, Daley

, Duhaime

, and McGhie

K..

2000. Hospital-and community-based surveillance of methicillin-resistant Staphylococcus aureus: previous hospitalization is the major risk factor. Infect. Control Hosp. Epidemiol., 21:724–727.

17.

Butler-Laporte

, Cheng

M.P.

, Cheng

A.P.

, McDonald

E.G.

, and Lee

T.C..

2016. Using MRSA screening tests to predict methicillin resistance in Staphylococcus aureus bacteremia. Antimicrob. Agents Chemother., 60:7444–7448.

18.

Zilberberg

M.D.

, Chaudhari

, Nathanson

B.H.

, Campbell

R.S.

, Emons

M.F.

, Fiske

, Hays

H.D.

, and Shorr

A.F..

2012. Development and validation of a bedside risk score for MRSA among patients hospitalized with complicated skin and skin structure infections. BMC Infect. Dis., 12:1.

19.

Miller

L.G.

, Remington

F.P.

, Bayer

A.S.

, Diep

, Tan

, Bharadwa

, Tsui

, Perlroth

, Shay

, and Tagudar

G..

2007. Clinical and epidemiologic characteristics cannot distinguish community-associated methicillin-resistant Staphylococcus aureus infection from methicillin-susceptible S. aureus infection: a prospective investigation. Clin. Infect. Dis., 44:471–482.

20.

Lesens

, Methlin

, Hansmann

, Remy

, Martinot

, Bergin

, Meyer

, and Christmann

D..

2003. Role of comorbidity in mortality related to Staphylococcus aureus bacteremia: a prospective study using the Charlson weighted index of comorbidity. Infect. Control Hosp. Epidemiol., 24:890–896.

21.

Chen

S.Y.

, Chiang

W.C.

, Ma

M.M.

, Hsueh

P.R.

, Chang

S.C.

, Fang

C.C.

, Chen

S.C.

, Chen

W.J.

, Chie

W.C.

, and Lai

M.S..

2012. Predicting methicillin resistance among community-onset Staphylococcus aureus bacteremia patients with prior healthcare-associated exposure. Eur. J. Clin. Microbiol. Infect. Dis., 31:2727–2736.

22.

Song

K.H.

, Jung

S.I.

, Lee

, Park

, Kiem

, Lee

, Kwak

, Kim

, Jang

H.C.

, and Kim

Y.S..

2017. Characteristics of cefazolin inoculum effect-positive methicillin-susceptible Staphylococcus aureus infection in a multicentre bacteraemia cohort. Eur. J. Clin. Microbiol. Infect. Dis., 36:285–294.

23.

Charlson

M.E.

, Pompei

, Ales

K.L.

, and MacKenzie

C.R..

1987. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J. Chronic Dis., 40:373–383.

24.

Kim

E.S.

, Song

J.S.

, Lee

H.J.

, Choe

P.G.

, Park

K.H.

, Cho

J.H.

, Park

W.B.

, Kim

S.H.

, Bang

J.H.

, and Kim

D.M..

2007. A survey of community-associated methicillin-resistant Staphylococcus aureus in Korea. J. Antimicrob. Chemother., 60:1108–1114.

25.

Song

J.H.

, Hsueh

P.R.

, Chung

D.R.

, Ko

K.S.

, Kang

C.I.

, Peck

K.R.

, Yeom

J.S.

, Kim

S.W.

, Chang

H.H.

, and Kim

Y.-S..

2011. Spread of methicillin-resistant Staphylococcus aureus between the community and the hospitals in asian countries: an ANSORP study. J. Antimicrob. Chemother., 66:1061–1069.

26.

Chen

C.J.

, and Huang

Y.C..

2014. New epidemiology of Staphylococcus aureus infection in Asia. Clin. Microbiol. Infect., 20:605–623.

27.

Kim

H.B.

2007. Community-associated methicillin-resistant Staphylococcus aureus (CA-MRSA). Korean J. Med., 72:120–130.

28.

Kim

E.S.

, Lee

H.J.

, Chung

G.T.

, Lee

Y.S.

, Shin

D.H.

, Jung

S.I.

, Song

K.H.

, Park

W.B.

, Kim

N.J.

, and Park

K.U..

2011. Molecular characterization of methicillin-resistant Staphylococcus aureus isolates in Korea. J. Clin. Microbiol., 49:1979–1982.

29.

Roghmann

M.-C.

2000. Predicting methicillin resistance and the effect of inadequate empiric therapy on survival in patients with Staphylococcus aureus bacteremia. Arch. Intern. Med., 160:1001–1004.

30.

Lee

J.Y.

, Chong

Y.P.

, Kim

, Hong

H.L.

, Park

S.J.

, Lee

E.S.

, Kim

M.N.

, Kim

S.H.

, Lee

S.O.

, and Choi

S.H..

2014. Bone and joint infection as a predictor of community-acquired methicillin-resistant Staphylococcus aureus bacteraemia: a comparative cohort study. J. Antimicrob. Chemother., 69:1966–1971.

31.

Wang

J.L.

, Chen

S.Y.

, Wang

J.T.

, Wu

G.H.M.

, Chiang

W.C.

, Hsueh

P.R.

, Chen

Y.C.

, and Chang

S.C..

2008. Comparison of both clinical features and mortality risk associated with bacteremia due to community-acquired methicillin-resistant Staphylococcus aureus and methicillin-susceptible S. aureus. Clin. Infect. Dis., 46:799–806.

32.

Skiest

D.J.

, Brown

, Cooper

T.W.

, Hoffman-Roberts

, Mussa

H.R.

, and Elliott

A.C..

2007. Prospective comparison of methicillin-susceptible and methicillin-resistant community-associated Staphylococcus aureus infections in hospitalized patients. J. Infect., 54:427–434.