Published online Dec 04, 2020.
https://doi.org/10.3947/ic.2020.52.4.503
Clinical Characteristics and Causative Pathogens of Infective Arthritis and Risk Factors for Gram-Negative Bacterial Infections
,1
Yun Suk Cho,1
Yu Jin Sohn,1
Jong Hoon Hyun,1
Sang Min Ahn,1
Woon Ji Lee,1,2
Jung Ho Kim,1,2
Hye Seong,1,2
Junhyoung Kim,1,2
Su Jin Jeong,1,2
Nam Su Ku,1,2
Joon Sup Yeom,1,2
Jin Young Ahn,1,2
and Jun Yong Choi1,2
Abstract
Background
The aim of this study was to describe the clinical and microbiological characteristics of infective arthritis and to analyze risk factors for Gram-negative bacterial infections that cause infective arthritis.
Materials and Methods
Patients admitted between 2009 - 2018 with infective arthritis in a single-tertiary hospital were evaluated retrospectively.
Results
A total of 181 patients were enrolled in this study. Of them, 135 were native joint infection patients and 46 were prosthetic joint infection patients. The most common site of infective arthritis was the knee (63.6%), followed by the shoulder (17.7%), and the hip (9.9%). The most frequently identified microorganisms were Staphylococcus aureus (51.1%), followed by Streptococci sp. (21.1%), Enterobacteriaceae (8.4%), and coagulase-negative-Staphylococci (CNS; 8.4%). Infections due to Gram-negative bacteria and fungi made up 13.7% and 3.2% of all cases, respectively. Additionally, 20% and 4.2% of the cases involved methicillin-resistant S. aureus (MRSA) and MRCNS. We found that bacteriuria, infective arthritis in the hip, and steroid use at admission are independent risk factors for Gram-negative bacterial infections.
Conclusion
Infective arthritis with methicillin-resistant microorganisms reached up to about 25% in a single-tertiary hospital in Korea. In case of suspected urinary tract infection, infective arthritis of the hip joint, or steroid use at admission time among infective arthritis patients, empirical treatment covering Gram-negative microorganisms can be considered.
Introduction
Infective arthritis can rapidly destroy articular cartilage, occasionally resulting in permanent joint disabilities. Therefore, appropriate and prompt use of antibiotics is important for minimizing its sequelae. Owing to the fact that Staphylococci and Streptococci were the most common causative agents of infective arthritis, empirical antibiotic spectrum usually include Staphylococci and Streptococci [1, 2, 3, 4]. A study in Korea also reported that 20.5% (n = 25) of patients with culture-positive infective arthritis were due to Staphylococcus aureus and 1.6% (n = 2) were due to Streptococci [5]. However, an increased frequency of methicillin-resistant S. aureus (MRSA) has been reported in various parts of the world [6, 7, 8, 9, 10]. According to a 2017 report from the Korean GlobaL Antimicrobial Resistance Surveillance System (Kor-GLASS), 53% of S. aureus infections were due to MRSA [11]. This implies that, in Korea, the antibiotic susceptibility profiles of microorganisms causing infective arthritis can change. In order to develop appropriate strategies for treating infective arthritis, data on the distributions of causative pathogens and the prevalence of antimicrobial resistance are required. However, data on the pathogen distributions and resistance patterns of pathogens involved in infective arthritis are limited in Korea. Therefore, we needed to investigate the current distribution of causative microorganisms of infective arthritis in Korean communities.
The proportion of infective arthritis with Gram-negative microorganisms ranged from 5% to 20% according to previous studies [2, 3, 4, 12, 13]. Among Gram-negative bacteria, Escherichia coli, Pseudomonas aeruginosa, and Haemophilus influenzae are frequently isolated pathogens in cases of infective arthritis. Although elderly or frail individuals, patients with recurrent urinary-tract infections, and recent abdominal surgery are considered to be risk factors for Gram-negative bacterial infections in cases with infective arthritis, the evidence was not clear [14]. Since Gram-positive bacteria are much more prevalent pathogens in infective arthritis than Gram-negative bacteria, empirical therapy for infective arthritis usually includes antibiotics for Gram-positive bacteria. The risk of inappropriate use of empirical antibiotics in Gram-negative bacterial infective arthritis is high. Thus, it is essential to verify risk factors of Gram-negative bacteria in order to successfully treat this disease.
This study aims to describe clinical manifestations and microbiological etiology of infective arthritis in a single tertiary hospital in Korea and to analyze risk factors for Gram-negative bacteria.
Materials and Methods
1. Study population
Patients admitted with infective arthritis in a single-tertiary hospital of Korea from January 2009 to December 2018, were evaluated retrospectively. The study population only included adults older than 18 years of age. Each patient was identified with ICD-10 diagnostic codes, including M00.00-29, M00.80-99, M01.00-09, M.01.20-69, M01.80-89, and T.84.5. Only patients with microorganisms identified from either synovial fluid or blood were included and patients with culture-negative infective arthritis were not included in this study. Patients with tuberculosis arthritis and brucella arthritis were also excluded. The criteria for diagnosis of septic arthritis were derived from those of Newman [15]. Group A was defined by identification of microorganisms from synovial fluid or joint tissue, while Group B was defined by evidence of joint inflammation and pathogen isolation from blood.
2. Study design and data collection
A retrospective cohort study was conducted in order to determine the clinical characteristics and microbiologic etiology of patients with infective arthritis and risk factors for Gram-negative microorganisms among subjects with infective arthritis. Medical records of patients were retrospectively reviewed for the following variables; age at admission, sex, definition criteria, underlying immunosuppression, steroid usage (identified when the patients took steroid at the time of admission regardless of dose and duration), comorbid conditions, previous history of infective arthritis, healthcare associated infection (defined with at least following history: recent antibiotics usage defined when antibiotics was used within 90 days, hospitalization for 2 days or more in the preceding 90 days, residence in a long term care facility, chronic dialysis within 30 days, and infective arthritis that developed more than 48 hours after admission, possible risk factors for infective arthritis (history of acupuncture and intra-articular injection, recent trauma, recent joint surgery within 90 days, prosthetic joint and other site of infections, including bacteriuria, adjacent skin and soft tissue infection, and infective endocarditis), presenting symptoms, site of infective arthritis, complications, white blood cell count (WBC), red blood cell count, platelet count, erythrocyte sedimentation rate (ESR), blood urea nitrate, serum creatinine, aspartate transaminase (AST), alanine transaminase (ALT), total bilirubin, albumin concentration, C-reactive protein (CRP), analysis of synovial fluid (WBC count, protein concentration, and glucose concentration), and results of microbiological analysis.
3. Microbiological tests
Identification testing was performed using the Vitek 2 automated analyzer system (bioMérieux, Marcy l'E'toile, France) until 2013. The Bruker Biotyper matrix-assisted laser desorption ionization–time of flight mass spectrometry (MALDI-TOF MS) system (Bruker Daltonics, Bremen, Germany) was introduced in 2013 and used for identification testing from 2013 onwards. All susceptibility testing was performed using the Vitek 2 automated analyzer system with a VITEK AST2 N212 card (bioMérieux). Antibiotic susceptibility was interpreted using the Clinical and Laboratory Standards Institute (CLSI) criteria [16].
4. Statistical analysis
Pearson's chi-square test was used to compare two variables. Fisher's exact test was used instead of Pearson's chi-square test when the expected value of at least one cell is <5. Variables of which P-value <0.1 from univariate analysis were included into multivariate logistic regression model for determining independent risk factors of infections with Gram-negative microorganisms. The results of statistical analyses were judged to be significant when the P-value was <0.05. Variance inflation factor (VIF) was used to test multi-collinearity. VIF >10 was regarded that multi-collinearity exist. Hosmer-Lemshow's test was done to test goodness-of-fit. SPSS software, version 23.0 (SPSS, IBM Copr., Armonk, NY, USA), was used for all statistical analyses.
5. Ethics approval and consent to participate
Ethical approval was obtained from the institutional review board at Severance Hospital (Approval No. 4-2020-1138). The need for patient consent was waived due to the retrospective nature of the study.
Results
1. Demographic data
Two hundred and forty eight cases were identified by ICD-10 diagnostic codes and culture results. Patients who were not admitted (n = 9), patients with no microbiological confirmation (n = 55), and tuberculosis arthritis patients (n = 3) were excluded. Thus, 181 patients remained and were evaluated in this study. The median age of the subjects was 68.0 (Range: 55 – 75.5; Table 1). The sex ratio was approximately equal, with 53.0% males. We also compared native joint patients to prosthetic joint patients. The prosthetic joint group was generally older than the native joint group, while more men were in the native joint group.
Table 1
Baseline characteristics of patients with septic arthritis
2. Comorbidities and potential risk factors
The most common underlying comorbidities were hypertension (n = 102; 56.4%), followed by diabetes mellitus (n = 64; 35.4%), and solid cancer (n = 21; 11.6%; Table 1). There were 21 cases (11.6%) of patients with previous infective arthritis. Other underlying conditions included end stage renal disease (n = 10; 5.5%), liver cirrhosis (n = 6; 3.3%), coronary artery disease (n = 16; 8.8%), rheumatologic joint disease (n = 10; 5.6%), hematologic malignancy (n = 5; 2.8%), and transplantation history (n = 5; 2.8%). The number of patients who had used any immunosuppressive agent, except steroids, at admission was nine (5.0%). The number patients who were actively using steroid was 19 (10.5%).
Healthcare-associated infection occurred in 62 cases (34.3%). Of these cases, 48 (70.6%) patients reported previous antibiotics usage, 2 (3.2%) patients had been recently admitted to hospital, and 5 (8.1%) patients had been transferred from a long-term care facility. The number of patients under chronic dialysis was 10 (16.1%), and the number of patients with infective arthritis that had developed in hospital was nine (14.5%).
In Table 1, we described potential risk factors of infective arthritis. The most common risk factor was a history of prosthetic joint replacement (n = 46; 25.4%). The second most common risk factor was bacteriuria (n = 31; 17.1%), followed by a history of intra-articular injection (n = 26; 14.4%), adjacent skin and soft tissue infection (n = 22; 12.2%), and recent joint surgery (n = 21; 11.6%). Bacteriuria was diagnosed upon identification of any microorganisms in urine samples that had been collected within 48 hours of admission. The number of patients who had a history of acupuncture was 12 (6.6%). There were no cases related to intravenous drug use.
There were no significant differences in comorbidities and risk factors between the native joint group and the prosthetic joint group except history of infective arthritis, intra-articular injection, and recent joint surgery. More patients with prosthetic joint had history of infective arthritis (n = 11; 23.9%) and recent joint surgery (n = 10; 21.7%) than patients with native joints (n = 10; 7.4% and n = 11; 8.1%, respectively). On the other hand, intra-articular injection history was more frequent in patients with native joint.
3. Symptoms, signs, and investigation at presentation
Median duration from onset of symptoms until admission was 5.0 days (range: 2.0 - 10.0; Table 2). Most patients had symptoms and signs of joint inflammation, such as arthralgia (n = 179; 100%), painful limitation of motion (n = 140; 93.3%), joint swelling (n = 143; 88.8%), and local heating sensation (n = 95; 88.8%). Fever was present in 65.7% of the patients and shock, defined as a blood pressure below 90/60 mmHg, was present in 12.7% of patients. When comparing the native joint group and the prosthetic joint group, there were no significant differences in clinical manifestations.
Table 2
Clinical presentation and laboratory findings of patients with infective arthritis
Leukocytosis, defined as blood WBC count above 10.8 × 103/mm3, was recorded in 51.4% (n = 93) of patients. The average of WBC count was 12002.9 ± 5679.1/mm3. Thrombocytosis was recorded in 19.9% (n = 36) of the patients, while thrombocytopenia was recorded in 18.8% (n = 34). Serum albumin levels decreased in 66.9% (n = 121) of the patients. ESR (reference range: 0.0 - 20.0 mm/hr) and CRP (reference range: 0.0 - 8.0 mg/L) were elevated in most of the cases, 96.1% and 97.8% respectively. The average of synovial WBC count was 82,138 ± 89,448/mm3. The average protein level and glucose level of synovial fluid were 4.3 ± 1.3 g/dL and 52.4 ± 58.0 mg/dL, respectively.
We also compared the native joint group with the prosthetic joint group. CRP and ALT were higher in the native joint group than in the prosthetic joint group. In the native joint group, the median level of CRP and ALT were 180.6 mg/L (range: 88.6 – 276.6 mg/L) and 22.5 IU/L (range: 14.0 – 37.0 IU/L), while the median levels were 126.9 mg/L (range: 48.5 – 224.6 mg/L) and 17.0 IU/L (range: 10.5 – 30.0 IU/L) in the prosthetic joint group. Other laboratory results showed no significant differences between the two groups.
4. Involved joints
The knees (n = 115; 63.6%) and shoulders (n = 32; 17.7%) were the joints most frequently affected, followed by hip joints (n = 18; 9.9%; Table 1, Fig. 1). Of the infected joints, 46 (24.5%) were prosthetic. Most of the affected prosthetic joints were knee joints (n = 40). Five infected hip joints were prosthetic and there was a case of prosthetic ankle joint infection. Other joints included sacroiliac joint (n = 4), sternoclavicular joint (n = 2), spine facet joint (n = 1), and metacarpopharyngeal joint (n = 1). Twelve patients (6.6%) presented with multiple joints involvement.
Figure 1
Site of infected joint.
Eighty-seven percentage (n = 40) of prosthetic joint infection patients also had knee joint infections, which was more frequent than in the native joint infection group (n = 75; 55.5%; P <0.001). All shoulder joint infection cases occurred in the native joint.
5. Distributions of causative microorganisms and antibiotics susceptibility
Ninety-two cases were defined with positive synovial fluid culture alone, while in 58 cases, both synovial fluid and blood culture were positive. Additionally, 31 cases were defined with blood culture alone. Almost all prosthetic joint infections were defined with positive synovial fluid culture results (n = 45; 98.8%), while only 77.8% (n = 105) of native joint infections were defined with positive synovial fluid culture. The most commonly identified pathogen was S. aureus (n = 97; 51.1%; Table 3). The second most common pathogen was Streptococci (n = 40; 21.1%), followed by coagulase-negative Staphylococci (n = 16; 8.4%), and Enterobacteriaceae (n = 16; 8.4%). Pseudomonas sp. made up 2.1% of the cases (n = 4). The number of cases with other Gram-positive microorganisms, including Bacillus sp., Enterococci sp., and Micrococcus sp., was five (2.6%) and there were six (3.2%) cases of other Gram-negative microorganisms. Other Gram-negative microorganisms included Burkholderia cepacia, Burkholderia pseudomallei, Acinetobacter baumannii, Stenotrophomonas maltophilia, and Achromobacter sp. Overall, 83.2% of the cases were caused by Gram-positive microorganisms, while 13.7% of cases were due to Gram-negative microorganisms. Only six fungal arthritis cases were identified (Fig. 2). We also analyzed the distribution of causative microorganisms by joint type, but there were no differences in causative microorganisms between the native joint group and prosthetic joint group.
Table 3
Microorganisms causing septic arthritis
Figure 2
Distribution of causative microorganisms according to the site of infected joints.
MSSA, methicillin-sensitive Staphylococcus aureus; MRSA, methicillin-resistant Staphylococcus aureus; MSCNS, methicillin-sensitive coagulase-negative Staphylococci; MRCNS, methicillin-resistant coagulase-negative Staphylococci; Other G+, other Gram-positive microorganisms; Other G-, other Gram-positve microorganisms.
The number of MRSA were 39.2% (n = 38) of all S. aureus cases and the number of MRCNS were 50% (n = 8) of all CNS cases. Among the MRSA cases, 35 cases were healthcare associated infections while 61 cases were community-acquired. We also analyzed antibiotic resistance profile in accordance with joint type, but there was no significant difference in antibiotics resistance profile between overall group, native joint patients, and prosthetic joint patients. The proportion of MRSA was higher in healthcare associated infection compared with community-acquired infection (odds ratio [OR], 3.185; 95% confidence internal [CI], 1.134 - 7.58; P = 0.007; Table 3). Gram-positive microorganisms were identified more frequently in community-acquired infection than in healthcare associated infection, It was mainly due to Streptococci. The number of cases with Streptococci was 29.4% (n = 35) of community-acquired infection cases, while the number was only 8.1% (n = 5) of healthcare associated infection cases. In the contrary, Gram-negative microorganisms were more frequently cultured in healthcare associated infection. Details of antibiotics susceptibility profiles are presented in Tables 4, 5, 6.
Table 4
Antibiotic resistance rates (%) of Staphylococcus aureus and coagulase-negative Staphylococci causing infective arthritis between 2009 – 2018
Table 5
Antibiotic resistance rates (%) of healthcare-associated Staphylococcus aureus and community-acquired Staphylococcus aureus causing infective arthritis between 2009 – 2018
Table 6
Antibiotic resistance rates (%) of Enterobacteriaceae causing infective arthritis between 2009 and 2018
6. Risk factors for Gram-negative organism infection
Univariate analysis of risk factors for infective arthritis caused by Gram-negative microorganisms showed association with female sex, solid cancer, previous usage of steroid, shock at presentation, hip joint infection, shoulder joint infection, bacteriuria, and healthcare associated infection (Table 7). A multivariate model identified the following as independent risk factors for Gram-negative organism infection: bacteriuria (OR 3.83; 95% CI, 1.42 - 10.34; P = 0.008), infective arthritis in hip (OR 3.84; 95% CI, 1.21 - 12.15; P = 0.022), and steroid use at admission (OR 3.82; 95% CI, 1.15 - 12.69; P = 0.029; Table 7). VIF showed that multi-collinearity didn't exist between each variables (bacteriuria, VIF = 1.02; hip joint involvement, VIF = 1.01; steroid use, VIF = 1.01). Hosmer-Lemshow's test showed goodness of fit for the multivariate logistic regression model (P = 0.619).
Table 7
Risk factors for infective arthritis due to Gram-negative microorganisms
Discussion
In this work, we describe the current distribution of causative microorganisms of infective arthritis in Korea. We show, as previous studies have described, that S. aureus and Streptococci are still the most common pathogens of infective arthritis [1, 2, 3, 4]. The proportion of Enterobacteriaceae and CNS was also similar to the aforementioned studies.
In the case of prosthetic joint infections, the distribution of causative microorganisms was different from the previous studies. S. aureus and CNS were known to contribute to between 50 and 60% of prothetic joint infections, and the proportions of prothetic joint infections caused by S. aureus and CNS were almost equal [17]. In comparison with the result, our study showed that the proportion of CNS was relatively low and that of S. aureus and Streptococci was high. This difference seems to be resulted from several limitations of this study. This study was conducted in a single tertiary hospital, and since this study is a retrospective study, there can also be a possibility that diagnostic codes were missed by clinicians. Those factors seem to affect the result of microbiological distribution.
The rate of antibiotic resistance was remarkable in this study. We found that 39.2% of the S. aureus cases and half of the CNS cases were methicillin-resistant. Overall, methicillin-resistant pathogens comprised a quarter of identified microorganisms. Becaure healthcare-associated infection was as high as 34.3% in this study, the antibiotic resistance rate could be overestimated. However, 27.6% of S. aureus cases were methicillin-resistant even in community-acquired infections, while 57.1% of S. aureus cases were methicillin-resistant in healthcare associated infections. In coagulase-negative Staphylococci infection, 55.6% of community-acquired infections and 42.9% of healthcare associated infections were methicillin-resistant, respectively. Therefore, we can say that the proportion of methicillin-resistant pathogens is notably high in infective arthritis not only in healthcare associated infections, but also in community-acquired infections.
According to a previous study, published in 2008 in Korea, reported that only 7.4% of identified microorganisms were MRSA in 1995 - 2006 [5]. Compared to the study by Seo et al. (2008) there has been a notable change in the antibiotic susceptibility of infective arthritis in South Korea. Several studies have also reported that the proportion of MRSA in infective arthritis increased. In 2005, a study from the United States of America reported 15 out of 59 septic arthritis cases (25%) involving MRSA [9]. Furthermore, a study from France also reported that there was an increased incidence of MRSA infections in 2009 – 2013 (30.8% of S. aureus cases) compared to 1979 – 2013 [18]. A study conducted between 2008 and 2011 in Taiwan, reported that MRSA arthritis was found in 38 (40.9%) cases of S. aureus [7]. Given these data, methicillin resistance in infective arthritis seems to be on the rise domestically and globally.
Antibiotic resistance in Gram-negative microorganisms was also at a high rate. Although the third generation cephalosporins and fluoroquinolone are the commonly used antibiotics for Gram-negative microorganisms, the resistance rate of cefotaxime, ciprofloxacin, and levofloxacin were 25.0%, 57.1%, and 15.4%, respectively. This trend was especially remarkable in healthcare-associated infections. In healthcare-associaed infections, the resistance rate of cefotaxime, ciprofloxacin, and levofloxacin rised as high as 57.1% (n = 4), 75.0% (n = 3) and 33.3% (n = 2).
A study conducted between 2016 to 2017 in Korea also showed high proportion of antibiotics resistance in Gram-negative microorganisms [19]. According to the study, the overall cefotaxime resistance rate was 34.7% in E. coli and 27.0% in K. pneumoniae. When it comes to healthcare-associated infections, the cefotaxime resistance rate was as high as 55.8% in E. coli and 48.2% in K. pneumoniae. Given these findings, we can say that antibiotics resistance of Gram-negative microorganisms is serious in infective arthritis, especially in healthcare-associated infections.
Significantly, in our study, we have also identified independent risk factors of Gram-negative infections in infective arthritis. These include bacteriuria, hip joint involvement, and usage of steroids. Although old age and frailty, recurrent urinary-tract infection, and recent abdominal surgery were previously known risk factors of Gram-negative infections in infective arthritis, the evidence was not clear [14]. A study, dealing with asymptomatic bacteriuria as a risk factor for prosthetic joint infection, reported that asymptomatic bacteriuria was associated with Gram-negative infection in patients with prosthetic joint infection [20]. Our study agrees with the findings made by Sousa et al. and demonstrates that bacteriuria at admission is a risk factor for Gram-negative infection in prosthetic and native joint infection. Since Gram-negative microorganisms were major pathogens of urinary tract infection, hematogenous spread of pathogens can explain the association between bacteriuria and Gram-negative infective arthritis.
Generally, it is more difficult to perform intraarticular injection in the hip joint than in the knee and shoulder joints. In our study, intraarticular injection was performed on only one patient out of 18 with hip joint involvement, while 25 of 163 patients with other joint involvement had history of intraarticular injection. This suggests that the hip joint is more often involved to infective arthritis via hematogenous spread than other joints. Different modes of pathogenesis can cause frequent involvement of Gram-negative microorganisms in the hip joints. Whereas, frequent Gram-negative infective arthritis in patients who use steroids may be due to the immunomodulatory effects of steroids. Given that immunocompromised patients are vulnerable to hematogenous spread of microorganisms living in the gastrointestinal tract and urinary tract, the immunomodulatory effect of steroids may also encourage the spread of Gram-negative microorganisms to the joints. Other immunomodulatory comorbidities and underlying conditions, including hematologic malignancy, transplantation, and uses of other immunosuppressant were too small in number to compare as risk factors. Thus, additional studies with a focus on the effects of other immunomodulatory conditions are in need.
A limitation of our study is that it was a retrospective single center study. There could, therefore, be information biased by the nature of a retrospective study. Given that patient histories of antibiotics usage were assessed only with in hospital electrical medical records, previous antibiotics usage may have been underestimated. Additionally, use of steroids were also evaluated with medicine at time of admission. The fact that our study was carried out in one tertiary hospital was also a limitation. This hospital, Severance Hospital, is one of the biggest hospitals in Korea, which has over 2,000 beds. Traditionally, there has been a tendency to refer severely ill patients to this hospital. Thus, it is possible that our study is not representative of the entire Korean population. Further prospective multicenter studies are needed to confirm the result of our work.
In conclusion, infective arthritis with methicillin-resistant microorganisms reached up to about 25% in a single-tertiary hospital in Korea. When urinary tract infection is suspected, infective arthritis occurs in the hip joint, or in patients who are using steroids among infective arthritis patients, empirical treatment covering Gram-negative organism can be considered.
Funding:This work was supported by research grants for deriving the major clinical and epidemiological indicators of people with HIV (Korea HIV/AIDS Cohort Study, 2019-ER5101-00), and a grant from the Ministry of Health & Welfare, Korea (grant number: HI14C1324).
Conflict of Interest:No conflict of interest.
Author Contributions:
Conceptualization: YL, YSC, YJS, JHH, SMA, WJL, JHK, JK, SJJ, NSK, JSY, JYA, JYC.
Data curation: YL.
Formal analysis: YL.
Methodology: YL, JYC.
Supervision: SJJ, NSK, JSY, JYA, JYC.
Validation: SJJ, NSK, JSY, JYA, JYC.
Visualization: YL.
Writing - original draft: YL.
Writing - review & editing: YL, JYC.
References
-
Le Dantec L, Maury F, Flipo RM, Laskri S, Cortet B, Duquesnoy B, Delcambre B. Peripheral pyogenic arthritis. A study of one hundred seventy-nine cases. Rev Rhum Engl Ed 1996;63:103–110.
-
-
Seo SS, Ha DJ, Kim CW, Kim KW, Seo JH. Etiologic transition of septic arthritis of the knee. J Korean Knee Soc 2008;20:44–49.
-
-
Clinical and Laboratory Standards Institute (CLSI). Performance standards for antimicrobial susceptibility testing. 26th ed. Wayne, PA: CLSI; 2016.CLSI supplement M100.
-
-
Lee H, Yoon EJ, Kim D, Jeong SH, Won EJ, Shin JH, Kim SH, Shin JH, Shin KS, Kim YA, Uh Y, Yang JW, Kim IH, Park C, Lee KJ. Antimicrobial resistance of major clinical pathogens in South Korea, May 2016 to April 2017: first one-year report from Kor-GLASS. Euro Surveill 2018;23:pii=1800047
-
Cited by
MeSH Terms
Figures
1 / 2
Tables
1 / 7
ORCID IDs
Funding Information
-
Korea HIV/AIDS Cohort Study
2019-ER5101-00
-
Ministry of Health and Welfare
HI14C1324
