- Email: [email protected]
Levofloxacin-resistant Stenotrophomonas maltophilia: risk factors and antibiotic susceptibility patterns in hospitalized patients
Journal of Hospital Infection xxx (xxxx) xxx Available online at www.sciencedirect.com
Journal of Hospital Infection journal homepage: www.elsevier.com/locate/jhin
Levofloxacin-resistant Stenotrophomonas maltophilia: risk factors and antibiotic susceptibility patterns in hospitalized patients C.H. Wang a, C.-M. Yu b, S.-T. Hsu c, R.-X. Wu d, * a
Division of Infectious Diseases and Tropical Medicine, Department of Internal Medicine, Tri-Service General Hospital, National Defense Medical Center, Taipei, Taiwan b Department of Clinical Pathology, National Defense Medical Center, Tri-Service General Hospital, Taipei, Taiwan c Infection Control Office, Tri-Service General Hospital, National Defense Medical Center, Taipei, Taiwan d Division of Infectious Diseases, Tri-Service General Hospital Penghu Branch, Penghu, Taiwan
A R T I C L E
I N F O
Article history: Received 3 August 2019 Accepted 2 September 2019 Available online xxx Keywords: Risk Resistance Stenotrophomonas maltophilia Levofloxacin Multiple-drug resistance
S U M M A R Y
Background: Levofloxacin has been considered as an alternative treatment for Stenotrophomonas maltophilia infection. However, levofloxacin-resistant S. maltophilia (LRSM) are emerging worldwide. Aim: To investigate LRSM risk factors in hospitalized patients and to determine antibiotic susceptibility patterns of LRSM isolates. Methods: In a retrospective matched caseecontrolecontrol study, LRSM patients (the case group) were compared with two control groups: levofloxacin-susceptible S. maltophilia (LSSM) patients (control group A) and non-S. maltophilia-infected patients (control group B). Conditional logistic regression was used to analyse risk factors for LRSM occurrence. Tigecycline, ceftazidime, colistin, and trimethoprim/sulfamethoxazole (TMP/SMX) susceptibilities in collected LRSM clinical isolates were determined. Findings: A total of 105 LRSM, 105 LSSM, and 105 non-S. maltophilia-infected patients were analysed. The first multivariate analysis (cases vs group A) revealed that previous fluoroquinolones use was significantly associated with LRSM occurrence, and the second multivariate analysis (cases vs group B) revealed that previous fluoroquinolone use, previous intensive care unit stay, and the number of previous exposures to different classes of antibiotics were significantly associated with LRSM occurrence. Of all the LRSM isolates tested for antibiotic susceptibility, ceftazidime, TMP/SMX, tigecycline, and colistin resistance rates were 42.0, 99.0, 78.0, and 40.0%, respectively. Conclusion: LRSM antibiotic susceptibility patterns revealed multiple-drug resistance, which further limits treatment options for clinicians. To reduce LRSM occurrence, proper use of antibiotics, especially fluoroquinolones, is mandatory. ª 2019 The Healthcare Infection Society. Published by Elsevier Ltd. All rights reserved.
* Corresponding author. Address: Division of Infectious Diseases, Tri-Service General Hospital Penghu Branch, Penghu, No. 90, Qianliao 880, Magong City, Penghu County, Taiwan. Tel.: þ886 6 9211116; fax: þ886 2 9211117. E-mail address: [email protected] (R.-X. Wu). https://doi.org/10.1016/j.jhin.2019.09.001 0195-6701/ª 2019 The Healthcare Infection Society. Published by Elsevier Ltd. All rights reserved. Please cite this article as: Wang CH et al., Levofloxacin-resistant Stenotrophomonas maltophilia: risk factors and antibiotic susceptibility patterns in hospitalized patients, Journal of Hospital Infection, https://doi.org/10.1016/j.jhin.2019.09.001
2
C.H. Wang et al. / Journal of Hospital Infection xxx (xxxx) xxx
Introduction
Study population
Stenotrophomonas maltophilia is an emerging multidrugresistant nosocomial pathogen contracted in hospital wards [1]. It mainly affects immunocompromised or severely debilitated patients and contributes to a variety of infections, including pneumonia, bacteraemia, and skin and soft tissue infections, with varying clinical presentations [2]. The mortality rate attributed to S. maltophilia infection can be as high as 37.5%, and it is difficult to manage because it exhibits high-level intrinsic resistance to a variety of structurally unrelated antibiotics, including b-lactams, aminoglycosides, and carbapenems, drastically reducing available antibiotic treatment options [3e5]. Historically, trimethoprimesulfamethoxazole (TMP-SMX) has been used as a first-line S. maltophilia infection treatment [6]. However, when faced with TMP-SMX-resistant S. maltophilia infection or TMP-SMX-intolerant patients, levofloxacin is used as an alternative, based on data from recent studies showing that its clinical efficacy is similar to that of TMP-SMX [7,8]. Recently, however, there has been an increase in the number of reports on the prevalence of levofloxacin resistance in S. maltophilia [9e11]. This may further limit treatment options for clinicians because of the inadequate data on the susceptibility of other possible in-vitro active antibiotics, beyond TMP-SMX and levofloxacin. Understanding the risk factors for levofloxacin resistance in S. maltophilia may help clinicians identify possible resistant strains, thereby reducing the risks of inappropriate empirical antibiotic resistance and facilitating the implementation of appropriate infection control measures. An unmatched caseecontrol study is commonly used for examining antibiotics as risk factors for the emergence of antimicrobial resistance. Nevertheless, unequal sample size between case and control groups, some confounding factors not controlled before analysis, and selection of patients with a susceptible bacterium as the only control groups, will distort measurement of association [12e14]. Therefore, we conducted this matched caseecontrolecontrol study to investigate risk factors for levofloxacin resistance in S. maltophilia in adult inpatients, as has been suggested [12]. In addition, susceptibility tests on some antibiotics, which, in previous studies, have demonstrated in-vitro activity against collected clinical levofloxacin-resistant S. maltophilia, were also performed [4].
This caseecontrolecontrol study involved the construction of two separate caseecontrol studies using one case group and two control groups (control group A and B). Between January 1st, 2014 and June 30th, 2015, S. maltophilia-infected patients were identified using the computerized medical records system. Patients identified within 48 h of admission and those aged <18 years were excluded. Levofloxacin-resistant S. maltophilia (LRSM) isolates included those which exhibited in-vitro resistance or intermediate susceptibility against levofloxacin. The case group consisted of inpatients for whom the presence of the LRSM isolate was first determined in clinical samples. Control group A consisted of patients with levofloxacin-susceptible S. maltophilia (LSSM) matched with the case group for age (within five years), sex, and infection site at 1:1 ratio. Similarly, control group B consisted of nonS. maltophilia-infected patients, randomly selected from the inpatient population matched with the case group for age (within five years), sex, and time at risk at 1:1 ratio. Time at risk, which represented the duration of the at-risk period for antibiotic-resistant organism acquisition, was defined as the number of days between the admission date and the time when the culture first tested positive for S. maltophilia in the case group and control group A. For control group B, it was defined as the number of days between hospital admission and discharge. Once an eligible pool of controls was established, they were randomly assigned using Microsoft Excel 2013 (Microsoft Corp., Redmond, WA, USA).
Methods Study design The study was designed to determine the risk factors for levofloxacin resistance in S. maltophilia in adult inpatients. The matched caseecontrolecontrol study was conducted between January 1st, 2014 and June 30th, 2015 in Tri-Service General Hospital in Taiwan. The hospital has 1100 beds and is a tertiary referral centre in Northern Taiwan. The study was approved by the Institutional Review Board of the TriService General Hospital (TSGHIRB approval number: 1103-05-154).
Data collection and definitions Using the computerized medical records system, clinical information was systematically collected from identified patients, and then matched with the controls. The collected information included demographic data, underlying diseases on admission, clinical outcomes, hospitalization-related factors, recent therapeutic measures, previous antibiotics exposure, and indwelling medical devices. The Charlson comorbidity index (CCI) was used as an aggregate measure of underlying diseases to indicate comorbidity burden [15]. Previous antibiotics exposure in the case group and control group A patients was defined as: 24 h therapy within two weeks prior to S. maltophilia isolation. In non-infected patients (control group B), previous antimicrobials exposure was defined as: 24 h therapy within two weeks prior to discharge index. Recent chemotherapy and surgery performed within 30 days before S. maltophilia isolation were recorded in the case group and control group A. In control group B (no S. maltophilia isolation), recent therapeutic measures were defined as recent chemotherapy and surgery within 30 days before discharge index. Indwelling invasive medical devices were also recorded.
Microbiology test Between January 1st, 2014 and June 30th, 2015, a total of 100 randomly collected levofloxacin non-susceptible clinical S. maltophilia isolates were tested for antibiotic susceptibility. The Vitek 2 automatic system for Gram-negative rods (bioMe ´rieux, Marcy l’Etoile, France) was used to identify
Please cite this article as: Wang CH et al., Levofloxacin-resistant Stenotrophomonas maltophilia: risk factors and antibiotic susceptibility patterns in hospitalized patients, Journal of Hospital Infection, https://doi.org/10.1016/j.jhin.2019.09.001
C.H. Wang et al. / Journal of Hospital Infection xxx (xxxx) xxx S. maltophilia isolates and to assess susceptibility according to the 2017 Clinical and Laboratory Standards Institute (CLSI) M100-S27 performance standards [16]. Tigecycline breakpoints were established according to the European Committee on Antimicrobial Susceptibility Testing for Enterobacteriaceae (1 mg/mL, susceptibility; 2.0 mg/mL, intermediate; and >2 mg/mL, resistance). Colistin breakpoints were established according to CLSI criteria for P. aeruginosa (2 mg/mL, susceptibility; 4.0 mg/mL, intermediate; and 8 mg/mL, resistance) [17].
Statistical analysis All data were analysed using SPSS v16.0 (SPSS Inc., Chicago, IL, USA). The results are presented as mean (standard deviation) or median (range), whereas categorical variables are presented as percentages. Bivariate analysis of categorical variables was conducted using the c2-test or Fisher’s exact test, as deemed appropriate, whereas Student’s t-test or ManneWhitney U-test (for data not normally distributed) was used to analyse continuous variables. The normality of continuous variables was verified using the KolmogoroveSmirnov test. Variables with P < 0.05 on bivariate analysis were included in the forward stepwise conditional multivariate analysis logistic regression model. Two multivariate models were produced from the data obtained. The first model compared the case group with control group A, and the second model compared the case group with control group B. All tests were two-tailed, and P < 0.05 was considered statistically significant.
Results Study population Between January 1st, 2014 and June 30th, 2015, the cultures of 809 patients tested positive for S. maltophilia; the most
3
frequent isolation sites were the respiratory tract (85.8%), followed by blood (4.7%), wound tissue (3.6%), abdominal ascites (2.6%), urine (2.2%), and central venous catheter (1.1%). Further, the isolates were mostly identified in intensive care unit (ICU) patients (48.2%), followed by medical ward (30.7%), surgical ward (10.9%), and respiratory centre (10.2%) patients; of the 809 patients, 133 were levofloxacin resistant (16.4%). Of the 133 LRSM patients, 24 with LRSM isolation within 48 h after admission were excluded, as well as four other patients, because of unsuccessful matching. Therefore, a total of 105 LRSM patients were included, with infection sites as follows: 101, respiratory tract; one, blood; one, urine; and two, wound. To analyse the risk factors associated with LRSM isolation, 105 LSSM patients and 105 nonS. maltophilia-infected patients were included for subsequent analyses. Patient characteristics, including demographics, underlying disease, and clinical outcomes, are shown in Table I.
Risk factor analysis: resistant case group vs control group A As shown in Table II, bivariate analysis revealed that the differences in hospitalization-related factors, CCI, recent therapeutic measures, and indwelling medical devices of the case group and control group A were not statistically significant (P > 0.05). With regard to previous antibiotics exposure, fluoroquinolones exposure (43.8% vs 8.6%; P < 0.001) was associated with LRSM occurrence, whereas previous fourth-generation cephalosporins exposure was inversely associated with LRSM occurrence (28.6% vs 42.9%, P ¼ 0.042). In the case group, of all the inpatients who received fluoroquinolones (N ¼ 46), 33, nine, and four had previously received levofloxacin, ciprofloxacin, and moxifloxacin, respectively, and of all the control group A patients receiving fluoroquinolones, five, two, and two had
Table I Stenotrophomonas maltophilia patient characteristics Variable
Demographics Age (years), median (range) Male Underlying disease Cerebrovascular disease Congestive heart failure Renal insufficiency Chronic obstructive pulmonary disease Liver cirrhosis Diabetes mellitus Connective tissue disease Malignancy Clinical outcome In-hospital mortality Hospital stay after isolation, median (range) a b c
Case groupa (N ¼ 105)
Control group Ab (N ¼ 105)
Control group Bc (N ¼ 105)
80 (21e95) 72 (68.5%)
80 (18e93) 72 (68.5%)
78 (26e96) 72 (78.5%)
47 (44.8%) 19 (18.1%) 35 (33.3%) 18 (17.1%) 4 (3.8%) 41 (39.0%) 6 (5.7%) 22 (21.0%)
37 13 24 23 4 36 1 26
24 (22.9%) 8 (7.6%) 27 (25.7%) 15 (14.3%) 10 (9.5%) 39 (37.1%) 3 (2.9%) 34 (32.4%)
22 (20.9%) 22 (1e362)
37 (35.2%) 23 (0e305)
(35.2%) (12.4%) (22.9%) (21.9%) (3.8%) (34.3%) (0.95%) (24.8%)
11 (10.4%) e
Patients with levofloxacin-resistant S. maltophilia (case group). Patients with levofloxacin-susceptible S. maltophilia (control group A). Patients without S. maltophilia infection (control group B).
Please cite this article as: Wang CH et al., Levofloxacin-resistant Stenotrophomonas maltophilia: risk factors and antibiotic susceptibility patterns in hospitalized patients, Journal of Hospital Infection, https://doi.org/10.1016/j.jhin.2019.09.001
4
C.H. Wang et al. / Journal of Hospital Infection xxx (xxxx) xxx
Table II Bivariate analysis of levofloxacin-resistant S. maltophilia isolation risk factors Case groupa (N ¼ 105) Control group Ab (N ¼ 105) Control group Bc (N ¼ 105) P-valued
Factors related to hospitalization Time at riskf Recent admission within 3 months Previous ICU stay Charlson comorbidity index Recent therapeutic measures Chemotherapy Surgery Previous antibiotics exposure Macrolides Aminoglycosides Penicillin-lactamase inhibitors Third-generation cephalosporins Fourth-generation cephalosporins Carbapenems Glycopeptides Fluoroquinolones Trimethoprimesulfamethoxazole Colistin Otherg No. of different classes of previous antibiotic treatment Previous indwelling invasive medical devices Central venous catheterh Nasogastric tube Urinary catheter Surgical drain Ventilator
25.0 35 84 4.0
(3e112) (33.3) (80.0) (0e11)
3 (2.9) 30 (28.6) 10 7 36 21 30 40 32 46 3 8 9 2.0
50 97 76 15 72
(9.5) (6.7) (34.3) (20.0) (28.6) (38.1) (30.5) (43.8) (2.9) (7.6) (8.5) (0e6.0)
(47.6) (92.4) (72.4) (14.3) (68.6)
23.0 34 80 3.0
(5e106) (32.4) (76.2) (0e11)
8 (7.6) 27 (25.7) 9 6 29 31 45 50 28 9 4 8 11 2.0
52 98 82 18 76
(8.6) (5.7) (27.6) (29.5) (42.9) (47.6) (26.7) (8.6) (3.8) (7.6) (10.4) (0e6.0)
(49.5) (93.3) (78.1) (17.1) (72.4)
P-valuee
e 35 (33.3) 30 (28.6) 3.0 (0e14)
0.340 0.884 0.506 0.089
e 1.000 <0.001 0.117
9 (8.6) 24 (22.9)
0.147 0.631
0.099 0.332
0.618 0.594 0.319 0.135 0.042 0.184 0.556 <0.001 0.706 1.000 0.819 0.328
0.054 0.069 0.338 0.226 0.014 <0.001 0.001 0.001 1.000 0.258 0.126 <0.001
0.789 0.782 0.306 0.565 0.547
0.024 <0.001 0.001 0.435 <0.001
3 1 33 15 14 15 3 22 3 4 0 1.0
34 57 50 19 24
(2.9) (0.95) (31.4) (14.3) (13.3) (14.3) (2.9) (21.0) (2.9) (3.8) (0) (0.0e5.0)
(32.4) (54.3) (47.6) (18.1) (22.9)
ICU, intensive care unit. a Patients with levofloxacin-resistant S. maltophilia (case group). b Patients with levofloxacin-susceptible S. maltophilia (control group A). c Patients without S. maltophilia infection (control group B). d Comparison of case group and control group A patients. e Comparison of case group and control group B patients. f Days of stay prior to S. maltophilia isolation. g Other antimicrobials, including tigecycline, linezolid, and daptomycin. h Including double-lumen catheter for haemodialysis, central venous catheter, and peripherally inserted central catheter.
previously received levofloxacin, ciprofloxacin, and moxifloxacin, respectively. Further multivariate analysis with forward stepwise conditioning logistic regression showed that only previous use of fluoroquinolones remained in the final models and, thus, was considered a levofloxacin-resistance acquisition risk factor (odds ratio (OR): 7.17; 95% confidence interval (CI): 3.05e16.84; P < 0.001).
Risk factor analysis: resistant case group vs control group B (non-S. maltophilia-infected group) As shown in Table II, bivariate analysis of hospitalizationrelated factors revealed that patients in the case group showed a higher risk of recent ICU admission compared with those in control group B (80.0% vs 28.6%, P < 0.001). There were no statistically significant differences in the CCI and recent therapeutic measures of the two groups (P > 0.05).
Table III Multivariate analysis of levofloxacin-resistant Stenotrophomonas maltophilia isolation risk factors Variable
OR (95% CI)
P-value
Case groupa vs control group Ab Fluoroquinolones 7.17 (3.05e16.84) <0.001 Case group vs control group Bc Previous ICU stay 15.25 (3.66e63.57) <0.001 No. of different classes of 2.82 (1.577e5.05) <0.001 previous antibiotic treatment Fluoroquinolones 3.41 (1.05e11.06) 0.041 OR, odds ratio; CI, confidence interval; ICU, intensive care unit. a Patients with levofloxacin-resistant S. maltophilia (case group). b Patients with levofloxacin-susceptible S. maltophilia (control group A). c Patients without S. maltophilia infection (control group B).
Please cite this article as: Wang CH et al., Levofloxacin-resistant Stenotrophomonas maltophilia: risk factors and antibiotic susceptibility patterns in hospitalized patients, Journal of Hospital Infection, https://doi.org/10.1016/j.jhin.2019.09.001
C.H. Wang et al. / Journal of Hospital Infection xxx (xxxx) xxx
5
Table IV Susceptibility of 100 levofloxacin-resistant Stenotrophomonas maltophilia isolates to different antibiotics Antimicrobial agent
Ceftazidime Levofloxacin Trimethoprimesulfamethoxazole Tigecyclinea Colistinb
MIC range (mg/L)
1 to 64 4 to 8 1 to 16 0.5 to 8 0.5 to 16
No. of indicated susceptibility category Susceptible
Intermediate
Resistant
49 0 1 10 58
9 36 e 12 2
42 64 99 78 40
MIC, minimum inhibitory concentration. a Tigecycline breakpoints established by the European Committee on Antimicrobial Susceptibility Testing for Enterobacteriaceae (1 mg/mL, susceptibility; 2.0 mg/mL, intermediate; and >2 mg/mL, resistance). b Colistin breakpoints established by the Clinical and Laboratory Standards Institute for P. aeruginosa (2 mg/mL, susceptibility; 4.0 mg/mL, intermediate; and 8 mg/mL, resistance).
Regarding previous antibiotics exposure, patients in the case group were more likely to be exposed to different classes of antibiotics compared with those in control group B (2.0 vs 1.0, P < 0.001). Regarding previous exposure to the different classes of antibiotics recorded in this study, patients in the case group had higher rates for receiving fourth-generation cephalosporins, carbapenems, glycopeptides, and fluoroquinolones than those in control group B (28.6% vs 13.3%, 38.1% vs 14.3%, 30.5% vs 2.9%, and 43.8% vs 21.0 %; P ¼ 0.014, <0.001, 0.001, and 0.001, respectively). Of all the patients in control group B receiving fluoroquinolones (N ¼ 22), 11, eight, and three had previously received levofloxacin, ciprofloxacin, and moxifloxacin, respectively. In addition, patients in the case group were likely to have invasive medical devices, including central venous catheters, nasogastric tubes, urinary catheters, and ventilators, compared with those in control group B (47.6% vs 32.4%, 92.4% vs 54.3%, 72.4% vs 47.6%, and 68.6% vs 22.9%, P ¼ 0.024, <0.001, 0.001, and <0.001, respectively). Further multivariate analysis with forward stepwise conditional logistic regression after adding significant variables (P < 0.05) in the aforementioned bivariate analysis was performed. The factors that were independently predictive of levofloxacin resistance occurrence were: previous use of fluoroquinolones (OR: 3.41; 95% CI: 1.05e11.06; P ¼ 0.04), recent ICU admission (15.25; 3.66e63.57; P < 0.001), and the number of previous exposures to different classes of antibiotics (2.82; 1.58e5.50; P < 0.001) (Table III).
Microbiological results Of the 100 LRSM clinical isolates, 24, six, 69, and one were obtained from blood, wound tissue, the respiratory tract, and urine, respectively. Ninety-nine isolates were resistant to TMP-SMX, 78 showed concomitant resistance to tigecycline, and 17 were resistant to all the tested antibiotics, whereas colistin showed greater invitro activity than the other tested antibiotics (its susceptibility was only 58%) (Table IV).
Discussion The main finding of this study was that previous fluoroquinolones use, ICU stay, and previous exposure to different antibiotic classes were independent risk factors for subsequent
LRSM occurrence. Additionally, the study revealed the high resistance rates to other potential in-vitro active antibiotics in LRSM isolates. Previous studies on the mechanism of quinolone resistance in S. maltophilia showed various mechanisms, including multidrug efflux pump systems overexpression, DNA gyrase protection proteins encoded by chromosomally quinoloneresistance determinants, and chromosomal mutations of the quinolone-resistance-determining regions in DNA gyrase or DNA topoisomerase IV [18e20]. It has also been suggested that the selective pressure of quinolones may contribute to the subsequent occurrence of quinolone-resistant S. maltophilia [21,22]. Despite extensive experimental studies on the quinolone-resistance mechanism in S. maltophilia, only two clinical studies regarding risk factors for levofloxacin-resistant S. maltophilia isolation have been reported [23,24]. In both studies, the number of levofloxacin-resistant S. maltophilia cases included was limited, which may have decreased the statistical strength of the study, resulting in possible type II errors. Furthermore, the inclusion of patients infected with susceptible organisms as the control group may have led to biased estimates of relative risk [25]. Our study, which was based on a modified caseecontrolecontrol study design with a substantial number of cases, revealed that previous fluoroquinolones use, previous ICU stay, and previous exposure to multiple antibiotic classes were risk factors for subsequent levofloxacin-resistant S. maltophilia isolation. Besides, our study results also provide clinical data to support experimental studies on the association between fluoroquinolones exposure and acquired resistance to levofloxacin in S. maltophilia. As demonstrated by previous studies, this study revealed that other risk factors, including previous ICU stay, and previous multiple classes of antibiotics exposure, were also risk factors for multidrug-resistant pathogen isolation [26,27]. Therefore, the implementation of antibiotic stewardship programmes, especially the careful use of fluoroquinolones, is mandatory to halt LRSM dissemination. The susceptibility tests performed in this study revealed that no tested antibiotics, including tigecycline, showed good in-vitro activity against LRSM isolates. This result was unexpected, since in a recent epidemiology survey, tigecycline showed great potency in vitro [28,29]. This discrepancy might have resulted from the mechanism of quinolone resistance in S. maltophilia. Different from other bacteria, quinolone-resistant S. maltophilia of clinical origin do not
Please cite this article as: Wang CH et al., Levofloxacin-resistant Stenotrophomonas maltophilia: risk factors and antibiotic susceptibility patterns in hospitalized patients, Journal of Hospital Infection, https://doi.org/10.1016/j.jhin.2019.09.001
6
C.H. Wang et al. / Journal of Hospital Infection xxx (xxxx) xxx
have mutations in topoisomerases and overproduction of multidrug efflux pumps may be the major mechanism of quinolone resistance [4,30,31]. S. maltophilia with multidrug efflux pump overexpression may not only be resistant to quinolones but also to other antibiotics [32,33]. Recent studies comparing the outcomes of TMP/SMX- and fluoroquinolone-treated S. maltophilia-infected patients showed that the two drugs had an equal clinical efficacy, although the incidence of adverse events was lower in the fluoroquinolone-treated patients than in the TMP/SMXtreated group [7,8]. Further, a systematic review by Ko et al. suggested the superiority of fluoroquinolones to TMPSMX in the treatment of S. maltophilia infections [34]. The results of our study, however, identified the worrisome consequence of possible multidrug-resistant S. maltophilia occurrence after quinolones exposure. Therefore, empirical levofloxacin use in treating S. maltophilia infection should be exercised with more caution. This study had several limitations. First, given that it was retrospective in nature, it might have resulted in selection and observational bias. Second, molecular studies on LRSM isolates were not performed. Therefore, the molecular mechanism of the quinolone resistance in LRSM isolates could not be verified. Finally, this was a single-centre study in Taiwan; hence, the results may not be generalizable. Thus, in future, a prospective large-scale multi-centre survey including molecular studies should be conducted to validate our results. In conclusion, this matched caseecontrolecontrol study identified that previous fluoroquinolones use, ICU stay, and previous exposure to different classes of antibiotics, were independent risk factors for LRSM occurrence. LRSM isolates showed multiple-drug resistance to potential active antibiotics in addition to levofloxacin. Careful antibiotics use, especially of levofloxacin, is necessary to combat LRSM dissemination. Conflict of interest statement None declared. Funding sources This work was supported by grants from the Tri-Service General Hospital Penghu Branch (TSGH-PH-107-02) and the Tri-Service General Hospital, National Defense Medical Center (TSGH-C103-187, TSGH-C104-195).
References [1] Brooke JS. Stenotrophomonas maltophilia: an emerging global opportunistic pathogen. Clin Microbiol Rev 2012;25:2e41. [2] Safdar A, Rolston KV. Stenotrophomonas maltophilia: changing spectrum of a serious bacterial pathogen in patients with cancer. Clin Infect Dis 2007;45:1602e9. [3] Falagas ME, Kastoris AC, Vouloumanou EK, Rafailidis PI, Kapaskelis AM, Dimopoulos G. Attributable mortality of Stenotrophomonas maltophilia infections: a systematic review of the literature. Future Microbiol 2009;4:1103e9. [4] Chang YT, Lin CY, Chen YH, Hsueh PR. Update on infections caused by Stenotrophomonas maltophilia with particular attention to resistance mechanisms and therapeutic options. Front Microbiol 2015;6:893.
[5] Abbott IJ, Slavin MA, Turnidge JD, Thursky KA, Worth LJ. Stenotrophomonas maltophilia: emerging disease patterns and challenges for treatment. Expert Rev Anti Infect Ther 2011;9:471e88. [6] Muder RR. Optimizing therapy for Stenotrophomonas maltophilia. Semin Respir Crit Care Med 2007;28:672e7. [7] Wang YL, Scipione MR, Dubrovskaya Y, Papadopoulos J. Monotherapy with fluoroquinolone or trimethoprimesulfamethoxazole for treatment of Stenotrophomonas maltophilia infections. Antimicrob Agents Chemother 2014;58:176e82. [8] Cho SY, Kang CI, Kim J, Ha YE, Chung DR, Lee NY, et al. Can levofloxacin be a useful alternative to trimethoprime sulfamethoxazole for treating Stenotrophomonas maltophilia bacteremia? Antimicrob Agents Chemother 2014;58:581e3. [9] Farrell DJ, Sader HS, Jones RN. Antimicrobial susceptibilities of a worldwide collection of Stenotrophomonas maltophilia isolates tested against tigecycline and agents commonly used for S. maltophilia infections. Antimicrob Agents Chemother 2010;54:2735e7. [10] Sader HS, Castanheira M, Mendes RE, Flamm RK. Frequency and antimicrobial susceptibility of Gram-negative bacteria isolated from patients with pneumonia hospitalized in ICUs of US medical centres (2015e17). J Antimicrob Chemother 2018;73:3053e9. [11] Gales AC, Seifert H, Gur D, Castanheira M, Jones RN, Sader HS. Antimicrobial susceptibility of Acinetobacter calcoaceticuse Acinetobacter baumannii complex and Stenotrophomonas maltophilia clinical isolates: results from the SENTRY Antimicrobial Surveillance Program (1997e2016). Open Forum Infect Dis 2019;6(Suppl 1):S34e46. [12] Rafailidis PI, Bliziotis IA, Falagas ME. Caseecontrol studies reporting on risk factors for emergence of antimicrobial resistance: bias associated with the selection of the control group. Microb Drug Resist 2010;16:303e8. [13] Harris AD, Karchmer TB, Carmeli Y, Samore MH. Methodological principles of caseecontrol studies that analyzed risk factors for antibiotic resistance: a systematic review. Clin Infect Dis 2001;32:1055e61. [14] Kaye KS, Engemann JJ, Mozaffari E, Carmeli Y. Reference group choice and antibiotic resistance outcomes. Emerg Infect Dis 2004;10:1125e8. [15] Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 1987;40:373e83. [16] Clinical and Laboratory Standards Institute. M100S 27. 27th ed. Wayne, PA: CLSI; 2017. [17] European Committee on Antimicrobial Susceptibility Testing (EUCAST). Breakpoint tables for interpretation of MICs and zone diameters. Version 7.1. Va ¨xjo ¨: EUCAST. 2017. Available at: http://www.eucast.org/ast_of_bacteria/previous_versions_of_ documents/[last accessed August 2019]. [18] Cha MK, Kang CI, Kim SH, Cho SY, Ha YE, Chung DR, et al. Emergence of fluoroquinolone-resistant Stenotrophomonas maltophilia in blood isolates causing bacteremia: molecular epidemiology and microbiologic characteristics. Diagn Microbiol Infect Dis 2016;85:210e2. [19] Chang LL, Chen HF, Chang CY, Lee TM, Wu WJ. Contribution of integrons, and SmeABC and SmeDEF efflux pumps to multidrug resistance in clinical isolates of Stenotrophomonas maltophilia. J Antimicrob Chemother 2004;53:518e21. [20] Shimizu K, Kikuchi K, Sasaki T, Takahashi N, Ohtsuka M, Ono Y, et al. Smqnr, a new chromosome-carried quinolone resistance gene in Stenotrophomonas maltophilia. Antimicrob Agents Chemother 2008;52:3823e5. [21] Garrison MW, Anderson DE, Campbell DM, Carroll KC, Malone CL, Anderson JD, et al. Stenotrophomonas maltophilia: emergence of multidrug-resistant strains during therapy and in an in vitro pharmacodynamic chamber model. Antimicrob Agents Chemother 1996;40:2859e64.
Please cite this article as: Wang CH et al., Levofloxacin-resistant Stenotrophomonas maltophilia: risk factors and antibiotic susceptibility patterns in hospitalized patients, Journal of Hospital Infection, https://doi.org/10.1016/j.jhin.2019.09.001
C.H. Wang et al. / Journal of Hospital Infection xxx (xxxx) xxx [22] Cheng AF, Li MK, Ling TK, French GL. Emergence of ofloxacinresistant Citrobacter freundii and Pseudomonas maltophilia after ofloxacin therapy. J Antimicrob Chemother 1987;20:283e5. [23] Pien CJ, Kuo HY, Chang SW, Chen PR, Yeh HW, Liu CC, et al. Risk factors for levofloxacin resistance in Stenotrophomonas maltophilia from respiratory tract in a regional hospital. J Microbiol Immunol Infect 2015;48:291e5. [24] Baek JH, Kim CO, Jeong SJ, Ku NS, Han SH, Choi JY, et al. Clinical factors associated with acquisition of resistance to levofloxacin in Stenotrophomonas maltophilia. Yonsei Med J 2014;55:987e93. [25] Kaye KS, Harris AD, Samore M, Carmeli Y. The caseecaseecontrol study design: addressing the limitations of risk factor studies for antimicrobial resistance. Infect Control Hosp Epidemiol 2005;26:346e51. [26] Drinka P, Niederman MS, El-Solh AA, Crnich CJ. Assessment of risk factors for multi-drug resistant organisms to guide empiric antibiotic selection in long term care: a dilemma. J Am Med Dir Assoc 2011;12:321e5. [27] Cardoso T, Ribeiro O, Aragao IC, Costa-Pereira A, Sarmento AE. Additional risk factors for infection by multidrug-resistant pathogens in healthcare-associated infection: a large cohort study. BMC Infect Dis 2012;12:375. [28] Pfaller MA, Huband MD, Streit JM, Flamm RK, Sader HS. Surveillance of tigecycline activity tested against clinical isolates from a global (North America, Europe, Latin America and Asia-Pacific) collection (2016). Int J Antimicrob Agents 2018;51:848e53.
7
[29] Pfaller MA, Flamm RK, Duncan LR, Mendes RE, Jones RN, Sader HS. Antimicrobial activity of tigecycline and cefoperazone/ sulbactam tested against 18,386 Gram-negative organisms from Europe and the Asia-Pacific region (2013e2014). Diagn Microbiol Infect Dis 2017;88:177e83. [30] Valdezate S, Vindel A, Sae ´z-Nieto JA, Baquero F, Canto ´n R. Preservation of topoisomerase genetic sequences during in vivo and in vitro development of high-level resistance to ciprofloxacin in isogenic Stenotrophomonas maltophilia strains. J Antimicrob Chemother 2005;56:220e3. [31] Garcia-Leon G, Salgado F, Oliveros JC, Sanchez MB, Martinez JL. Interplay between intrinsic and acquired resistance to quinolones in Stenotrophomonas maltophilia. Environ Microbiol 2014;16:1282e96. [32] Sanchez MB, Martinez JL. Overexpression of the efflux pumps SmeVWX and SmeDEF is a major cause of resistance to cotrimoxazole in Stenotrophomonas maltophilia. Antimicrob Agents Chemother 2018;62. AAC.00301e18. [33] Wang CH, Lin JC, Chang FY, Yu CM, Lin WS, Yeh KM. Risk factors for hospital acquisition of trimethoprimesulfamethoxazole resistant Stenotrophomonas maltophilia in adults: a matched caseecontrol study. J Microbiol Immunol Infect 2017;50:646e52. [34] Ko JH, Kang CI, Cornejo-Juarez P, Yeh KM, Wang CH, Cho SY, et al. Fluoroquinolones versus trimethoprimesulfamethoxazole for the treatment of Stenotrophomonas maltophilia infections: a systematic review and meta-analysis. Clin Microbiol Infect 2019;25:546e54.
Please cite this article as: Wang CH et al., Levofloxacin-resistant Stenotrophomonas maltophilia: risk factors and antibiotic susceptibility patterns in hospitalized patients, Journal of Hospital Infection, https://doi.org/10.1016/j.jhin.2019.09.001
Journal of Hospital Infection journal homepage: www.elsevier.com/locate/jhin
Levofloxacin-resistant Stenotrophomonas maltophilia: risk factors and antibiotic susceptibility patterns in hospitalized patients C.H. Wang a, C.-M. Yu b, S.-T. Hsu c, R.-X. Wu d, * a
Division of Infectious Diseases and Tropical Medicine, Department of Internal Medicine, Tri-Service General Hospital, National Defense Medical Center, Taipei, Taiwan b Department of Clinical Pathology, National Defense Medical Center, Tri-Service General Hospital, Taipei, Taiwan c Infection Control Office, Tri-Service General Hospital, National Defense Medical Center, Taipei, Taiwan d Division of Infectious Diseases, Tri-Service General Hospital Penghu Branch, Penghu, Taiwan
A R T I C L E
I N F O
Article history: Received 3 August 2019 Accepted 2 September 2019 Available online xxx Keywords: Risk Resistance Stenotrophomonas maltophilia Levofloxacin Multiple-drug resistance
S U M M A R Y
Background: Levofloxacin has been considered as an alternative treatment for Stenotrophomonas maltophilia infection. However, levofloxacin-resistant S. maltophilia (LRSM) are emerging worldwide. Aim: To investigate LRSM risk factors in hospitalized patients and to determine antibiotic susceptibility patterns of LRSM isolates. Methods: In a retrospective matched caseecontrolecontrol study, LRSM patients (the case group) were compared with two control groups: levofloxacin-susceptible S. maltophilia (LSSM) patients (control group A) and non-S. maltophilia-infected patients (control group B). Conditional logistic regression was used to analyse risk factors for LRSM occurrence. Tigecycline, ceftazidime, colistin, and trimethoprim/sulfamethoxazole (TMP/SMX) susceptibilities in collected LRSM clinical isolates were determined. Findings: A total of 105 LRSM, 105 LSSM, and 105 non-S. maltophilia-infected patients were analysed. The first multivariate analysis (cases vs group A) revealed that previous fluoroquinolones use was significantly associated with LRSM occurrence, and the second multivariate analysis (cases vs group B) revealed that previous fluoroquinolone use, previous intensive care unit stay, and the number of previous exposures to different classes of antibiotics were significantly associated with LRSM occurrence. Of all the LRSM isolates tested for antibiotic susceptibility, ceftazidime, TMP/SMX, tigecycline, and colistin resistance rates were 42.0, 99.0, 78.0, and 40.0%, respectively. Conclusion: LRSM antibiotic susceptibility patterns revealed multiple-drug resistance, which further limits treatment options for clinicians. To reduce LRSM occurrence, proper use of antibiotics, especially fluoroquinolones, is mandatory. ª 2019 The Healthcare Infection Society. Published by Elsevier Ltd. All rights reserved.
* Corresponding author. Address: Division of Infectious Diseases, Tri-Service General Hospital Penghu Branch, Penghu, No. 90, Qianliao 880, Magong City, Penghu County, Taiwan. Tel.: þ886 6 9211116; fax: þ886 2 9211117. E-mail address: [email protected] (R.-X. Wu). https://doi.org/10.1016/j.jhin.2019.09.001 0195-6701/ª 2019 The Healthcare Infection Society. Published by Elsevier Ltd. All rights reserved. Please cite this article as: Wang CH et al., Levofloxacin-resistant Stenotrophomonas maltophilia: risk factors and antibiotic susceptibility patterns in hospitalized patients, Journal of Hospital Infection, https://doi.org/10.1016/j.jhin.2019.09.001
2
C.H. Wang et al. / Journal of Hospital Infection xxx (xxxx) xxx
Introduction
Study population
Stenotrophomonas maltophilia is an emerging multidrugresistant nosocomial pathogen contracted in hospital wards [1]. It mainly affects immunocompromised or severely debilitated patients and contributes to a variety of infections, including pneumonia, bacteraemia, and skin and soft tissue infections, with varying clinical presentations [2]. The mortality rate attributed to S. maltophilia infection can be as high as 37.5%, and it is difficult to manage because it exhibits high-level intrinsic resistance to a variety of structurally unrelated antibiotics, including b-lactams, aminoglycosides, and carbapenems, drastically reducing available antibiotic treatment options [3e5]. Historically, trimethoprimesulfamethoxazole (TMP-SMX) has been used as a first-line S. maltophilia infection treatment [6]. However, when faced with TMP-SMX-resistant S. maltophilia infection or TMP-SMX-intolerant patients, levofloxacin is used as an alternative, based on data from recent studies showing that its clinical efficacy is similar to that of TMP-SMX [7,8]. Recently, however, there has been an increase in the number of reports on the prevalence of levofloxacin resistance in S. maltophilia [9e11]. This may further limit treatment options for clinicians because of the inadequate data on the susceptibility of other possible in-vitro active antibiotics, beyond TMP-SMX and levofloxacin. Understanding the risk factors for levofloxacin resistance in S. maltophilia may help clinicians identify possible resistant strains, thereby reducing the risks of inappropriate empirical antibiotic resistance and facilitating the implementation of appropriate infection control measures. An unmatched caseecontrol study is commonly used for examining antibiotics as risk factors for the emergence of antimicrobial resistance. Nevertheless, unequal sample size between case and control groups, some confounding factors not controlled before analysis, and selection of patients with a susceptible bacterium as the only control groups, will distort measurement of association [12e14]. Therefore, we conducted this matched caseecontrolecontrol study to investigate risk factors for levofloxacin resistance in S. maltophilia in adult inpatients, as has been suggested [12]. In addition, susceptibility tests on some antibiotics, which, in previous studies, have demonstrated in-vitro activity against collected clinical levofloxacin-resistant S. maltophilia, were also performed [4].
This caseecontrolecontrol study involved the construction of two separate caseecontrol studies using one case group and two control groups (control group A and B). Between January 1st, 2014 and June 30th, 2015, S. maltophilia-infected patients were identified using the computerized medical records system. Patients identified within 48 h of admission and those aged <18 years were excluded. Levofloxacin-resistant S. maltophilia (LRSM) isolates included those which exhibited in-vitro resistance or intermediate susceptibility against levofloxacin. The case group consisted of inpatients for whom the presence of the LRSM isolate was first determined in clinical samples. Control group A consisted of patients with levofloxacin-susceptible S. maltophilia (LSSM) matched with the case group for age (within five years), sex, and infection site at 1:1 ratio. Similarly, control group B consisted of nonS. maltophilia-infected patients, randomly selected from the inpatient population matched with the case group for age (within five years), sex, and time at risk at 1:1 ratio. Time at risk, which represented the duration of the at-risk period for antibiotic-resistant organism acquisition, was defined as the number of days between the admission date and the time when the culture first tested positive for S. maltophilia in the case group and control group A. For control group B, it was defined as the number of days between hospital admission and discharge. Once an eligible pool of controls was established, they were randomly assigned using Microsoft Excel 2013 (Microsoft Corp., Redmond, WA, USA).
Methods Study design The study was designed to determine the risk factors for levofloxacin resistance in S. maltophilia in adult inpatients. The matched caseecontrolecontrol study was conducted between January 1st, 2014 and June 30th, 2015 in Tri-Service General Hospital in Taiwan. The hospital has 1100 beds and is a tertiary referral centre in Northern Taiwan. The study was approved by the Institutional Review Board of the TriService General Hospital (TSGHIRB approval number: 1103-05-154).
Data collection and definitions Using the computerized medical records system, clinical information was systematically collected from identified patients, and then matched with the controls. The collected information included demographic data, underlying diseases on admission, clinical outcomes, hospitalization-related factors, recent therapeutic measures, previous antibiotics exposure, and indwelling medical devices. The Charlson comorbidity index (CCI) was used as an aggregate measure of underlying diseases to indicate comorbidity burden [15]. Previous antibiotics exposure in the case group and control group A patients was defined as: 24 h therapy within two weeks prior to S. maltophilia isolation. In non-infected patients (control group B), previous antimicrobials exposure was defined as: 24 h therapy within two weeks prior to discharge index. Recent chemotherapy and surgery performed within 30 days before S. maltophilia isolation were recorded in the case group and control group A. In control group B (no S. maltophilia isolation), recent therapeutic measures were defined as recent chemotherapy and surgery within 30 days before discharge index. Indwelling invasive medical devices were also recorded.
Microbiology test Between January 1st, 2014 and June 30th, 2015, a total of 100 randomly collected levofloxacin non-susceptible clinical S. maltophilia isolates were tested for antibiotic susceptibility. The Vitek 2 automatic system for Gram-negative rods (bioMe ´rieux, Marcy l’Etoile, France) was used to identify
Please cite this article as: Wang CH et al., Levofloxacin-resistant Stenotrophomonas maltophilia: risk factors and antibiotic susceptibility patterns in hospitalized patients, Journal of Hospital Infection, https://doi.org/10.1016/j.jhin.2019.09.001
C.H. Wang et al. / Journal of Hospital Infection xxx (xxxx) xxx S. maltophilia isolates and to assess susceptibility according to the 2017 Clinical and Laboratory Standards Institute (CLSI) M100-S27 performance standards [16]. Tigecycline breakpoints were established according to the European Committee on Antimicrobial Susceptibility Testing for Enterobacteriaceae (1 mg/mL, susceptibility; 2.0 mg/mL, intermediate; and >2 mg/mL, resistance). Colistin breakpoints were established according to CLSI criteria for P. aeruginosa (2 mg/mL, susceptibility; 4.0 mg/mL, intermediate; and 8 mg/mL, resistance) [17].
Statistical analysis All data were analysed using SPSS v16.0 (SPSS Inc., Chicago, IL, USA). The results are presented as mean (standard deviation) or median (range), whereas categorical variables are presented as percentages. Bivariate analysis of categorical variables was conducted using the c2-test or Fisher’s exact test, as deemed appropriate, whereas Student’s t-test or ManneWhitney U-test (for data not normally distributed) was used to analyse continuous variables. The normality of continuous variables was verified using the KolmogoroveSmirnov test. Variables with P < 0.05 on bivariate analysis were included in the forward stepwise conditional multivariate analysis logistic regression model. Two multivariate models were produced from the data obtained. The first model compared the case group with control group A, and the second model compared the case group with control group B. All tests were two-tailed, and P < 0.05 was considered statistically significant.
Results Study population Between January 1st, 2014 and June 30th, 2015, the cultures of 809 patients tested positive for S. maltophilia; the most
3
frequent isolation sites were the respiratory tract (85.8%), followed by blood (4.7%), wound tissue (3.6%), abdominal ascites (2.6%), urine (2.2%), and central venous catheter (1.1%). Further, the isolates were mostly identified in intensive care unit (ICU) patients (48.2%), followed by medical ward (30.7%), surgical ward (10.9%), and respiratory centre (10.2%) patients; of the 809 patients, 133 were levofloxacin resistant (16.4%). Of the 133 LRSM patients, 24 with LRSM isolation within 48 h after admission were excluded, as well as four other patients, because of unsuccessful matching. Therefore, a total of 105 LRSM patients were included, with infection sites as follows: 101, respiratory tract; one, blood; one, urine; and two, wound. To analyse the risk factors associated with LRSM isolation, 105 LSSM patients and 105 nonS. maltophilia-infected patients were included for subsequent analyses. Patient characteristics, including demographics, underlying disease, and clinical outcomes, are shown in Table I.
Risk factor analysis: resistant case group vs control group A As shown in Table II, bivariate analysis revealed that the differences in hospitalization-related factors, CCI, recent therapeutic measures, and indwelling medical devices of the case group and control group A were not statistically significant (P > 0.05). With regard to previous antibiotics exposure, fluoroquinolones exposure (43.8% vs 8.6%; P < 0.001) was associated with LRSM occurrence, whereas previous fourth-generation cephalosporins exposure was inversely associated with LRSM occurrence (28.6% vs 42.9%, P ¼ 0.042). In the case group, of all the inpatients who received fluoroquinolones (N ¼ 46), 33, nine, and four had previously received levofloxacin, ciprofloxacin, and moxifloxacin, respectively, and of all the control group A patients receiving fluoroquinolones, five, two, and two had
Table I Stenotrophomonas maltophilia patient characteristics Variable
Demographics Age (years), median (range) Male Underlying disease Cerebrovascular disease Congestive heart failure Renal insufficiency Chronic obstructive pulmonary disease Liver cirrhosis Diabetes mellitus Connective tissue disease Malignancy Clinical outcome In-hospital mortality Hospital stay after isolation, median (range) a b c
Case groupa (N ¼ 105)
Control group Ab (N ¼ 105)
Control group Bc (N ¼ 105)
80 (21e95) 72 (68.5%)
80 (18e93) 72 (68.5%)
78 (26e96) 72 (78.5%)
47 (44.8%) 19 (18.1%) 35 (33.3%) 18 (17.1%) 4 (3.8%) 41 (39.0%) 6 (5.7%) 22 (21.0%)
37 13 24 23 4 36 1 26
24 (22.9%) 8 (7.6%) 27 (25.7%) 15 (14.3%) 10 (9.5%) 39 (37.1%) 3 (2.9%) 34 (32.4%)
22 (20.9%) 22 (1e362)
37 (35.2%) 23 (0e305)
(35.2%) (12.4%) (22.9%) (21.9%) (3.8%) (34.3%) (0.95%) (24.8%)
11 (10.4%) e
Patients with levofloxacin-resistant S. maltophilia (case group). Patients with levofloxacin-susceptible S. maltophilia (control group A). Patients without S. maltophilia infection (control group B).
Please cite this article as: Wang CH et al., Levofloxacin-resistant Stenotrophomonas maltophilia: risk factors and antibiotic susceptibility patterns in hospitalized patients, Journal of Hospital Infection, https://doi.org/10.1016/j.jhin.2019.09.001
4
C.H. Wang et al. / Journal of Hospital Infection xxx (xxxx) xxx
Table II Bivariate analysis of levofloxacin-resistant S. maltophilia isolation risk factors Case groupa (N ¼ 105) Control group Ab (N ¼ 105) Control group Bc (N ¼ 105) P-valued
Factors related to hospitalization Time at riskf Recent admission within 3 months Previous ICU stay Charlson comorbidity index Recent therapeutic measures Chemotherapy Surgery Previous antibiotics exposure Macrolides Aminoglycosides Penicillin-lactamase inhibitors Third-generation cephalosporins Fourth-generation cephalosporins Carbapenems Glycopeptides Fluoroquinolones Trimethoprimesulfamethoxazole Colistin Otherg No. of different classes of previous antibiotic treatment Previous indwelling invasive medical devices Central venous catheterh Nasogastric tube Urinary catheter Surgical drain Ventilator
25.0 35 84 4.0
(3e112) (33.3) (80.0) (0e11)
3 (2.9) 30 (28.6) 10 7 36 21 30 40 32 46 3 8 9 2.0
50 97 76 15 72
(9.5) (6.7) (34.3) (20.0) (28.6) (38.1) (30.5) (43.8) (2.9) (7.6) (8.5) (0e6.0)
(47.6) (92.4) (72.4) (14.3) (68.6)
23.0 34 80 3.0
(5e106) (32.4) (76.2) (0e11)
8 (7.6) 27 (25.7) 9 6 29 31 45 50 28 9 4 8 11 2.0
52 98 82 18 76
(8.6) (5.7) (27.6) (29.5) (42.9) (47.6) (26.7) (8.6) (3.8) (7.6) (10.4) (0e6.0)
(49.5) (93.3) (78.1) (17.1) (72.4)
P-valuee
e 35 (33.3) 30 (28.6) 3.0 (0e14)
0.340 0.884 0.506 0.089
e 1.000 <0.001 0.117
9 (8.6) 24 (22.9)
0.147 0.631
0.099 0.332
0.618 0.594 0.319 0.135 0.042 0.184 0.556 <0.001 0.706 1.000 0.819 0.328
0.054 0.069 0.338 0.226 0.014 <0.001 0.001 0.001 1.000 0.258 0.126 <0.001
0.789 0.782 0.306 0.565 0.547
0.024 <0.001 0.001 0.435 <0.001
3 1 33 15 14 15 3 22 3 4 0 1.0
34 57 50 19 24
(2.9) (0.95) (31.4) (14.3) (13.3) (14.3) (2.9) (21.0) (2.9) (3.8) (0) (0.0e5.0)
(32.4) (54.3) (47.6) (18.1) (22.9)
ICU, intensive care unit. a Patients with levofloxacin-resistant S. maltophilia (case group). b Patients with levofloxacin-susceptible S. maltophilia (control group A). c Patients without S. maltophilia infection (control group B). d Comparison of case group and control group A patients. e Comparison of case group and control group B patients. f Days of stay prior to S. maltophilia isolation. g Other antimicrobials, including tigecycline, linezolid, and daptomycin. h Including double-lumen catheter for haemodialysis, central venous catheter, and peripherally inserted central catheter.
previously received levofloxacin, ciprofloxacin, and moxifloxacin, respectively. Further multivariate analysis with forward stepwise conditioning logistic regression showed that only previous use of fluoroquinolones remained in the final models and, thus, was considered a levofloxacin-resistance acquisition risk factor (odds ratio (OR): 7.17; 95% confidence interval (CI): 3.05e16.84; P < 0.001).
Risk factor analysis: resistant case group vs control group B (non-S. maltophilia-infected group) As shown in Table II, bivariate analysis of hospitalizationrelated factors revealed that patients in the case group showed a higher risk of recent ICU admission compared with those in control group B (80.0% vs 28.6%, P < 0.001). There were no statistically significant differences in the CCI and recent therapeutic measures of the two groups (P > 0.05).
Table III Multivariate analysis of levofloxacin-resistant Stenotrophomonas maltophilia isolation risk factors Variable
OR (95% CI)
P-value
Case groupa vs control group Ab Fluoroquinolones 7.17 (3.05e16.84) <0.001 Case group vs control group Bc Previous ICU stay 15.25 (3.66e63.57) <0.001 No. of different classes of 2.82 (1.577e5.05) <0.001 previous antibiotic treatment Fluoroquinolones 3.41 (1.05e11.06) 0.041 OR, odds ratio; CI, confidence interval; ICU, intensive care unit. a Patients with levofloxacin-resistant S. maltophilia (case group). b Patients with levofloxacin-susceptible S. maltophilia (control group A). c Patients without S. maltophilia infection (control group B).
Please cite this article as: Wang CH et al., Levofloxacin-resistant Stenotrophomonas maltophilia: risk factors and antibiotic susceptibility patterns in hospitalized patients, Journal of Hospital Infection, https://doi.org/10.1016/j.jhin.2019.09.001
C.H. Wang et al. / Journal of Hospital Infection xxx (xxxx) xxx
5
Table IV Susceptibility of 100 levofloxacin-resistant Stenotrophomonas maltophilia isolates to different antibiotics Antimicrobial agent
Ceftazidime Levofloxacin Trimethoprimesulfamethoxazole Tigecyclinea Colistinb
MIC range (mg/L)
1 to 64 4 to 8 1 to 16 0.5 to 8 0.5 to 16
No. of indicated susceptibility category Susceptible
Intermediate
Resistant
49 0 1 10 58
9 36 e 12 2
42 64 99 78 40
MIC, minimum inhibitory concentration. a Tigecycline breakpoints established by the European Committee on Antimicrobial Susceptibility Testing for Enterobacteriaceae (1 mg/mL, susceptibility; 2.0 mg/mL, intermediate; and >2 mg/mL, resistance). b Colistin breakpoints established by the Clinical and Laboratory Standards Institute for P. aeruginosa (2 mg/mL, susceptibility; 4.0 mg/mL, intermediate; and 8 mg/mL, resistance).
Regarding previous antibiotics exposure, patients in the case group were more likely to be exposed to different classes of antibiotics compared with those in control group B (2.0 vs 1.0, P < 0.001). Regarding previous exposure to the different classes of antibiotics recorded in this study, patients in the case group had higher rates for receiving fourth-generation cephalosporins, carbapenems, glycopeptides, and fluoroquinolones than those in control group B (28.6% vs 13.3%, 38.1% vs 14.3%, 30.5% vs 2.9%, and 43.8% vs 21.0 %; P ¼ 0.014, <0.001, 0.001, and 0.001, respectively). Of all the patients in control group B receiving fluoroquinolones (N ¼ 22), 11, eight, and three had previously received levofloxacin, ciprofloxacin, and moxifloxacin, respectively. In addition, patients in the case group were likely to have invasive medical devices, including central venous catheters, nasogastric tubes, urinary catheters, and ventilators, compared with those in control group B (47.6% vs 32.4%, 92.4% vs 54.3%, 72.4% vs 47.6%, and 68.6% vs 22.9%, P ¼ 0.024, <0.001, 0.001, and <0.001, respectively). Further multivariate analysis with forward stepwise conditional logistic regression after adding significant variables (P < 0.05) in the aforementioned bivariate analysis was performed. The factors that were independently predictive of levofloxacin resistance occurrence were: previous use of fluoroquinolones (OR: 3.41; 95% CI: 1.05e11.06; P ¼ 0.04), recent ICU admission (15.25; 3.66e63.57; P < 0.001), and the number of previous exposures to different classes of antibiotics (2.82; 1.58e5.50; P < 0.001) (Table III).
Microbiological results Of the 100 LRSM clinical isolates, 24, six, 69, and one were obtained from blood, wound tissue, the respiratory tract, and urine, respectively. Ninety-nine isolates were resistant to TMP-SMX, 78 showed concomitant resistance to tigecycline, and 17 were resistant to all the tested antibiotics, whereas colistin showed greater invitro activity than the other tested antibiotics (its susceptibility was only 58%) (Table IV).
Discussion The main finding of this study was that previous fluoroquinolones use, ICU stay, and previous exposure to different antibiotic classes were independent risk factors for subsequent
LRSM occurrence. Additionally, the study revealed the high resistance rates to other potential in-vitro active antibiotics in LRSM isolates. Previous studies on the mechanism of quinolone resistance in S. maltophilia showed various mechanisms, including multidrug efflux pump systems overexpression, DNA gyrase protection proteins encoded by chromosomally quinoloneresistance determinants, and chromosomal mutations of the quinolone-resistance-determining regions in DNA gyrase or DNA topoisomerase IV [18e20]. It has also been suggested that the selective pressure of quinolones may contribute to the subsequent occurrence of quinolone-resistant S. maltophilia [21,22]. Despite extensive experimental studies on the quinolone-resistance mechanism in S. maltophilia, only two clinical studies regarding risk factors for levofloxacin-resistant S. maltophilia isolation have been reported [23,24]. In both studies, the number of levofloxacin-resistant S. maltophilia cases included was limited, which may have decreased the statistical strength of the study, resulting in possible type II errors. Furthermore, the inclusion of patients infected with susceptible organisms as the control group may have led to biased estimates of relative risk [25]. Our study, which was based on a modified caseecontrolecontrol study design with a substantial number of cases, revealed that previous fluoroquinolones use, previous ICU stay, and previous exposure to multiple antibiotic classes were risk factors for subsequent levofloxacin-resistant S. maltophilia isolation. Besides, our study results also provide clinical data to support experimental studies on the association between fluoroquinolones exposure and acquired resistance to levofloxacin in S. maltophilia. As demonstrated by previous studies, this study revealed that other risk factors, including previous ICU stay, and previous multiple classes of antibiotics exposure, were also risk factors for multidrug-resistant pathogen isolation [26,27]. Therefore, the implementation of antibiotic stewardship programmes, especially the careful use of fluoroquinolones, is mandatory to halt LRSM dissemination. The susceptibility tests performed in this study revealed that no tested antibiotics, including tigecycline, showed good in-vitro activity against LRSM isolates. This result was unexpected, since in a recent epidemiology survey, tigecycline showed great potency in vitro [28,29]. This discrepancy might have resulted from the mechanism of quinolone resistance in S. maltophilia. Different from other bacteria, quinolone-resistant S. maltophilia of clinical origin do not
Please cite this article as: Wang CH et al., Levofloxacin-resistant Stenotrophomonas maltophilia: risk factors and antibiotic susceptibility patterns in hospitalized patients, Journal of Hospital Infection, https://doi.org/10.1016/j.jhin.2019.09.001
6
C.H. Wang et al. / Journal of Hospital Infection xxx (xxxx) xxx
have mutations in topoisomerases and overproduction of multidrug efflux pumps may be the major mechanism of quinolone resistance [4,30,31]. S. maltophilia with multidrug efflux pump overexpression may not only be resistant to quinolones but also to other antibiotics [32,33]. Recent studies comparing the outcomes of TMP/SMX- and fluoroquinolone-treated S. maltophilia-infected patients showed that the two drugs had an equal clinical efficacy, although the incidence of adverse events was lower in the fluoroquinolone-treated patients than in the TMP/SMXtreated group [7,8]. Further, a systematic review by Ko et al. suggested the superiority of fluoroquinolones to TMPSMX in the treatment of S. maltophilia infections [34]. The results of our study, however, identified the worrisome consequence of possible multidrug-resistant S. maltophilia occurrence after quinolones exposure. Therefore, empirical levofloxacin use in treating S. maltophilia infection should be exercised with more caution. This study had several limitations. First, given that it was retrospective in nature, it might have resulted in selection and observational bias. Second, molecular studies on LRSM isolates were not performed. Therefore, the molecular mechanism of the quinolone resistance in LRSM isolates could not be verified. Finally, this was a single-centre study in Taiwan; hence, the results may not be generalizable. Thus, in future, a prospective large-scale multi-centre survey including molecular studies should be conducted to validate our results. In conclusion, this matched caseecontrolecontrol study identified that previous fluoroquinolones use, ICU stay, and previous exposure to different classes of antibiotics, were independent risk factors for LRSM occurrence. LRSM isolates showed multiple-drug resistance to potential active antibiotics in addition to levofloxacin. Careful antibiotics use, especially of levofloxacin, is necessary to combat LRSM dissemination. Conflict of interest statement None declared. Funding sources This work was supported by grants from the Tri-Service General Hospital Penghu Branch (TSGH-PH-107-02) and the Tri-Service General Hospital, National Defense Medical Center (TSGH-C103-187, TSGH-C104-195).
References [1] Brooke JS. Stenotrophomonas maltophilia: an emerging global opportunistic pathogen. Clin Microbiol Rev 2012;25:2e41. [2] Safdar A, Rolston KV. Stenotrophomonas maltophilia: changing spectrum of a serious bacterial pathogen in patients with cancer. Clin Infect Dis 2007;45:1602e9. [3] Falagas ME, Kastoris AC, Vouloumanou EK, Rafailidis PI, Kapaskelis AM, Dimopoulos G. Attributable mortality of Stenotrophomonas maltophilia infections: a systematic review of the literature. Future Microbiol 2009;4:1103e9. [4] Chang YT, Lin CY, Chen YH, Hsueh PR. Update on infections caused by Stenotrophomonas maltophilia with particular attention to resistance mechanisms and therapeutic options. Front Microbiol 2015;6:893.
[5] Abbott IJ, Slavin MA, Turnidge JD, Thursky KA, Worth LJ. Stenotrophomonas maltophilia: emerging disease patterns and challenges for treatment. Expert Rev Anti Infect Ther 2011;9:471e88. [6] Muder RR. Optimizing therapy for Stenotrophomonas maltophilia. Semin Respir Crit Care Med 2007;28:672e7. [7] Wang YL, Scipione MR, Dubrovskaya Y, Papadopoulos J. Monotherapy with fluoroquinolone or trimethoprimesulfamethoxazole for treatment of Stenotrophomonas maltophilia infections. Antimicrob Agents Chemother 2014;58:176e82. [8] Cho SY, Kang CI, Kim J, Ha YE, Chung DR, Lee NY, et al. Can levofloxacin be a useful alternative to trimethoprime sulfamethoxazole for treating Stenotrophomonas maltophilia bacteremia? Antimicrob Agents Chemother 2014;58:581e3. [9] Farrell DJ, Sader HS, Jones RN. Antimicrobial susceptibilities of a worldwide collection of Stenotrophomonas maltophilia isolates tested against tigecycline and agents commonly used for S. maltophilia infections. Antimicrob Agents Chemother 2010;54:2735e7. [10] Sader HS, Castanheira M, Mendes RE, Flamm RK. Frequency and antimicrobial susceptibility of Gram-negative bacteria isolated from patients with pneumonia hospitalized in ICUs of US medical centres (2015e17). J Antimicrob Chemother 2018;73:3053e9. [11] Gales AC, Seifert H, Gur D, Castanheira M, Jones RN, Sader HS. Antimicrobial susceptibility of Acinetobacter calcoaceticuse Acinetobacter baumannii complex and Stenotrophomonas maltophilia clinical isolates: results from the SENTRY Antimicrobial Surveillance Program (1997e2016). Open Forum Infect Dis 2019;6(Suppl 1):S34e46. [12] Rafailidis PI, Bliziotis IA, Falagas ME. Caseecontrol studies reporting on risk factors for emergence of antimicrobial resistance: bias associated with the selection of the control group. Microb Drug Resist 2010;16:303e8. [13] Harris AD, Karchmer TB, Carmeli Y, Samore MH. Methodological principles of caseecontrol studies that analyzed risk factors for antibiotic resistance: a systematic review. Clin Infect Dis 2001;32:1055e61. [14] Kaye KS, Engemann JJ, Mozaffari E, Carmeli Y. Reference group choice and antibiotic resistance outcomes. Emerg Infect Dis 2004;10:1125e8. [15] Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 1987;40:373e83. [16] Clinical and Laboratory Standards Institute. M100S 27. 27th ed. Wayne, PA: CLSI; 2017. [17] European Committee on Antimicrobial Susceptibility Testing (EUCAST). Breakpoint tables for interpretation of MICs and zone diameters. Version 7.1. Va ¨xjo ¨: EUCAST. 2017. Available at: http://www.eucast.org/ast_of_bacteria/previous_versions_of_ documents/[last accessed August 2019]. [18] Cha MK, Kang CI, Kim SH, Cho SY, Ha YE, Chung DR, et al. Emergence of fluoroquinolone-resistant Stenotrophomonas maltophilia in blood isolates causing bacteremia: molecular epidemiology and microbiologic characteristics. Diagn Microbiol Infect Dis 2016;85:210e2. [19] Chang LL, Chen HF, Chang CY, Lee TM, Wu WJ. Contribution of integrons, and SmeABC and SmeDEF efflux pumps to multidrug resistance in clinical isolates of Stenotrophomonas maltophilia. J Antimicrob Chemother 2004;53:518e21. [20] Shimizu K, Kikuchi K, Sasaki T, Takahashi N, Ohtsuka M, Ono Y, et al. Smqnr, a new chromosome-carried quinolone resistance gene in Stenotrophomonas maltophilia. Antimicrob Agents Chemother 2008;52:3823e5. [21] Garrison MW, Anderson DE, Campbell DM, Carroll KC, Malone CL, Anderson JD, et al. Stenotrophomonas maltophilia: emergence of multidrug-resistant strains during therapy and in an in vitro pharmacodynamic chamber model. Antimicrob Agents Chemother 1996;40:2859e64.
Please cite this article as: Wang CH et al., Levofloxacin-resistant Stenotrophomonas maltophilia: risk factors and antibiotic susceptibility patterns in hospitalized patients, Journal of Hospital Infection, https://doi.org/10.1016/j.jhin.2019.09.001
C.H. Wang et al. / Journal of Hospital Infection xxx (xxxx) xxx [22] Cheng AF, Li MK, Ling TK, French GL. Emergence of ofloxacinresistant Citrobacter freundii and Pseudomonas maltophilia after ofloxacin therapy. J Antimicrob Chemother 1987;20:283e5. [23] Pien CJ, Kuo HY, Chang SW, Chen PR, Yeh HW, Liu CC, et al. Risk factors for levofloxacin resistance in Stenotrophomonas maltophilia from respiratory tract in a regional hospital. J Microbiol Immunol Infect 2015;48:291e5. [24] Baek JH, Kim CO, Jeong SJ, Ku NS, Han SH, Choi JY, et al. Clinical factors associated with acquisition of resistance to levofloxacin in Stenotrophomonas maltophilia. Yonsei Med J 2014;55:987e93. [25] Kaye KS, Harris AD, Samore M, Carmeli Y. The caseecaseecontrol study design: addressing the limitations of risk factor studies for antimicrobial resistance. Infect Control Hosp Epidemiol 2005;26:346e51. [26] Drinka P, Niederman MS, El-Solh AA, Crnich CJ. Assessment of risk factors for multi-drug resistant organisms to guide empiric antibiotic selection in long term care: a dilemma. J Am Med Dir Assoc 2011;12:321e5. [27] Cardoso T, Ribeiro O, Aragao IC, Costa-Pereira A, Sarmento AE. Additional risk factors for infection by multidrug-resistant pathogens in healthcare-associated infection: a large cohort study. BMC Infect Dis 2012;12:375. [28] Pfaller MA, Huband MD, Streit JM, Flamm RK, Sader HS. Surveillance of tigecycline activity tested against clinical isolates from a global (North America, Europe, Latin America and Asia-Pacific) collection (2016). Int J Antimicrob Agents 2018;51:848e53.
7
[29] Pfaller MA, Flamm RK, Duncan LR, Mendes RE, Jones RN, Sader HS. Antimicrobial activity of tigecycline and cefoperazone/ sulbactam tested against 18,386 Gram-negative organisms from Europe and the Asia-Pacific region (2013e2014). Diagn Microbiol Infect Dis 2017;88:177e83. [30] Valdezate S, Vindel A, Sae ´z-Nieto JA, Baquero F, Canto ´n R. Preservation of topoisomerase genetic sequences during in vivo and in vitro development of high-level resistance to ciprofloxacin in isogenic Stenotrophomonas maltophilia strains. J Antimicrob Chemother 2005;56:220e3. [31] Garcia-Leon G, Salgado F, Oliveros JC, Sanchez MB, Martinez JL. Interplay between intrinsic and acquired resistance to quinolones in Stenotrophomonas maltophilia. Environ Microbiol 2014;16:1282e96. [32] Sanchez MB, Martinez JL. Overexpression of the efflux pumps SmeVWX and SmeDEF is a major cause of resistance to cotrimoxazole in Stenotrophomonas maltophilia. Antimicrob Agents Chemother 2018;62. AAC.00301e18. [33] Wang CH, Lin JC, Chang FY, Yu CM, Lin WS, Yeh KM. Risk factors for hospital acquisition of trimethoprimesulfamethoxazole resistant Stenotrophomonas maltophilia in adults: a matched caseecontrol study. J Microbiol Immunol Infect 2017;50:646e52. [34] Ko JH, Kang CI, Cornejo-Juarez P, Yeh KM, Wang CH, Cho SY, et al. Fluoroquinolones versus trimethoprimesulfamethoxazole for the treatment of Stenotrophomonas maltophilia infections: a systematic review and meta-analysis. Clin Microbiol Infect 2019;25:546e54.
Please cite this article as: Wang CH et al., Levofloxacin-resistant Stenotrophomonas maltophilia: risk factors and antibiotic susceptibility patterns in hospitalized patients, Journal of Hospital Infection, https://doi.org/10.1016/j.jhin.2019.09.001