Drug susceptibility distributions of Mycobacterium chimaera and other non-tuberculous mycobacteria.

SUSCEPTIBILITY Drug Susceptibility Distributions of Mycobacterium chimaera and Other Nontuberculous Mycobacteria Bettina Schulthess,a,b Daniel Schäfle,a Nicole Kälin,a,b Tamara Widmer,a,b Peter Sandera,b Institute of Medical Microbiology, University of Zurich, Zurich, Switzerland a National Reference Center for Mycobacteria, Zurich, Switzerland b Recent outbreaks of cardiac surgery-associated Mycobacterium chimaera infections have highlighted the importance of species differentiation within the Mycobacterium avium complex and pointed to a lack of antibiotic susceptibility data for M. chimaera. Using the MGIT 960/EpiCenter TB eXiST platform, we have determined antibiotic susceptibility patterns of 48 clinical M. chimaera isolates and 139 other nontuberculous mycobacteria, including 119 members of the M. avium complex and 20 Mycobacterium kansasii isolates toward clofazimine and other drugs used to treat infections with slow-growing nontuberculous mycobacteria (NTM). MIC50, MIC90, and tentative epidemiological cutoff (ECOFF) values for clofazimine were 0.5 mg/liter, 1 mg/liter, and 2 mg/liter, respectively, for M. chimaera. Comparable values were observed for other M. avium complex members, whereas lower MIC50 (#0.25 mg/liter), MIC90 (0.5 mg/liter), and ECOFF (1 mg/liter) values were found for M. kansasii. Susceptibility to clarithromycin, ethambutol, rifampin, rifabutin, amikacin, moxifloxacin, and linezolid was in general similar for M. chimaera and other members of the M. avium complex, but increased for M. kansasii. The herein determined MIC distributions, MIC90, and ECOFF values of clofazimine for M. chimaera and other NTM provide the basis for the definition of clinical breakpoints. Further studies are needed to establish correlation of in vitro susceptibility and clinical outcome. ABSTRACT KEYWORDS Mycobacterium chimaera, Mycobacterium avium complex, drug susceptibility testing, clofazimine, resistance, antibiotic resistance M ycobacterium chimaera is a slow-growing nontuberculous mycobacterium (NTM) that was established in 2004 as a new species within the Mycobacterium avium complex (1). In the past, the number of infections with M. chimaera was underestimated, as commercial mycobacterial identification systems such as line probe assays failed to identify M. chimaera to species level and thus classified M. chimaera as M. avium complex, M. avium, or Mycobacterium intracellulare (1, 2). M. chimaera is differentiated from other members of the M. avium complex by a unique 16S rRNA gene sequence and internal transcribed spacer (ITS) region (1). Recently, a global outbreak of cardiac surgery-associated M. chimaera infections highlighted the importance of species identification within the M. avium complex (3, 4). The outbreak was linked to contaminated water reservoirs of heater-cooler devices that spread M. chimaera by aerosols during open chest surgery (5, 6). Severe, disseminated M. chimaera infections with a high case fatality rate were observed (7). Due to the limited ability of commercial identification methods to adequately identify M. chimaera, few studies have reported drug susceptibility data on M. chimaera. Recent studies analyzed antimicrobial susceptibility of M. chimaera using a commercial microdilution system, the SLOWMYCO Sensititre panel from Trek Diagnostic Systems, and reported similar susceptibility patterns for M. chimaera as for other members of the M. avium complex (8–11). Recommended treatment options for disseminated M. May 2021 Volume 65 Issue 5 e02131-20 Antimicrobial Agents and Chemotherapy Citation Schulthess B, Schäfle D, Kälin N, Widmer T, Sander P. 2021. Drug susceptibility distributions of Mycobacterium chimaera and other nontuberculous mycobacteria. Antimicrob Agents Chemother 65:e02131-20. https://doi.org/10.1128/AAC.02131-20. Copyright © 2021 American Society for Microbiology. All Rights Reserved. Address correspondence to Bettina Schulthess, . Received 12 October 2020 Returned for modification 12 January 2021 Accepted 14 February 2021 Accepted manuscript posted online 22 February 2021 Published 19 April 2021 aac.asm.org 1 Schulthess et al. Antimicrobial Agents and Chemotherapy TABLE 1 Number and origin of NTM isolates included in this study Speciesa M. avium (MAC) M. intracellulare (MAC) M. chimaera (MAC) Other MAC M. kansasii aMAC, No. (%) of respiratory isolates 55 (69) 30 (97) 34 (71) 6 (75) 12 (60) No. (%) of nonrespiratory isolates 8 (10) 0 (0) 13 (27) 1 (12.5) 5 (25) No. (%) of isolates with unknown origin 17 (21) 1 (3) 1 (2) 1 (12.5) 3 (15) Total 80 31 48 8 20 M. avium complex. chimaera infections include combination therapy with macrolides, rifamycins, ethambutol, amikacin, and clofazimine (7). Clofazimine is not yet included in the SLOWMYCO Sensititre antibiotic panel and consequently clofazimine MIC data for M. chimaera are scarce. We have previously established automated quantitative drug susceptibility testing (DST) for slow-growing NTM using the MGIT 960/EpiCenter TB eXiST platform (12, 13). We here report on MIC distributions of clofazimine and other drugs used to treat NTM infections for 48 clinical M. chimaera isolates and 139 other nontuberculous mycobacteria, including 119 members of the M. avium complex and 20 Mycobacterium kansasii isolates. RESULTS M. chimaera clofazimine MIC distribution. MICs of clofazimine were determined for 48 clinical, nonduplicate M. chimaera isolates using the MGIT 960/EpiCenter TB eXiST system (Becton Dickinson, Sparks, MD). A clofazimine concentration range of 0.25 mg/liter to 4 mg/liter was tested in 2-fold serial dilutions. Out of 48 M. chimaera isolates, 34 (71%) were of respiratory origin and 13 (27%) isolates were of nonrespiratory origin (Table 1). For one isolate, the source was unknown. Clofazimine MIC values for M. chimaera ranged from #0.25 mg/liter to 2 mg/liter (Fig. 1A, Table 2). MIC50 and MIC90 values of 0.5 mg/liter and 1 mg/liter were determined. A tentative epidemiological cutoff (ECOFF) was set at 2 mg/liter by visual inspection of the MIC distribution (Fig. 1). The clofazimine MIC distribution of M. chimaera was compared to MIC distributions of 119 M. avium complex isolates, including M. avium (n = 80), M. intracellulare (n = 31), Mycobacterium yongonense (n = 3), Mycobacterium timonense (n = 2), Mycobacterium bouchedurhonense (n = 1), Mycobacterium colombiense (n = 1), and Mycobacterium vulneris (n = 1) (Fig. 1B to E, Table 2). The MIC range, MIC50, MIC90, and tentative ECOFF values of M. chimaera and other M. avium complex isolates were comparable. The clofazimine MIC distribution of M. kansasii, a slow-growing nontuberculous mycobacterium not related to the M. avium complex, showed lower MIC50 (#0.25 mg/liter), MIC90 (0.5 mg/liter), and tentative ECOFF value (1 mg/liter) compared to M. chimaera and other M. avium complex species (Fig. 1F). Susceptibility distributions of additional drugs used for the treatment of M. chimaera infections. Susceptibil (...truncated)