Stable antimicrobial resistance patterns of Mycobacterium abscessus complex antibiograms in Singapore from 2013 to 2022: a retrospective review

INTRODUCTION Nontuberculous mycobacteria (NTM) are opportunistic pathogens that can cause a wide spectrum of diseases in humans. The prevalence of NTM disease has been rising in recent years,[1] leading to its emergence as a global public health concern. One of the most clinically important NTM species is the rapidly growing mycobacteria Mycobacterium abscessus complex (MABC), comprising three subspecies, abscessus, bolletii and massiliense.[2–6] Mycobacteria Mycobacterium abscessus complex can cause a wide spectrum of diseases, including pulmonary infection, lymphadenitis, skin and soft tissue infection, ocular infection and disseminated infection.[2,5,7] The incidence and prevalence of MABC infection has been increasing in the last decade,[1,8] and it is the most frequently isolated NTM in Singapore.[7,9,10] Mycobacteria Mycobacterium abscessus complex is an extremely difficult pathogen to treat, requiring a multidrug combination regimen for protracted durations,[11] with disappointingly high rates of recurrence.[4,5,12,13] The choice of antibiotics is complicated by the high rates of intrinsic resistance in MABC to many antibiotics, along with its ability to develop adaptive resistance through changes in gene or protein expression, or acquired resistance through mutations in target genes.[3,4,12–16]M. abscessus subspecies bolletii and most subspecies abscessus isolates possess inducible macrolide resistance due to the expression of erm(41), an erythromycin ribosome methylase that modifies the 23S rRNA, conferring macrolide resistance.[2–6] Drug susceptibility testing (DST) of MABC is difficult and controversial. Currently, macrolides are the only oral agents that show a correlation between in vitro susceptibility and treatment response in patients with MABC pulmonary disease; for other antimicrobial agents, there is no established correlation between susceptibility and clinical response.[17] The use of companion antibiotics with uncertain efficacy, coupled with the widespread use of the same antibiotics for other indications,[18] raises concern that MABC resistance may be increasing over time. Infection with MABC strains resistant to key antibiotics such as macrolides can result in refractory disease, with poor outcomes and high mortality.[5,6,13] A previous study of MABC antimicrobial resistance in Singapore between 2006 and 2011 found that 90% of MABC isolates were susceptible to amikacin and 96% to clarithromycin.[10] Since then, updated information on the prevalance of antibiotic resistance in Singapore has been lacking. To address this gap, we conducted a retrospective review of MABC isolates obtained from 2013 to 2022 in our centre, aiming to assess antimicrobial resistance rates and identify any resistance trends over time. METHODS Laboratory records of MABC isolates from samples collected between 1 January 2013 and 31 December 2022 in Singapore General Hospital were included. Routine identification of isolates was carried out using matrix-assisted laser desorption ionisation time-of-flight mass spectrometry (Vitek MS; BioMérieux SA, Marcy-l’Étoile, France) or a reverse hybridisation line probe assay (INNO-LiPA Mycobacteria v2; Fujirebio Holdings Inc, Tokyo, Japan). Drug susceptibility testing was performed on clinician request for isolates assessed to be significant. Isolates for which susceptibility testing was performed were included. For patients with multiple isolates, only the first isolate with available DST data were included. Drug susceptibility testing was carried out by broth microdilution (RAPMYCO plates, Sensititre; Thermo Fisher Scientific, MA, Waltham, MA, USA), with Mycobacterium peregrinum for quality control and interpreted according to the Clinical and Laboratory Standards Institute (CLSI) guidelines.[19–21] The plates were incubated at 28–30°C, and the minimum inhibitory concentrations of amikacin, cefoxitin, ciprofloxacin, clarithromycin, doxycycline, imipenem, linezolid, moxifloxacin and trimethoprim–sulfamethoxazole (using 80% growth inhibition for trimethoprim–sulfamethoxazole) were determined after 3–5 days. The plates were further incubated for 14 days to detect inducible macrolide resistance for clarithromycin. Between 2013 and 2022, there were no significant changes in CLSI testing methodology or breakpoints for MABC. Graphical representation of data was performed on Prism (GraphPad Prism, GraphPad Software, Boston, Massachusetts, USA). Ethics approval was not required for this study (CIRB Ref: 2023/2045). RESULTS A total of 443 MABC isolates were identified, averaging 44 (range 31–61) isolates per year. Among these, 264 (59.6%) isolates were from pulmonary sites, including 204 from the sputum and 52 from bronchoalveolar lavage. Blood samples accounted for 21 (4.7%) isolates. The remaining 158 (35.7%) were from other extrapulmonary sites, including 19 from eye samples, eight from lymph nodes and nine from peritoneal dialysis catheters. Only nine (2.0%) isolates were identified to subspecies level, as this speciation was performed upon clinician request. The number and percentages of susceptible, intermediate and resistant isolates are shown in Table 1. Mycobacteria Mycobacterium abscessus complex showed the highest susceptibility to amikacin (95.0% of isolates susceptible), followed by clarithromycin (76.1%) and linezolid (23.0%); ≤5% of isolates were susceptible to the other antibiotics tested.Table 1: Antimicrobial susceptibilities of MABC isolates from 2013 to 2022 (N=443).The annual percentage of susceptible, intermediate and resistant isolates for each antibiotic is presented in Figure 1. For both clarithromycin and amikacin, the percentages remained steady over the decade. Clarithromycin resistance included both intrinsic and inducible resistance, as data was obtained after a 14-day incubation to detect inducible macrolide resistance. For imipenem, there was a noticeable trend: the percentage of resistant isolates decreased from 25.0% in 2013 to 3.9% in 2022, while the intermediate percentage increased from 72.5% in 2013 to 94.2% in 2022, and susceptible percentage remained unchanged. For other antibiotics, the susceptibility patterns remained stable, with over 90% of isolates showing resistance to ciprofloxacin, doxycycline, moxifloxacin and trimethoprim–sulfamethoxazole.Figure 1: Graphs show antimicrobial susceptibilites of Mycobacterium abscessus complex isolates from 2013 to 2022—percentages of isolates that are susceptible, intermediate or resistant to each antibiotic.DISCUSSION Mycobacteria Mycobacterium abscessus complex is one of the most prevalent and important causes of NTM disease worldwide.[1,8] Despite this, a standard treatment regimen remains elusive. Selecting a combination multidrug regimen based on DST, particularly for macrolides and amikacin, is recommended.[11] However, DST can take up to 2 weeks, making knowledge about local antimicrobial resistance patterns essential for empirical treatment in the interim. Our study, which analysed a large number of isolates over a decade, provides insight into resistance trends over time. We found no significant shifts or trends in resistance patterns to routinely tested antibiotics in MABC isolates from 2013 to 2022 in our centre. Notably, the percentages of susceptible, intermediate and resistant isolates remained similar to those reported in previous studies conducted in Singapore from 2006 to 2011[10] and from 2017 to 2019.[22] While clarithromycin resistance appeared to increase from 3% between 2006 and 2011 to 22.8% in our study, this increase is likely due to a change of CLSI testing methodology in 2011, when extended incubation with clarithromycin to detect inducible macrolide resistance was introduced.[20] Our study showed that clarithromycin resistance rates remained fairly constant from 2013 to 2022. The preservation and stability of high levels of susceptibility to macrolides and amikacin over this time frame are reassuring, supporting the continued use of these antibiotics as key components of MABC treatment. For imipenem, the percentage of resistant isolates appeared to decrease while that of intermediate isolates appeared to increase from 2013 to 2022, but the significance of this trend is currently unknown. There are several limitations in this retrospective study. Firstly, at our centre, MABC is identified to the subspecies level only on request. Therefore, we were not able to delineate antimicrobial resistance trends between the MABC subspecies. Secondly, susceptibility testing is typically only performed when treatment is being considered, potentially excluding isolates from patients on watchful waiting, or those deemed colonisers or contaminants. Thirdly, only the first isolate of each patient was included, which means this study does not account for the possible development of resistance during treatment. In addition, MABC isolates do not always reflect the diversity present within the patient, which could include subclones with differing antimicrobial resistance profiles.[16] Lastly, susceptibility testing for bedaquiline, clofazimine, eravacycline, omadacycline, tedizolid and tigecycline is not routinely carried out, limiting our study’s scope. Furthermore, while combination antimicrobial testing of MABC has helped to identify synergistic drug combinations[23] and could potentially be useful given the need for multidrug therapy in MABC treatment, a standardised method for testing and the relationship between in vitro results and clinical outcomes have not yet been established. If testing for these antibiotics and combination antimicrobial testing become more widespread, they should be included in future studies. This study offers some reassurance regarding the antimicrobial susceptibility trends in MABC, particularly for macrolides and amikacin. While specific data or recommendations remain lacking, caution should be exercised to mitigate the emergence of macrolide- or amikacin-resistant MABC strains.[6,13] Measures include avoiding the initiation of macrolide monotherapy for immunomodulatory purposes in patients with non-cystic fibrosis bronchiectasis with a possibility or history of macrolide-susceptible NTM lung disease, and avoiding functional monotherapy of MABC disease through the selection of poor companion antibiotics, such as macrolide-containing regimens without amikacin. Continuous surveillance and research will be vital in addressing this emerging public health threat. Acknowledgement We thank our research coordinator Ms Charlene Suk Teng Cheong, Department of Infectious Diseases, Singapore General Hospital; Ms Lynn May Yee Choy, Central TB Laboratory, Singapore General Hospital; and members of the Departments of Infectious Diseases and Microbiology, Singapore General Hospital, for their helpful discussions and support. Financial support and sponsorship Nil. Conflicts of interest There are no conflicts of interest.