Epidemiology of Resistance Determinants Identified in Meropenem-Nonsusceptible Enterobacterales Collected as Part of a Global Surveillance Study, 2018 to 2019

ABSTRACT The objective of this study was to describe the frequency of resistance determinants in meropenem-nonsusceptible (MEM-NS) Enterobacterales isolates collected in 2018 and 2019 as a part of the ATLAS global surveillance program. Among a total of 39,368 Enterobacterales isolates collected in 2018 and 2019, 5.7% were MEM-NS (MIC ≥2 μg/mL). Among the different regions, the proportion of MEM-NS isolates ranged from 1.9% (North America) to 8.4% (Asia/Pacific). The majority of MEM-NS isolates collected were of the species Klebsiella pneumoniae (71.5%). Among the MEM-NS Enterobacterales isolates collected, metallo-β-lactamases (MBL) were identified in 36.7%, KPC in 25.5%, and OXA-48-like in 24.1%. The predominance of resistance mechanisms among MEM-NS isolates varied by region: MBLs were dominant in isolates collected in Africa and Middle East (AfME, 49%) and Asia/Pacific (59.4%), OXA-48-like carbapenemases were predominant in Europe (30%), and KPC in Latin America (51.9%) and North America (53.6%). NDM β-lactamases accounted for the majority of MBLs identified (88.4%). Of the 38 carbapenemase variants identified, NDM-1 (68.7%), KPC-2 (54.6%), OXA-48 (54.3%), and VIM-1 (76.1%) were the common variants within their respective families. Among the MEM-NS isolates, 7.9% co-carried two carbapenemases. Notably, the proportion of MEM-NS Enterobacterales increased from 4.9% in 2018 to 6.4% in 2019. The results of this study show a continuation of the trend of increasing carbapenem-resistance within clinical Enterobacterales with mechanisms of resistance varying across different regions. The existential threat to public health posed by the continued spread of nearly untreatable pathogens requires a multifaceted approach to prevent the collapse of modern medicine.

hyperproduction AmpC combined with mutations of porins can be involved as well (6)(7)(8). Carbapenemases produced by Enterobacterales have been identified among three out of four Ambler classes of b-lactamases-class A, class B, and class D. The most notable among class A enzymes include KPC and some variants of GES, while class D carbapenemases are most often those closely related to OXA-48. Both class A and D enzymes have serine residues at their active sites. Ambler class B enzymes are metallo-b-lactamases (MBLs) and have metal ion cofactors in their active sites (9). The most widespread MBLs among clinical Enterobacterales isolates include NDM, IMP, and VIM (1, 3,6). The spectrum of activity differs across enzymes of the different classes; KPCs hydrolyze all clinically available b-lactam agents but, can be inhibited by diazabicyclooctane b-lactamase inhibitors, MBLs hydrolyze all clinically available b-lactam agents except aztreonam, and are insensitive to clinically available b-lactamase inhibitors (BLIs), and OXA-48-like enzymes which generally hydrolyze penicillins and carbapenems, and are poorly inhibited by available BLIs (6). The OXA-48-type b-lactamases are now routinely encountered in infections caused by carbapenem-resistant Enterobacterales. These enzymes are of high and growing clinical significance due to the importance of carbapenems in treatment of health care-associated infections by Gram-negative bacteria, the wide and increasing dissemination of OXA-48 enzymes on plasmids, and the challenges posed by their detection (10).
A previous study assessed the distribution of b-lactamase resistance determinants in meropenem-nonsusceptible Enterobacterales among isolates collected between 2012 and 2017 (7). The current study is a continuation of the previous study with the objective of describing the epidemiology of b-lactamase resistance determinants in meropenem-nonsusceptible (MEM-NS) Enterobacterales among isolates collected in 2018 and 2019.
Similar to global proportions, NDM (99%, 297/300) was predominant among MBL genes identified among isolates collected in APAC. IMP was the only other MBL identified in isolates collected in APAC (Fig. 3C).
Comparison of carbapenemase carriage across organisms revealed that isolates of K.   Fig. 4A). Among the different geographical regions, the highest proportion of MEM-NS isolates carrying two carbapenemases were collected from APAC (22.4%, 113/505; Fig. 4).
Across all geographical regions, there was an increase in the proportion of MEM-NS isolates in this study compared to the previous study. In that previous study, in AfME, the number of MEM-NS isolates collected increased over 2.5 times (5.5% versus 1.97%) (7). The current study also identified a marked increase in the proportion of MEM-NS isolates from 2018 to 2019 across some regions: Latin America (4.7% versus 9.3%), APAC (6.8% versus 9.9%), and AfME (4.7% versus 6.3%; Fig. 4). Among isolates of species examined, the distribution of MEM-NS isolates in this study were in line with that of the previous study-K. pneumoniae (72.3% versus 76.7%, 2018 to 2019 versus 2012 to 2017), E. cloacae (8.8% versus 6.6%), and E. coli (6.7% versus 5.1%).
The majority of MEM-NS isolates assessed in this study carried MBLs (36.7%). This is in contrast to the previous study where the majority of the MEM-NS isolates carried KPC-type carbapenemase (47.4%) and only 20.6% carried MBLs (7). This increase in MBLs over the periods examined could be attributed to the marked increase in the proportion of MBL-positive isolates among the MEM-NS isolates in Latin America (2018 to 2019 versus 2012 to 2017: 30.7% versus 7.4%); North America (15.5% versus 4.9%), and Europe (27.8% versus 18.5%). The evident increase in the proportion of MBL-positive isolates among the MEM-NS isolates in the current study is consistent with several other studies that have also reported an increase and spread of MBLs globally and across different regions (1, 4, 11-13). Among all regions, the lowest frequency of MBLpositive isolates among the MEM-NS isolates was noted in North America in our study (15.5%), which is corroborated by another study (12). However, a marked increase in the frequency of MBLs among the MEM-NS isolates was also observed in North America from 2018 to 2019 in our study (10% versus 22%; Fig. 4). Overall, the proportion of MBL-positive isolates has been growing over the years which is of great concern due to their increasing resistance to carbapenems which were considered as the lastresort antimicrobials for treating infections (6). The majority of the MBLs in this study were identified in isolates of K. pneumoniae (58.9%), E. cloacae (14.8%), E. coli (10.2%),  and Providencia spp. (6.1%). These data are in line with that of the previous study for K. pneumoniae (49.5%), E. cloacae (19.5%), and E. coli (6.2%; Table 1). The previous study reported an increase in frequency of NDM-positivity, globally, and particularly in Latin America, Europe, and APAC, from 2012 to 2014 to 2015 to 2017 (6). Interestingly, the overall proportion of NDM-positive isolates among those that were MBLpositive, increased in this study compared to the previous study (2012 to 2017 versus 2018 to 2019: 61% versus 88.4%), suggesting a global spread of NDM carrying Enterobacterales (7). This is further supported by findings from other studies that have also indicated the increasing frequency of NDM carrying isolates worldwide (7,(14)(15)(16). In this study, NDM-positive isolates were detected in all regions and the highest proportion of these isolates were from APAC and AfME, similar to the previous study. Although the current study identified NDM-1 as the major NDM-variant, the proportion of isolates carrying NDM-1, among those that were NDM-positive, reduced compared to the previous study (2018 to 2019 versus 2012 to 2017: 68.7% versus 84.2%), and also from 2018 (70%) to 2019 (67.9%; Table 2). Notably, isolates carrying NDM-5 and NDM-7 were found mainly from AfME and APAC in this study which is corroborated by the previous study. Our study revealed an increase in proportion of these isolates, among those that were NDM-positive, compared to the previous study (AfME, 25.9% versus 17.6%; APAC 52.5% versus 31.3%) (7).

Frequency of Mem-NS isolates
In this study, the proportion of carbapenemase-positive isolates carrying OXA-48like b-lactamases globally was comparable between 2018 and 2019. However, among the different geographical regions, an increase was noted only in the proportion of isolates collected in APAC and AfME from 2018 to 2019 (Fig. 4). Interestingly, Kazmierzack et al. also noted an increase in the proportion of OXA-48-like from 2012 to 2014 to 2015 to 2017 globally and in APAC and AfME (7). Among the different OXA-48-like variants identified in this study, OXA-48 was the most frequently identified, which was in line with the previous study (7). Of note, In the current study, there was a higher proportion of carbapenemase-positive isolates carrying OXA-232 and OXA-181 in APAC and AfME compared to other regions. Interestingly, the previous study also reported a similar trend. OXA-232 accounted for more of the OXA-48-like b-lactamases (31.3%) than did OXA-181 (12.9%) in the current study. However, in the past years, among the isolates that carried OXA-48-like carbapenemases, the proportion of isolates carrying OXA-181 had been reported to be more than those with OXA-232 (17). It should also be noted that OXA-232 carbapenemases hydrolyze carbapenem substrates less efficiently than either OXA-181 or OXA-48 (18). This overall increase in the proportion of OXA-232, among the MEM-NS isolates that carried OXA-48-like carbapenemases evident in the current study, could be driven by increasing frequency in APAC (71.7%), which is further supported by reports of outbreaks in that region since 2015 (17).
The proportion of isolates co-carrying two carbapenemases among those that were MEM-NS was 7.8% in this study (Fig. 3A), up from 2.7% reported in the previous study. This increase may be attributed to the increase in the proportion of isolates co-carrying MBLs and OXA-48-like carbapenemases in the current study (2% versus 6.7%, 2012 to 2017 versus 2018 to 2019). The increasing frequency of isolates with dual carbapenemases further complicates strategies to detect and overcome antimicrobial resistance.
This study has limitations. The data from this study cannot be used for epidemiological assessment or as a marker for prevalence of resistance mechanisms as each site collected a predefined number of isolates of given species. There were variations in the number of participating centers between years as well as the distribution of centers in each region. Changes in participating centers and countries must be taken into consideration when evaluating regional trends as well as comparing trends with previous studies and interpretations must be drawn cautiously. Additionally, carbapenemase carriage was only assessed in MEM-NS isolates which could have resulted in failure to identify some carbapenemases, especially among isolates carrying OXA-48-like or other carbapenemases with low activity that do not consistently confer resistance to meropenem and potentially biasing the results toward resistance mechanisms more specific to meropenem than carbapenems in general. Isolates collected from China were excluded in the study as they were characterized differently from the rest of the isolates. Notably, this study, but not the study performed previously, included isolates collected in India. Isolates collected in India comprised 23.4% of isolates collected in APAC for this study yet accounted for 75.8% of MEM-NS isolates for that region (data not shown), skewing the frequency of carbapenemase carriage toward trends in what may be specific to that country.
Despite the limitations, the data presented in this study provides crucial insights on the distribution of MEM-NS Enterobacterales and the carbapenemases that cause resistance among these organisms. The results of this study show an increasing trend of MEM-NS Enterobacterales globally with mechanisms of resistance varying across different regions. Among the MEM-NS isolates, the rise in the proportion of those carrying MBLs, particularly NDM, and those co-carrying different carbapenemases is of concern considering the increased resistance, paucity of treatment options, and the increased risk of mortality (1, 14,19,20). Hence, it is important to continuously monitor the trends of carbapenem resistance, both at global and regional levels. Understanding the trends in carbapenem-resistance mechanisms is important for the implementation of optimal stewardship programs, development of novel antimicrobial agents, and the allocation of public health resources, all of which are essential in the continued struggle against antimicrobial-resistant infections.