Carbapenem-resistant Enterobacterales (CRE) are an urgent public health threat. In 2022, an estimated 13,387 CRE infections occurred among hospitalized patients in the United States (1). Different mechanisms may contribute to carbapenem resistance, but carbapenemase-producing CRE (CP-CRE) are of particular concern. Carbapenemase genes confer broad resistance to β-lactam antimicrobial drugs and are often located on mobile genetic elements, enabling them to spread between bacterial species. Metallo-β-lactamases (MBLs), including New Delhi metallo-β-lactamase (NDM)–producing CRE, are especially problematic because few antimicrobial drugs have activity against them (2). We report an increase in bla_NDM among CP-CRE collected through the Centers for Disease Control and Prevention’s Emerging Infections Program (EIP) Multi-site Gram-negative Surveillance Initiative.

The EIP Multi-site Gram-negative Surveillance Initiative conducts active, population- and laboratory-based surveillance of CRE (3–5). During 2016–2023, a total of 10 EIP sites participated in surveillance for carbapenem-resistant Enterobacter cloacae complex, Escherichia coli, and Klebsiella species (K. pneumoniae, K. oxytoca, and K. aerogenes) in isolates from a usually sterile site or from urine. Each site submitted a convenience sample of isolates to the Centers for Disease Control and Prevention for testing, including in-house reference broth microdilution incorporating an MBL screen, species identification, and real-time PCR for carbapenemase genes (3). We tested all isolates for bla_KPC, bla_NDM, and bla_OXA-48-like genes; we also tested isolates with a positive MBL screen for bla_VIM and bla_IMP (4). We defined CP-CRE as isolates positive for a carbapenemase gene by real-time PCR and resistant to ertapenem, imipenem, or meropenem as determined by broth microdilution and Clinical and Laboratory Standards Institute breakpoints (Appendix Table) (6). Using the Pearson χ² test, we compared the percentages of bla_NDM among CP-CRE collected in 2016 and 2023. Annual reports with epidemiologic, clinical, and laboratory data are available online (5).

Figure

Figure. Percentage of carbapenemase genes detected by real-time PCR among 1,288 CP-CRE chosen for a study analyzing an increase in bla_NDMamong carbapenemase-producing, carbapenem-resistant Enterobacterales, United States, 2016–2023. IMP,...

Among 1,288 CP-CRE identified, bla_KPC was the most common carbapenemase gene detected among all confirmed CP-CRE during 2016–2023 (72.7%, 937 isolates), followed by bla_NDM (20.6%, 265 isolates), bla_OXA-48-like (7.5%, 96 isolates), bla_IMP (0.6%, 8 isolates), and bla_VIM (0.5%, 7 isolates) (Figure); 25 (1.9%) isolates harbored >1 carbapenemase gene (Appendix Table). The percentage of bla_NDM isolates increased from 5.4% in 2016 (n = 6) to 39.8% in 2023 (n = 99) (p<0.00001). Conversely, bla_KPC decreased from 92.8% in 2016 (n = 103) to 53.0% in 2023 (n = 132) (p<0.00001) (Figure). Again comparing 2016 and 2023, we observed an increase in bla_NDM among E. coli (1.8% to 22.8%), Klebsiella spp. (3.6% to 11.6%), and E. cloacae complex (0.6% to 5.2%) (Table). NDM-producing Enterobacterales were more resistant to β-lactam combination agents than were K. pneumoniae carbapenemase–producing Enterobacterales (Appendix Table).

We report a notable shift in the type of carbapenemase genes among a convenience sample of 1,288 CP-CRE collected in the United States during 2016–2023. Although bla_KPC remained the most common carbapenemase gene, we observed a decrease in the proportion of bla_KPC coupled with an increase in bla_NDM. This shift was most striking among E. coli, with bla_NDM representing 73% of all carbapenemase-producing E. coli in 2023; in contrast, we observed bla_NDM among only 14.3% of carbapenemase-producing E. coli in 2016.

The increase of bla_NDM is alarming given that NDM-producing CRE are more resistant than other CRE isolates (7). Furthermore, newer β-lactam combination agents, including ceftazidime-avibactam, meropenem-vaborbactam, and imipenem-relebactam, are ineffective against MBL enzymes, and treatment options are limited (2,8). Aztreonam/avibactam and cefiderocol have demonstrated in vitro efficacy, but resistance has also been reported (7).

Our study is limited because it is based on a convenience sample of isolates from population-based surveillance and might be affected by sampling bias. We collected isolates from 10 EIP sites and national trends may not be extrapolated based on these data. This study is not able to determine whether there are changes in the incidence of NDM-producing CRE, however, our findings align with recent reports of rising NDM-positive CRE in New York City and among the Antimicrobial Resistance Laboratory Network (9,10). Whole-genome sequencing analysis is needed to determine gene variants and whether increases are driven by specific sequence types.

In conclusion, we report a concerning increase in bla_NDM among a convenience sample of CP-CRE collected across 10 EIP sites in the United States. Further investigation is needed to assess if this is a nationwide trend, to analyze epidemiologic data comparing characteristics of patients infected with bla_NDM-CRE and those with bla_KPC, and to examine whole-genome sequencing data to determine if the observed increase is related to clonal expansion. Our findings should alert clinicians to the increase in bla_NDM and encourage mechanism testing in clinical laboratories.

Top

Suggested citation for this article: Ansari UA, Campbell D, Brandenburg JM, Duffy N, Grass JE, Guh AY, et al. Increase in bla_NDM among carbapenemase-producing, carbapenem-resistant Enterobacterales, United States, 2016–2023. Emerg Infect Dis. 2026 Jun [date cited]. https://doi.org/10.3201/eid3206.251404