Neisseria gonorrhoeae infection remains a major global public health concern because of increasing antimicrobial resistance (AMR), including reduced susceptibility to extended-spectrum cephalosporins (1,2). Extragenital infections, particularly in the pharynx and rectum, are often asymptomatic and can contribute to persistence and transmission of resistant organisms (3–5).

We analyzed samples from 189 N. gonorrhoeae–positive clinical encounters collected during 2022–2024 at Tripler Army Medical Center (TAMC; Honolulu, Hawaii, USA) and Madigan Army Medical Center (MAMC; Tacoma, Washington, USA). The human research protection offices from both institutions, under the umbrella of the Defense Health Agency, reviewed the study and determined it to be research not involving human subjects. Investigators adhered to US policies for the protection of human subjects as prescribed in 45 Code of Federal Regulation 46.

We analyzed data from specimens that included urine and swabs from oropharyngeal, rectal, and urogenital sites (Table). We deidentified data and analyzed it at the encounter level; we were unable to exclude repeat encounters from the same patient. We performed targeted next-generation sequencing using the Urinary Pathogen ID/AMR Panel (Illumina, https://www.illumina.com). We identified AMR markers using predefined sequence-matching rules specific to each gene or mutation (e.g., exact match for point mutations such as rpsJ V57M and curated allele grouping for penA variants). Because short-read data and panel design did not support reliable allele-level or phylogenetic resolution, we analyzed penA variants as a single category. This study did not include phylogenetic reconstruction, multilocus sequence typing, or transmission network analysis; therefore, our findings describe marker prevalence and co-occurrence patterns rather than clonal relationships.

Figure

Figure. Prevalence of AMR markers in study of AMR patterns in extragenital Neisseria gonorrhoeaeinfections, US military medical centers, 2022–2024. Heatmap shows percent of encounters in which each genotypic marker...

Across all encounters, resistance-associated markers frequently co-occurred. The most prevalent markers were macB (77.2%) and rpsJ V57M (75.7%), and we detected penA variants in 63.0% of isolates (Figure). We observed co-occurrence of these markers in 60.3% of encounters, which suggested common resistance-associated marker combinations within the cohort. Findings from TAMC demonstrated higher prevalence of rpsJ V57M (84.2% vs. 65.9%; p = 0.004) and penA variants (70.3% vs. 54.5%; p = 0.034) than for those from MAMC.

Markers associated with fluoroquinolone resistance were also present (gyrA S91F, 35.4%). We frequently detected structural efflux pump genes (macB, farA, mtrE); however, because those genes are part of the core genome, their presence alone does not predict phenotypic resistance. Regulatory mutations (e.g., mtrR variants) and combined genetic contexts are more directly associated with resistance expression.

We observed anatomic site–specific differences. The plasmid-mediated β-lactamase gene bla_TEM-1 was more prevalent in extragenital specimens; we detected bla_TEM-1 in 10/17 (58.8%) of rectal swab samples and 10/24 (41.7%) of oropharyngeal swab samples, compared with 22/125 (17.6%) of urine specimens (Figure). Compared with oropharyngeal swab samples, rectal specimens had increased odds of bla_TEM-1 detection (odds ratio [OR] 6.3; p = 0.004), whereas urine specimens had lower odds (OR 0.29; p = 0.012). In adjusted analyses controlling for age, sex, and institution, rectal specimens showed increased odds and wide 95% CIs (adjusted OR 2.57 [95% CI 0.94–7.00]).

Chromosomal resistance-associated markers such as rpsJ V57M were more common in urine specimens (80.0%) than in oropharyngeal specimens (50.0%; OR 3.90; p = 0.014). Conversely, we detected farA more frequently in urine (66.4%) than in extragenital (oropharyngeal and rectal) sites (<12.5%) (Figure). Those findings suggest that different anatomic niches may harbor distinct resistance-associated profiles.

Among 189 encounters, 24.9% occurred in persons with documented use of HIV pre-exposure prophylaxis. bla_TEM-1 was more common among users (36.2% vs. 21.8%; p = 0.056), whereas farA was less frequent (21.3% vs. 62.7%; OR 0.16, p<0.001). Associations were attenuated after adjustment.

Antimicrobial drug treatment varied across encounters. Doxycycline exposure was common; however, doxycycline is not recommended for treating uncomplicated gonorrhea and was likely prescribed for concurrent or presumptive chlamydial infection or syndromic management (6). A high (78.3%) percentage of isolates exposed to doxycycline contained rpsJ V57M, consistent with tetracycline-associated resistance markers (7).

Our study targeted sequencing and short-read data, which limited resolution of gene variants (e.g., bla_TEM subtypes) and precluded plasmid reconstruction or phylogenetic analyses. The absence of phenotypic susceptibility testing limited our ability to interpret clinical resistance. Retrospective data and incomplete sampling across anatomic sites might affect prevalence estimates. Finally, lack of temporal and patient-level linkage data limited inferences regarding transmission dynamics.

In summary, we identified frequent co-occurrence of resistance-associated markers in N. gonorrhoeae from US military medical centers and found enrichment of bla_TEM-1 in extragenital sites. Our findings support the value of anatomic site–specific surveillance, especially for extragenital infections in sexually transmitted infections, and molecular AMR marker detection (1,4,8,9).

Top

Suggested citation for this article: Dombach JL, MacArthur JL, Mekonnen C, La THA, Nxedhlana V, Norris MH, et al. Antimicrobial resistance in extragenital Neisseria gonorrhoeae infections, US military centers, 2022–2024. Emerg Infect Dis. 2026 Aug [date cited]. https://doi.org/10.3201/eid3208.260337