We conducted a cross-sectional analysis of cisgender MSM who attended participating STD clinics in 12 eGISP and 8 SURRG jurisdictions in 2018 to 2019 (Supplemental Text, http://links.lww.com/OLQ/A748). We included MSM with a positive gonococcal culture obtained from any anatomic site (urogenital, rectal, pharyngeal) and antimicrobial susceptibility testing (AST) performed by agar dilution. We restricted isolate collection to consecutive MSM who had related clinical and epidemiologic data submitted to CDC and whose records contained a unique patient ID. For the purpose of this analysis, MSM were defined as a patient who self-identified as male gender and reported having any male gender sex partners within the past 3 months; consequently, patients with self-reported female gender, transgender persons, and patients who reported only female, nonbinary, or unknown gender sex partners were excluded.

Per the eGISP protocol, urethral gonococcal isolates were collected from male patients presenting with symptomatic urethritis, and pharyngeal and rectal isolates from patients who reported sexual exposure at oral and/or rectal anatomic sites with a N. gonorrhoeae nucleic acid amplification test performed.⁷ Per the SURRG protocol, urethral, pharyngeal, and rectal isolates were collected from patients who reported sexual exposure at urogenital, pharyngeal, and/or rectal anatomic sites with a N. gonorrhoeae nucleic acid amplification test performed. For eGISP, per participating clinic, collection was limited to the first 25 isolates from male patients with symptomatic urethritis and the first 25 pharyngeal and rectal isolates from patients who reported sexual exposure at oral or rectal anatomic sites. There was no maximum collection limit for clinics participating in SURRG. There were no minimum monthly isolate contributions required by participating clinics in either project.

Specimens collected for N. gonorrhoeae isolation were inoculated on selective media and incubated at 36°C ± 1°C in 5% CO₂ by local laboratory staff. Gonococcal isolates were subcultured on noninhibitory medium (e.g., chocolate agar) with 1% IsoVitalex to obtain a pure culture of the isolate and for subsequent species identification. For eGISP, local laboratories were only required to do presumptive identification; whereas, for SURRG, local laboratories also used matrix-assisted laser desorption-ionization time-of-flight mass spectrometry and API-NH assay for identification of N. gonorrhoeae.^32s,33s After 18 to 20 hours of incubation, growth from the pure culture was suspended in trypticase soy broth containing 20% (vol/vol) glycerol and frozen to −70°C, and shipped monthly to 1 of 4 participating Antibiotic Resistance Laboratory Network (ARLN) regional laboratories. Antibiotic Resistance Laboratory Network regional laboratories performed N. gonorrhoeae AST by agar dilution. Antimicrobials tested included azithromycin (0.008–16 μg/mL), cefixime (0.002–1 μg/mL), ceftriaxone (0.001–1 μg/mL), ciprofloxacin (0.001–32 μg/mL), penicillin (0.008–64 μg/mL), and tetracycline (0.06–64 μg/mL). Isolates with a cefixime MIC of ≥1 μg/mL, ceftriaxone MIC of ≥1 μg/mL, or azithromycin MIC of ≥16 μg/mL were shipped to the CDC and tested to the end point MIC at the CDC Division of STD Prevention laboratory. All isolates were tested for β-lactamase activity using the nitrocefin test by the regional ARLN laboratories. Isolates with azithromycin MICs of ≥2 μg/mL, cefixime MICs of ≥0.25 μg/mL, or ceftriaxone MICs of ≥0.125 μg/mL were shipped from regional laboratories to the CDC for confirmatory AST. To ensure the accuracy of agar dilution AST results, control N. gonorrhoeae strains (F18/ATCC 49226, WHO L and WHO U) with known MICs were included for each susceptibility test run at both the regional ARLN laboratories and at the CDC Division of STD Prevention laboratory.^34s,35s For additional quality assurance, ARLN regional laboratories participated in external quality assessment by testing a panel of 15 unidentified N. gonorrhoeae strains (provided by the CDC) twice yearly.

Antimicrobial susceptibility testing results were reported as MIC for antigonococcal antimicrobials. The results were interpreted according to the criteria recommended by the Clinical and Laboratory Standards Institute for penicillin resistance (MIC ≥2 μg/mL or β-lactamase positive), ciprofloxacin resistance (MIC ≥1.0 μg/mL), and tetracycline resistance (MIC ≥2.0 μg/mL).^34s The Clinical and Laboratory Standards Institute has not established the criteria for resistance to azithromycin, cefixime, and ceftriaxone. Therefore, the breakpoints used in these analyses to define elevated MIC for these antimicrobials include those previously described by the CDC GISP: azithromycin MIC ≥2.0 μg/mL, cefixime MIC ≥0.25 μg/mL, and ceftriaxone MIC ≥0.125 μg/mL.^36s

Clinical and demographic data documented at the time of visit were abstracted from medical records by local clinic staff. Patients and each of their specimens were assigned a unique ID such that multiple specimens contributed by a single patient could be identified for a single visit and/or tracked across multiple clinic visits. In addition to anatomic site of infection, the following additional variables were included in this analysis: gender of sex partners (MSM exclusively; men who have sex with men and women), age, race/Hispanic ethnicity, and HIV status (self-reported and/or per medical chart). Given possible geographic differences in antimicrobial susceptibility patterns, we additionally categorized each clinic site by US census region (Northeast, Midwest, South, West) for stratified analyses.

Demographic, clinical, and epidemiologic data were linked to AST results at the CDC.

We calculated frequencies of demographic and epidemiologic characteristics of cisgender MSM from whom N. gonorrhoeae isolates were collected. To describe antimicrobial susceptibility across anatomic sites, we calculated the MIC range, geometric mean MICs with 95% confidence intervals (CIs) for each antimicrobial by anatomic site, and the number and proportion of isolates with elevated or resistant MICs. We calculated the MIC distributions (percentage of isolates by MICs) by antimicrobial for pharyngeal, rectal, and urethral isolates. To account for correlated observations (i.e., multiple isolates provided by the same patients either at the same visit or across visits), we used linear mixed-effects models to calculate geometric mean MICs and mixed-effects logistic regression models to calculate the proportion of isolates with elevated or resistant MICs across anatomic sites. We used the likelihood ratio test to assess any MIC difference across anatomic sites (overall test); if the overall P value was less than 0.05, we performed post hoc pairwise comparisons among anatomic site pairs. We used the Holm-Bonferroni method (Holm’s sequential Bonferroni procedure) to control the family-wise error rate for the multiple comparisons. We also assessed geometric mean MICs and proportion with elevated or resistant MICs across anatomic sites of infection for each antimicrobial by geographic location using the statistical models described previously.

Analyses were conducted using SAS version 9.4 (SAS Institute, Cary, NC) and R version 3.6.3. software (R Core Team, Vienna, Austria). Data were collected as sentinel surveillance activities and were exempt from CDC institutional review board review.

From January 1, 2018, to December 31, 2019, 13,363 isolates were collected from cisgender men attending participating STD clinics. Of these, 12,585 (94%) isolates were collected from men with known gender of sex partners; 6576 (49%) isolates were collected from 5456 unique cisgender MSM and are included in the analysis. Most participants contributed one isolate only (n = 4570; 83.8%). Among all participants, 13.1% (n = 714) contributed 2 isolates, 2.4% (n = 129) contributed 3 isolates, and <1% (n = 43) contributed more than 3 isolates. Most participants had only 1 clinic visit (n = 4964; 91.0%); the maximum number of clinic visits by any one MSM was 5 visits over the 2 years (n = 1).

Of the 6576 isolates, 3974 were urethral, 1553 were rectal, and 1049 were pharyngeal isolates (Table 1). There was variation in the geographic distribution of isolates across anatomic sites. Most pharyngeal and rectal isolates (62.3% and 60.4%, respectively) were contributed by participants in the West, whereas most urethral isolates were collected in the Northeast (40.4%) and West (40.0%) regions. Across anatomic sites, more than 90% of isolates were collected from MSM who only identified recent male sex partners. There was some variation in the distribution of ages and race/Hispanic ethnicity of MSM from whom isolates were collected across anatomic sites. For example, 16.9% of urethral isolates were from MSM aged 20 to 24 years, whereas 22.7% of pharyngeal isolates were from MSM aged 20 to 24 years. Overall, 17.5% of isolates were collected from MSM who had a positive HIV status documented at the time of isolate collection; this varied according to anatomic site of infection (e.g., 19.4% of urogenital isolates vs. 11.1% of pharyngeal isolates).

The MIC distributions by antimicrobial for pharyngeal, rectal, and urethral isolates were unimodal across antimicrobials, with the exception of ciprofloxacin, which was bimodal (Fig. 1). There were no significant differences in the geometric mean MICs for azithromycin, ciprofloxacin, penicillin, and tetracycline across anatomic sites (Table 2). For cefixime, there were differences in the geometric mean MICs across anatomic sites (P < 0.05); the geometric mean MIC for pharyngeal isolates (geometric mean MIC, 0.0192 μg/mL) was higher compared with rectal isolates (geometric mean MIC, 0.0181 μg/mL; adjusted P < 0.05) and urethral isolates (geometric mean MIC, 0.0176 μg/mL; adjusted P < 0.05). The rectal geometric mean MIC for cefixime was not significantly different from the urethral geometric mean MIC. For ceftriaxone, there were also differences in the geometric mean MICs across anatomic sites (overall, P < 0.05); the geometric mean MIC for pharyngeal isolates (geometric mean MIC, 0.0108 μg/mL) was higher compared with rectal isolates (geometric mean MIC, 0.00987 μg/mL; adjusted P < 0.05) and urethral isolates (geometric mean MIC, 0.0098 μg/mL; adjusted P < 0.05). The rectal geometric mean MIC for ceftriaxone was not significantly different from the urethral geometric mean MIC.

There were no differences in the proportion of isolates with elevated or resistant MICs across anatomic sites for azithromycin, cefixime, ciprofloxacin, penicillin, and tetracycline; however, for ceftriaxone, the proportion of isolates with elevated MICs at the pharynx was higher than the proportion with elevated MICs at the rectum or urethra (P < 0.05). For ceftriaxone, 0.67% (7 of 1049; 95% CI, 0.32–1.4) of pharyngeal isolates had elevated MICs compared with 0.18% (7 of 3974; 95% CI, 0.08–0.37) of urethral isolates and 0.13% (2 of 1553; 95% CI, 0.03–0.51) of rectal isolates.

In some geographic regions, there were differences in geometric mean MICs for azithromycin, cefixime, and ceftriaxone across anatomic sites of infection (Table S1, http://links.lww.com/OLQ/A749). For azithromycin, the geometric mean MIC differed across anatomic sites in the Northeast with higher pharyngeal geometric mean MIC (P < 0.05) compared with rectal and urethral. For cefixime and ceftriaxone, the geometric mean MICs differed across anatomic sites in the Midwest with higher pharyngeal geometric mean MICs than rectal and urethral geometric mean MICs (P < 0.05). There were no significant differences in the geometric mean MICs for ciprofloxacin, penicillin, and tetracycline according to anatomic site by geographic region.

The proportion of pharyngeal isolates with elevated or resistant MICs for ceftriaxone was higher than the other anatomic sites in each geographic region; however, there were no statistically significant differences in the proportion with elevated or resistant MICs for azithromycin, cefixime, ceftriaxone, ciprofloxacin, penicillin, and tetracycline according to anatomic site of infection by geographic region (Table S2, http://links.lww.com/OLQ/A749).