Prevalence of Cefixime-Resistant Neisseria gonorrhoeae in Melbourne, Australia, 2021–2022

Abstract

Sexually transmitted infections (STIs) have been rising worldwide over the last decade [1]. Gonorrhea is of particular concern because cases resistant to ceftriaxone, the last remaining fully effective antimicrobial against Neisseria gonorrhoeae, have occurred [2]. Dual antimicrobial therapy (ceftriaxone plus azithromycin) was introduced in the early 2010s for first-line treatment for gonorrhea in many countries. However, several countries, including the United States [3] and United Kingdom [4], changed to ceftriaxone monotherapy in the late 2010s due to observed rises in azithromycin resistance in N gonorrhoeae after the introduction of dual therapy and the likelihood of selecting for macrolide-resistant Mycoplasma genitalium [5]. Alternative regimens for gonorrhea treatment are urgently required to address the rise in antimicrobial resistance in N gonorrhoeae.

Cefixime is a third-generation oral cephalosporin antibiotic. In 2002, the US Centers for Disease Control and Prevention recommended a single oral dose of cefixime (400 mg) as one of the first-line treatments for gonorrhea [6]. In 2012, this recommendation ceased because the rising rates of cefixime resistance raised concerns about the increasing resistance in all cephalosporins. In 2021, the Centers for Disease Control and Prevention recommended a single dose of ceftriaxone (500 mg) as the only first-line treatment for uncomplicated gonorrhea; however, a single oral dose of cefixime (800 mg) can be considered an alternative treatment if ceftriaxone administration is not available or feasible [3]. Cefixime was not considered the first-line treatment because of decreased susceptibility or resistance in surveillance data and the concerns about its effectiveness for oropharyngeal gonorrhea treatment [7].

A global antimicrobial resistance surveillance study for N gonorrhoeae in 2017 to 2018 found that only 51 countries reported data on cefixime susceptibility, and there have been no data in Australia [8]. Thus, we conducted this study aimed to determine the prevalence of cefixime-resistant N gonorrhoeae among individuals attending a sexual health clinic in Melbourne, Australia.

METHODS

This cross-sectional study was conducted at the Melbourne Sexual Health Centre (MSHC) between 9 August 2021 and 18 July 2022. The MSHC is a major public sexual health service in Victoria, Australia. It provides about 50 000 clinical consultations each year. Individuals at higher risk for STIs and those with STI symptoms were preferentially triaged into the clinic. HIV and STI testing and treatment are free of charge.

Individuals were screened for N gonorrhoeae by nucleic acid amplification test with the Aptima Combo 2 Assay (Hologic Panther system; Hologic). Untreated individuals with a positive test result were recalled to the MSHC for treatment, and a swab was taken for culture. Swabs were inoculated onto the Modified Thayer Martin Agar with a selective medium that contains a GC agar base (Difco; supplied by the Media Preparation Unit, The University of Melbourne) [9] and incubated at 35 to 37 °C in 5% CO₂ for 48 hours. Oxidase-positive gram-negative diplococci colonies resembling N gonorrhoeae underwent confirmatory testing by the Vitek-MS MALDI TOF system (BioMérieux). Ceftriaxone, cefixime, and azithromycin minimum inhibitory concentrations (MICs) were determined by agar dilution at the Microbiological Diagnostic Unit Public Health Laboratory in Victoria, Australia. MIC testing was performed from a 20- to 24-hour culture on chocolate agar incubated at 35 ± 1 °C in 5% CO₂. In some cases, isolates were stored in glycerol storage broth at −80 °C prior to subculture. MIC plates were prepared per standard M07 of the Clinical and Laboratory Standards Institute via antimicrobial powders obtained from Sigma Aldrich (ceftriaxone, ciprofloxacin, penicillin, spectinomycin, and tetracycline) or Pfizer (azithromycin) [10]. The European Committee on Antimicrobial Susceptibility Testing (EUCAST) breakpoint tables and interpretive criteria were used to interpret the MICs [11]. Isolates with ceftriaxone or cefixime MICs >0.125 mg/L were considered resistant and azithromycin MICs >1 mg/L as likely resistant (EUCAST epidemiologic cutoff value).

For individuals with isolates from multisite infections collected on the same day (eg, genital, oropharyngeal, and rectal), we randomly selected 1 isolate per individual and excluded the remaining isolates. For those with samples collected on different days during the study period, the isolates from different episodes (1 per episode) were included in the final analysis. In other words, only 1 isolate per patient episode was included in the final analysis.

We calculated the proportion of isolates with resistance to cefixime, azithromycin, and ceftriaxone and the corresponding 95% CIs with the exact binomial method. Univariable logistic regression with a generalized estimating equation was performed to identify factors that were associated with cefixime-resistant N gonorrhoeae. Variables with a P value <.20 in the univariable analyses were included in the multivariable analyses. The exchangeable correlation matrix was used. Crude and adjusted odds ratios and the corresponding 95% CIs were reported. All statistical analyses were performed in Stata (version 17; StataCorp). This study was approved by the Alfred Hospital Ethics Committee, Melbourne, Australia (685/22).

RESULTS

Between August 2021 and July 2022, 1297 N gonorrhoeae isolates from 1049 individuals underwent antimicrobial susceptibility testing for cefixime, azithromycin, and ceftriaxone​. Of the 114 patients who had more than 1 isolate, we randomly selected 1 isolate per person per episode, and the remaining 121 isolates were excluded. The remaining 1176 isolates from 1049 clients were included in the final analysis. Of the 1176 isolates, 104 people had isolates from multiple episodes (126 isolates) and the median time between episodes per person was 95 days (IQR, 59–139).

The median age of the individuals was 31 years (IQR, 27–37), and most were gay, bisexual, and other men who have sex with men (MSM; 83.5%, 982/1176). About 71.5% (841/1176) were HIV negative and not taking HIV preexposure prophylaxis (PrEP); 17.0% (200/1176) were taking HIV PrEP; and 11.5% (135/1176) were people with HIV. Most isolates were collected from urogenital sites (including urethral, vaginal, and cervical specimens; 40.2%, 473/1176), followed by the rectum (35.8%, 421/1176) and oropharynx (24.0%, 282/1176).

The proportion of isolates with resistance to cefixime was 6.3% (95% CI, 5.0%–7.8%; 74/1176), and azithromycin was 4.9% (95% CI, 3.8%–6.3%; 58/1176; Figure 1). There were no isolates with resistance to cefixime and azithromycin. No isolates demonstrated resistance to ceftriaxone (0%; 95% CI, 0%–.3%; 0/1176). Cefixime resistance rates varied by anatomic site, with 7.4% (21/282) resistance at the oropharynx, followed by 6.3% (30/473) from urogenital sites and 5.5% (23/421) from the rectum. Women had the highest level of cefixime resistance (16.4%, 10/61), followed by men who have sex with women (6.4%, 7/109) and MSM (5.8%, 57/982).

Figure 1.

The distribution of minimum inhibitory concentrations of cefixime and azithromycin for Neisseria gonorrhoeae isolates. The vertical line indicates the resistance breakpoint.

Open in new tabDownload slide

Multivariable logistic regression showed that women had higher odds (adjusted odds ratio, 2.72; 95% CI, 1.22–6.06) of being infected with N gonorrhoeae resistant to cefixime as compared with MSM after adjusting for continent of birth and sex work status (Table 1). Cefixime resistance was not associated with age, anatomic site, HIV status, PrEP use, and sex overseas in the last 12 months. Of the 10 women carrying cefixime-resistant isolates of N gonorrhoeae, most of them were born in Asia (n = 6); half of them had worked as sex workers (n = 5); and the isolates were collected from the vagina (n = 5), followed by the cervix (n = 2), oropharynx (n = 2), and urine (n = 1).

DISCUSSION

The proportion of isolates with resistance to cefixime was 6.3% among sexually active individuals attending an Australian sexual health clinic. This is above the maximum 5% antimicrobial resistance rate recommended by the World Health Organization (WHO) for an antimicrobial to be suitable as a standard treatment regime. Our previous study tested 7588 N gonorrhoeae isolates in Victoria and found that the prevalence of resistance to ceftriaxone was 2.9% and that to azithromycin was 8.2% in 2018 [12]. This level of azithromycin resistance (>5% per the WHO-recommended antimicrobial resistance threshold) among isolates of N gonorrhoeae from Victoria is of concern as it threatens the efficacy of dual antimicrobial therapy (ceftriaxone plus azithromycin), which is currently recommended as first-line treatment for gonorrhea in Australia's guidelines [13]. With the concern of antimicrobial resistance, the MSHC dropped azithromycin for gonorrhea treatment in late 2021. Furthermore, ceftriaxone plus doxycycline instead of azithromycin is recommended for rectal gonorrhea and chlamydia coinfections per the Australian STI guidelines [14]. Other strategies will be required to counteract the rise in antimicrobial resistance in N gonorrhoeae.

Our data showed that 6.3% of N gonorrhoeae isolates among clinic attendees were resistant to cefixime. This estimate is similar to estimates from some European countries, such as Italy (5.0%, 5/100), Switzerland (5.6%, 3/54), and Greece (6.0%, 5/83), but substantially higher than that in the United States (0.3%, 15/5160) [8]. However, the United States uses the Clinical and Laboratory Standards Institute susceptibility breakpoint for cefixime (MIC ≤0.25 mg/L) [15], which is higher than the EUCAST's breakpoint (MIC <0.125 mg/L) [11]. We also found that women (n = 61) have a higher level of cefixime resistance (16%) when compared with men (6%). We found that most women with isolates resistant to cefixime were born in Asia. It is possible that these women may have been exposed to cefixime in the past for other infections, including urinary tract infections, or they might have had sex with individuals from Asia or within a sexual network where cefixime-resistant N gonorrhoeae might have been imported from Asia. There have been limited data on the prevalence of resistance to cefixime in Asia [8]. In 2022, a Chinese study tested 2804 N gonorrhoeae isolates and reported a high prevalence of resistance to cefixime (16.0%) with a MIC >0.125 mg/L as the breakpoint [16]. Further genomic work may be required to identify the N gonorrhoeae strain with resistance to cefixime to better understand why a high level of resistance is concentrated in certain populations.

The major strengths of this study are its large sample size (N = 1176) and that it represents the first data on cefixime susceptibility from Australia. The 2017–2018 WHO global antimicrobial resistance surveillance data showed that of the 51 countries that reported data on cefixime susceptibility, the Philippines was the only country in the Western Pacific region to do so (4.2%, 4/95), and just 3 countries (ie, United States, Canada, and Chile) tested >1000 isolates [8]. Our study provides additional estimates not only in Australia but also in the Western Pacific region. However, our study has several limitations. First, it was conducted at a single urban sexual health service, and sexual health clinic attendees may have had higher levels of antibiotic exposure or more diverse sexual networks as compared with the general population. Additionally, about half the attendees at our clinics are travelers where importation of resistant N gonorrhoeae may have occurred, and this may not fully represent the prevalence of resistant N gonorrhoeae in the broader population [17, 18]. Most data in other countries, though, come from STI centers and, like our study, may have overestimated the prevalence of resistance in the entire community. Second, this study was conducted in 2021 to 2022 with major lockdowns during the COVID-19 pandemic. Clinic attendees during this period might have different characteristics due to the changes in sexual practices and STI screening patterns during COVID-19 [19]. The lockdowns also meant that there was likely to be less international travel; as such, our data are more likely than usual to represent locally endemic strains rather than recently imported ones.

Cefixime is currently not available in Australia. As the current prevalence of cefixime-resistant N gonorrhoeae among sexually active individuals in Melbourne exceeds the WHO's recommended antimicrobial resistance threshold of 5%, these data suggest that cefixime would have limited benefits for gonorrhea treatment in Australia. Additionally, while a meta-analysis concluded that cefixime (400 and 800 mg) for gonorrhea treatment is highly effective in treating urogenital infections (97% and 98% cure rates, respectively), it is less effective for eradicating oropharyngeal infections (91% and 82% cure rates) [7]. There are limited options available for first-line treatment of gonorrhea; thus, further development and investigation of alternative therapies for gonorrhea treatment are required.

Notes

Acknowledgment. We thank Rohit Sasidharan at the Melbourne Sexual Health Centre for his help in extracting electronic data for the study and the technical staff at the Microbiological Diagnostic Unit Public Health Laboratory for performing antimicrobial susceptibility testing.

Author contributions. E. P. F. C., L. A. B., D. A. W., and C. K. F. conceived the study. E. P. F. C. performed data analyses, interpreted data, and wrote the manuscript. K. S., V. D. P., D. A. W., and B. P. H. supervised the laboratory testing. D. A. W. and B. P. H. provided laboratory support on antimicrobial susceptibility testing. K. S., V. D. P., N. L. S., K. L. S., D. A. W., and B. P. H. helped with data interpretation on antimicrobial susceptibility testing. M. Y. C., C. S. B., L. A. B., L. A. V., I. A., K. M., and C. K. F. helped with data interpretation on clinical perspective and antimicrobial resistance. All authors revised the manuscript for intellectual content and approved the final version of the manuscript. The corresponding author had full access to all the data in the study and had final responsibility for the decision to submit for publication.

Data availability. All data generated or analyzed during this study are included in this published article.

Financial support. This work was supported by an National Health and Medical Research Council Emerging Leadership Investigator Grant (GNT1172873 to E. P. F. C. and GNT1174555 to D. A. W.); and an National Health and Medical Research Council Leadership Investigator Grant (GNT1172900 to C. K. F., GNT1173361 to C. S. B., GNT2017383 to K. L. S., and GNT1196103 to B. P. H.).

References

Author notes