Phenotypic and genotypic properties of Neisseria gonorrhoeae isolates in Norway in 2009: antimicrobial resistance warrants an immediate change in national management guidelines
- First Online:
- Cite this article as:
- Hjelmevoll, S.O., Golparian, D., Dedi, L. et al. Eur J Clin Microbiol Infect Dis (2012) 31: 1181. doi:10.1007/s10096-011-1426-4
- 148 Views
Despite rapidly diminishing treatment options for Neisseria gonorrhoeae and high levels of ciprofloxacin resistance worldwide, Norwegian guidelines still recommend ciprofloxacin as empirical treatment for gonorrhea. The present study aimed to characterize phenotypical and genotypical properties of N. gonorrhoeae isolates in Norway in 2009. All viable N. gonorrhoeae isolates (n = 114) from six university hospitals in Norway (2009) were collected, representing 42% of all notified gonorrhea cases. Epidemiological data were collected from the Norwegian Surveillance System for Communicable Diseases and linked to phenotypical and genotypical characteristics for each N. gonorrhoeae isolate. Resistance levels to the antimicrobials examined were: ciprofloxacin 78%, azithromycin 11%, cefixime 3.5%, ceftriaxone 1.8%, and spectinomycin 0%. The minimum inhibitory concentrations of gentamicin varied from 1.5 to 8 mg/L. Forty-one (36%) of the isolates were β-lactamase-producing, 17 displayed penA mosaic alleles, and 72 different N. gonorrhoeae multiantigen sequence types (ST; 37 novel) were identified. The most common ST was ST1407 (n = 11), containing penA mosaic allele. Four of these isolates displayed intermediate susceptibility/resistance to cefixime. The N. gonorrhoeae strains circulating in Norway were highly diverse. The level of ciprofloxacin resistance was high and the Norwegian management guidelines should promptly exclude ciprofloxacin as an empirical treatment option for gonorrhea.
Treatment of gonorrhea is a major global concern because Neisseria gonorrhoeae has developed resistance to all previously recommended first-line antimicrobials, such as penicillins, tetracyclines, and ciprofloxacin [1, 2]. Extended-spectrum cephalosporins (ESCs), ceftriaxone (injectable) and cefixime (oral), are currently the only effective antimicrobials for empirical first-line treatment in many countries. However, verified clinical failures with cefixime have been described in Japan , and most recently also in Norway  and the UK . Recently, the first N. gonorrhoeae strain with high-level clinical resistance to ceftriaxone was identified in Japan and subsequently characterized in detail . Accordingly, the antimicrobial options for the treatment of gonorrhea have rapidly diminished, and monitoring of antimicrobial susceptibility, contact tracing, test-of-cure, and molecular typing are more important than ever.
Gonococcal resistance to ciprofloxacin has been reported since the early 1990s [1, 2, 7], and most countries in Asia [1, 2, 7], Europe , and the USA  have abandoned ciprofloxacin as an empirical treatment of gonorrhea. Surveillance of fluoroquinolone resistance started in 1995 in Norway and by 2007, 39% of the examined N. gonorrhoeae displayed intermediate susceptibility or resistance to ciprofloxacin . Despite this information and a recent study showing 47% resistance to ciprofloxacin among Norwegian N. gonorrhoeae isolates , the Norwegian management guidelines still recommend ciprofloxacin as empirical treatment . This is especially worrisome as general practitioners, who likely follow these guidelines in detail, treat approximately half of the patients diagnosed with N. gonorrhoeae in Norway . There is also a trend toward a higher percentage (from 20 to 30% in the early 1990s to approximately 40% in the last 8 years) of diagnosed gonorrhea being contracted abroad (Norwegian Surveillance System for Communicable Diseases (MSIS); www.msis.no), predominantly in Asia and other European countries. Furthermore, there is an increase in the proportion of men who have sex with men (MSM) diagnosed with gonorrhea, i.e., from 8% in 1993 to more than 50% of all diagnosed gonorrhea cases in 2010 (www.msis.no).
In the present study, we investigated the genotypical and phenotypical properties (including antimicrobial resistance determination), of all viable N. gonorrhoeae isolates diagnosed in 2009 at six university hospitals in Norway. Laboratory data were linked to epidemiological information from the Norwegian Surveillance System for Communicable Diseases (NSSCD).
Materials and methods
A total of 126 clinical N. gonorrhoeae isolates, cultured in 2009, were collected from six university hospitals in Norway. Of these, 114 isolates from unique patients were retrieved, and these isolates were initially cultured at the University Hospital of North Norway, Tromsø (n = 3); St. Olavs Hospital, Trondheim University Hospital, Trondheim (n = 6); Stavanger University Hospital, Stavanger (n = 16); Haukeland University Hospital, Bergen (n = 25); Oslo University Hospital, Ullevål Hospital, Oslo (n = 60); and Akershus University Hospital, Oslo (n = 4). This study population represented 42% of all notified gonorrhea cases in Norway in 2009.
For each patient, NSSCD provided information regarding gender, age, symptoms, sexual orientation, and exposure to infection, i.e., where and by whom (commercial sex worker, regular partner or casual contact).
Statistics and ethics
Pearson Chi-squared tests were performed using SPSS 18 for Windows (SPSS, Chicago, IL, USA) to show any statistical difference in ciprofloxacin resistance between MSM and heterosexual men, and between ciprofloxacin resistance in foreign and domestically contracted N. gonorrhoeae.
The study was approved by the Regional Committee for Medical and Health Research Ethics (Rek-Nord), Tromsø, Norway.
Antibiotic susceptibility of Neisseria gonorrhoeae isolates (n = 114) cultured in 2009 in Norway
In vitro resistance (%)
Ciprofloxacin (S ≤ 0.03/R > 0.06)a
Ceftriaxone (S ≤ 0.12/R > 0.12)a
Cefixime (S ≤ 0.12/R > 0.12)a
Azithromycin (S ≤ 0.25/R > 0.5)a
Spectinomycin (S ≤ 64/R > 64)a
MIC range: 1.5 to 8 mg/L
Genomic DNA was isolated using the BUGS`n BEADS STI-fast kit (NorDiag ASA, Oslo, Norway) on the NorDiag Bullet, according to the manufacturer’s instructions. A conventional PCR was used to screen for penA mosaic alleles, as previously described . The more variable internal segments of porB (490 bp) and tbpB gene (390 bp), examined in N. gonorrhoeae multiantigen sequence typing (NG-MAST) , as well as the full-length porB gene were sequenced as previously described . Different allele numbers and sequence types (STs) were assigned on the NG-MAST web site (www.ng-mast.net).
Study population and demographics
Key epidemiological data for Neisseria gonorrhoeae cases (N = 114) in 2009 in Norway
Same sex transmission
Reason for testing
At own request
Comparison of epidemiological data between the Norwegian gonorrhea cases and the study population in 2009
Norway in totala (N = 269)
This study (N = 114)
Same sex transmission
Infected in Norway
Infected in Thailand
Antimicrobial susceptibility testing and genotyping
The results of the antimicrobial susceptibility testing of all N. gonorrhoeae isolates (n = 114) are summarized in Table 1. In total, 78% (n = 89) of the isolates were resistant, and one isolate (0.9%) had intermediate susceptibility to ciprofloxacin. Four isolates (3.5%) showed intermediate susceptibility/resistance to cefixime (0.19–0.38 mg/L), two isolates (1.8%) displayed intermediate susceptibility/resistance to ceftriaxone (0.19 mg/L), and 12 isolates (11%) were resistant to azithromycin. No intermediate susceptibility or resistance to spectinomycin was found, and the gentamicin MICs were low (1.5–8 mg/L). β-lactamase production, causing high-level resistance to penicillins, was found in 41 isolates (36%). A penA mosaic allele was identified in 17 isolates (15%), including all isolates with intermediate susceptibility/resistance to cefixime (n = 4).
Among the 31 MSM in the present study, 24 (77%) had strains resistant to ciprofloxacin while 64 of the 81 heterosexuals (79%) had ciprofloxacin-resistant strains. No significant difference was found (p = 0.854). All four isolates displaying intermediate susceptibility/resistance to cefixime were cultured from males (2 MSM) in the age group 39–46 years infected in Oslo (n = 3) and France (n = 1). These four (all NG-MAST ST1407) isolates all contained a penA mosaic allele, which is a predictive indicator of decreased susceptibility to ESCs, especially cefixime . A further 13 isolates contained a penA mosaic allele, and excluding the four isolates mentioned above, these had the highest MIC of cefixime (0.094–0.125 mg/L). Of the 17 isolates with the penA mosaic allele, 11 were determined as ST1407 (10 of these were isolated from MSM). All but two penA mosaic allele-positive isolates were from gonorrhea cases contracted in Norway, 14 of them in Oslo. The two foreign infections were one MSM and one heterosexual, both infected in France (ST1407 and ST3168 respectively).
Correlations between laboratory data and epidemiological information
In total, 38 of the men (39%) were infected abroad and Thailand was the most frequent place for contracting N. gonorrhoeae (50% among these men; Table 2). For women, 29% of the patients were infected abroad. Symptoms were the reason for seeking medical attention for 92 of the men (95%) and 9 of the women (53%; Table 2).
Of the domestically contracted isolates and the isolates contracted abroad, 55 out of 68 (81%) and 34 out of 45 (76%) respectively were ciprofloxacin-resistant. Seventy-one percent (5 out of 7) of the women infected abroad had ciprofloxacin-resistant isolates, and 9 of the women infected domestically (90%) had ciprofloxacin-resistant isolates. For men, 29 (27 heterosexual) out of 38 of those infected abroad (76%) and 46 (23 heterosexual) out of 58 of those infected domestically (79%) had ciprofloxacin-resistant isolates. No significant difference was found (p = 0.729).
The two isolates displaying intermediate susceptibility/resistance to ceftriaxone were both from MSM infected domestically. Two of the four isolates showing intermediate susceptibility/resistance to cefixime were from domestically infected MSM, one from a domestically infected heterosexual man, and one heterosexual man infected in another European country. Among the 12 azithromycin-resistant strains 2 were from women (1 domestically infected), 1 was from a heterosexual male infected abroad and 9 from men (5 of these were MSM) infected domestically.
This is the first study investigating phenotypical and genotypical characteristics of gonococcal infections diagnosed in Norway in relation to epidemiological data. The study population was representative of all notified gonococcal cases in Norway in 2009 with regard to gender, sexual orientation, and place of contraction of infection.
The treatment guidelines in Norway [10, 19] still recommend ciprofloxacin as empirical treatment for gonococcal infections. Treatment is only to be deviated when there is a verified treatment failure or when antimicrobial susceptibility testing indicates resistance to ciprofloxacin. The present study showed that the level of ciprofloxacin resistance was higher among isolates from individuals infected domestically (82%) compared with abroad (72%). According to the World Health Organization (WHO), the empirical first-line treatment should be changed when the resistance exceeds 5% in the general population (and mainly any resistance in a high-risk group of frequent transmitters) . Accordingly, in Norway ciprofloxacin (78% resistance) and azithromycin (11% resistance) should not be used for empirical treatment of gonorrhea, i.e. if no antimicrobial resistance testing has been performed to show that the gonococcal strains are susceptible. Worryingly, the level of intermediate susceptibility/resistance to cefixime and ceftriaxone was 3.5% and 1.8% respectively. The susceptibility to these ESCs, the last remaining treatment options in many countries, is also decreasing worldwide [1, 2, 6, 20]. Furthermore, cefixime treatment failures of urogenital gonorrhea have also been verified in Japan  and recently in Norway  and England . So far, clinical failures using ceftriaxone for treatment of urogenital gonorrhea have not been reported. However, recently, one verified treatment failure of pharyngeal gonorrhea was reported in Sweden , and the first N. gonorrhoeae strain with high-level resistance to ceftriaxone, most likely related to a treatment failure, has also been found in Japan and characterized in detail . It is crucial that the susceptibility of cefixime and ceftriaxone is monitored carefully (locally, nationally, and internationally), and this should also include surveillance and appropriate verification of suspected treatment failures. By enhanced use of antimicrobial susceptibility testing, improved antimicrobial resistance testing and continued use of test-of-cure for all positive patient samples, ESCs can still be used for empirical treatment. The use of test-of-cure may reveal treatment success, despite laboratory resistance for several antimicrobials. However, it is likely only a matter of time before ESC-resistant gonococcal strains spread internationally. Accordingly, it is crucial to focus research on developing new antimicrobials and/or strategies for the treatment of gonorrhea.
In the present study, the proportion of isolates containing a penA mosaic allele was higher among MSM (39%) than in heterosexuals (6.2%), which also caused the higher level of isolates with intermediate susceptibility/resistance to cefixime found among MSM, partly explained by the circulating ST1407.
The epidemiological data showed that 39% of all infections were imported. Several of the predominant STs identified in Norway have been described circulating in many other countries, which strengthens the need to follow international treatment guidelines . For example, ST1407, which contains a penA mosaic allele, has been found to cause decreased susceptibility to ESCs in many countries worldwide . Moreover, ST292 has been previously found in other European countries such as Sweden , France , and Scotland . In general, the NG-MAST results also showed that the strains circulating in Norway were heterogeneous; 72 NG-MAST STs, 37 not previously described, were found among 114 isolates. This high number of new STs is probably partially explicable by the fact that this is the first N. gonorrhoeae molecular surveillance report from Norway and the widespread import of N. gonorrhoeae strains into Norway. The low number of gonorrhea cases and the lack of large sexual networks may also suggest effective contact tracing, which is a main focus in Norway. Nevertheless, a previous study  and recent surveillance reports (unpublished data) have suggested that the culture-based diagnostics results in false-negative tests in Norway. This is especially true for rectal samples in MSM who practice exclusively receptive anal sex. The Norwegian geography with long distances from physicians who sample the patients to the diagnostic laboratory increases the risk of false-negative culture diagnostics. We encourage the use of transport media suitable for both molecular tests and culture, and ideally to test all samples using both tests.
In conclusion, the N. gonorrhoeae strains circulating in Norway in 2009 were highly heterogeneous. The level of resistance to first-line treatment ciprofloxacin (as well as to azithromycin) was high and the Norwegian management guidelines need to be revised promptly. The susceptibility of the ESCs, cefixime and ceftriaxone, is decreasing globally and should be monitored carefully (locally, nationally, and internationally), including surveillance and appropriate verification of suspected treatment failures.
This work was financially supported by the Örebro County Council Research Committee, the Foundation for Medical Research at Örebro University Hospital, Sweden, and the University of Tromsø, Tromsø, Norway. The collection of isolates and initial diagnosis was performed and funded by the participating University hospitals as part of their routine diagnosis.