A new sexually transmitted infection ( STI ) in Geneva ? Ciprofloxacin-resistant Neisseria gonorrhoeae , 2002 – 2005

Principles: Neisseria gonorrhoeae (NG) resistant to ciprofloxacin (CR) was documented for the first time in Geneva in 2002 and increased from 7% that year to 47% in 2005. We describe NG cases during this period and compare characteristics of CR and ciprofloxacin susceptible (CS) cases. Method: Geneva microbiological laboratories identified NG cases. Antimicrobial susceptibility testing (AST) was performed on a sample of reported cases. The attending physicians completed questionnaires on demographic and epidemiological characteristics. Risk exposures were assessed by comparing CR and CS cases using logistic regression. Results: 238 NG cases were reported. Of 91 on which AST was performed, 23 (25%) were CR; 91% of these were male vs 87% of CS patients. Men having sex with men (MSM) represented 38% of CR cases compared with 31% of CS cases (p >0.05). Among CR cases 65% were Swiss compared with 56% of CS cases. Median age was 35 years for both. Casual sexual contacts were reported for 88% of CR cases and 72% of CS cases (p >0.05). The difference between CR and CS cases in terms of sexual activity outside Switzerland (50% and 19% respectively) remained significant after adjusting for sexual preference and nationality (OR: 7.0, CI 95: 1.99–24.6). Conclusion: Although CR infection was more common among Swiss MSM, only sexual activity outside Switzerland was independently associated with CR. Physicians should request AST before treatment and reconsider first-line use of ciprofloxacin. Surveillance of gonococcal antimicrobial resistance is essential in monitoring epidemiologic trends and updating recommendations on first-line treatment.

Principles: Neisseria gonorrhoeae (NG) resistant to ciprofloxacin (CR) was documented for the first time in Geneva in 2002 and increased from 7% that year to 47% in 2005.We describe NG cases during this period and compare characteristics of CR and ciprofloxacin susceptible (CS) cases.
Method: Geneva microbiological laboratories identified NG cases.Antimicrobial susceptibility testing (AST) was performed on a sample of reported cases.The attending physicians completed questionnaires on demographic and epidemiological characteristics.Risk exposures were assessed by comparing CR and CS cases using logistic regression.
Results: 238 NG cases were reported.Of 91 on which AST was performed, 23 (25%) were CR; 91% of these were male vs 87% of CS patients.Men having sex with men (MSM) represented 38% of CR cases compared with 31% of CS cases (p >0.05).Among CR cases 65% were Swiss com-pared with 56% of CS cases.Median age was 35 years for both.Casual sexual contacts were reported for 88% of CR cases and 72% of CS cases (p >0.05).The difference between CR and CS cases in terms of sexual activity outside Switzerland (50% and 19% respectively) remained significant after adjusting for sexual preference and nationality (OR: 7.0, CI 95: 1.99-24.6).

Conclusion:
Although CR infection was more common among Swiss MSM, only sexual activity outside Switzerland was independently associated with CR.Physicians should request AST before treatment and reconsider first-line use of ciprofloxacin.Surveillance of gonococcal antimicrobial resistance is essential in monitoring epidemiologic trends and updating recommendations on first-line treatment.
Reporting Neisseria gonorrhoeae to public health authorities has been mandatory for laboratories in Switzerland since 1987.Notifications of gonococcal infections increased by 97% between 1997 (269) and 2003 (531), growing by 119% in males and 48% in females [15].
In Geneva, notifications of gonococcal infections increased nearly twofold between 2001 and 2002 [16,17] and rose again slightly in 2004.
Neisseria gonorrhoeae resistant to ciprofloxacin (CR) was documented for the first time in Geneva  in 2002 by Unilabs Genève [18].Quinolone resistance in Geneva increased from 7% in 2002 to 47% in 2005 (Figure 1).
A Neisseria gonorrhoeae case is diagnosed by phenotypic identification and/or by molecular amplification (COBAS AMPLICOR NG test).For each laboratory case, a questionnaire on demographics, clinical and epidemiological characteristics, including possible risk exposures, is completed by the attending physician (questionnaire in French available on request).
In addition, all laboratories were encouraged to notify the General Directorate of Health (DGS) of the results of antimicrobial susceptibility testing when diagnosis was obtained through culture.Microbiological methods used to test antimicrobial susceptibility were based on the CLSI recommendations [19], using the disk diffusion technique in combination with a beta-lactamase test.The minimum inhibitory concentration (MICs) was determined by the E-test method (AB Biodisk, Solna, Sweden) to define a ciprofloxacin-resistant strain.A strain was classified as susceptible to ciprofloxacin if the MIC was <0.06 mg/L or the diameter ≥47 mm; it was classified as intermediate if MIC was 0.06-1 mg/L and resistant if CMI was >1.
Associations between quinolone-resistant gonococcal infection and specific exposures were examined by univariate analysis, calculating odds ratios (OR) and 95% confidence intervals (95 CI).The c 2 test was used to compare proportions between groups.Multivariate analysis was performed using logistic regression models; variables of clinical and statistical importance (having a p value <0.2 in univariate analysis) were used to build the model.Multivariate models were built by the backward method and the model with the best log-likelihood value was chosen.Analysis was performed with SPSS v. 14.

Results
A total of 238 cases were reported between 2002 and October 2005.Median age was 35 years and 75% of cases were aged under 41 years.The male/female sex ratio was 9:1.79% of cases named Switzerland as the country where infection was acquired.67% reported acquiring the infection during a single sexual encounter.40% of male cases reported homosexual contacts, and of the 68 individuals with known HIV status 24 (35%) were HIV positive.
Seven (3%) of all reported cases had at least two separate episodes of gonococcal infection during a maximum period of 5 months.Four of them were males reporting a homosexual contact.One was HIV positive.Three of these cases had a second infection within one month and this was considered a recurrence of the same infection.Of these three recurrences, two were ciprofloxacinresistant.
Four variables were kept in the final logistic regression model (log-likelihood 73.2, p value 0.008): sexual contact outside Switzerland, type of sexual contact (homosexual/heterosexual), nationality and age (Table 2).Adjusting for those factors only sexual contact outside Switzerland was found to be associated with CR infection (adjusted OR 7.0, CI 95: 1.99-24.6).
As a consequence of the detection of an increasing number of Neisseria gonorrhoeae cases in 2002 in Geneva, and of increasing resistance to quinolone in 2004 by Unilabs Genève laboratory, Geneva cantonal health authorities supplemented laboratory-based notification with a voluntary physician-based surveillance system to improve the description and monitoring of cases' epidemiological features.
We report the results obtained by the Geneva surveillance system in describing all gonorrhoea cases comparing patients with quinolone-resistant and quinolone-susceptible strains.

Discussion
Neisseria gonorrhoeae ciprofloxacin resistance is emerging in Geneva: 7% of strains tested in 2002 were CR compared to 47% at the end of October 2005, nearly a sevenfold increase.Similar trends have been observed in other European countries [6].
Descriptive analysis suggests that quinolone-resistant infection emerging in Geneva is more common than quinolone-sensitive infection among Swiss men aged below 40 who had sexual contacts with other men.When adjusting for potential confounders, sexual contact outside Switzerland, as the presumptive way of acquiring N. gonorrhoeae infection, is the only significant risk factor for a ciprofloxacin-resistant gonococcal infection.
These results are subject to several limitations.Antimicrobial susceptibility testing was only available for one third of all cases; however, considering the distribution of the main variables among tested and non-tested patients, those with a known susceptibility profile may be considered representative of all the patients.Information was obtained on a voluntary basis through physicians and CR patients were, perhaps, more likely to be tested in the event of recurrence, a potential source of selection bias.This may cause the prevalence of antibiotic resistance among patients with N. gonorrhoeae infection to be overestimated.The sample size available for the analysis of risk factors associated with CR was small.Notwithstanding the validity of the statistical results, caution should be exercised in referring them to the general population.
Physicians should be encouraged to request culture and sensitivity tests prior to treatment and to reconsider ciprofloxacin treatment for all patients with Neisseria gonorrhoeae infection, particularly in regard to persons at increased risk.
Antibiotic resistance is increasingly compromising the effective treatment of gonorrhoea and has become a major public health concern.Clinical treatment and public health strategies for control of emergent STI will benefit from the combined availability of microbiological and epidemiological data.Our findings demonstrate the importance of gonococcal antimicrobial resistance surveillance at the national level if we are to remain responsive to the changes in epidemiological trends and regularly update recommendations on first-line treatment of STI.

Figure 1
Figure 1Percentage of isolates susceptible and resistant to ciprofloxacin among Neisseria gonorrhoeae cases.Geneva, January 2002 -October 2005.