点击显示 收起
Departments of Infectious Diseases and Microbiology and Infectious Disease Epidemiology, Imperial College London, St. Mary's Campus, HIV and STI Department, Communicable Disease Surveillance Centre, Health Protection Agency Centre for Infections
Centre for Sexual Health and HIV Research, Departments of Primary Care and Population Sciences, Royal Free and University College London Medical School, London, United Kingdom
The percentage of quinolone-resistant Neisseria gonorrhoeae isolated in London increased between 2000 and 2003, from 0.9% to 7.9% of total isolates. This increase was investigated by genotyping resistant isolates and comparing demographic and behavioral data. In 2000, resistant isolates predominantly had unique sequence types (STs) that were associated with imported infection, whereas, in 2002 and 2003, large ST clusters of indistinguishable isolates were associated with endemic acquisition. Resistant isolates that belonged to these large clusters were typically from patients who had similar epidemiological characteristics (such as ethnicity and sexual orientation) and behavioral characteristics (such as multiple sex partners and previous gonorrhea). In London, quinolone resistance is no longer associated with importation from areas of high prevalence and is spreading endemically in high-risk groups.
Neisseria gonorrhoeae was highly susceptible to the fluoroquinolone ciprofloxacin when the drug was first introduced for treatment of gonorrhea in the late 1980s. Initially, resistance to ciprofloxacin was sporadic and was associated with importation from certain regions of the world, such as Southeast Asia [1]. However, in many countries, it is no longer appropriate to use ciprofloxacin for gonococcal treatment, because of high levels of resistance [2, 3]. Resistance to ciprofloxacin generally indicates resistance to all fluoroquinolones.
Ciprofloxacin had been one of the recommended first-line therapies for treatment of N. gonorrhoeae infection in England and Wales [4] until 2003, when it was recommend that a change be made to the treatment guidelines to remove it from the list of first-line therapies [5], and new guidelines are being prepared that recommend the use of third-generation cephalosporins [6]. In 2002 and 2003, the prevalence of quinolone-resistant N. gonorrhoeae (QRNG) infection increased to >5% in all areas of England and Wales, as determined through national surveillance (The Gonococcal Resistance to Antimicrobials Surveillance Programme ) [7, 8]. The prevalence of QRNG infection increased rapidly during the first 4 years of surveillance, particularly between 2001 and 2002, with an overall change of 64% per year.
London has 32 different genitourinary medicine (GUM) clinics, at which the majority of gonococcal infections are treated. Partner notification (PN) by which gonorrhea-transmission contacts can be detected is problematic, given the large number of clinics and the large number of anonymous sex partners among core groups, such as men who have sex with men (MSM) and young people, who report high rates of sex partner change. The absence of adequate data from PN makes it difficult to study the transmission of gonorrhea in a large metropolitan population. Genotyping of N. gonorrhoeae has been shown to confirm sexual contacts and to reveal transmission links not apparent by PN [9].
The aim of the present study was to genotype, by N. gonorrhoeae multiantigen sequence typing (NG-MAST) [10], all QRNG recovered during GRASP in 2000, 2001, 2002, and 2003 in London. These data were combined with the demographic and behavioral data that are collected as part of GRASP, to analyze the transmission of QRNG strains over the 4-year time period.
Materials and methods.
GRASP is the national surveillance program for monitoring antimicrobial resistance in N. gonorrhoeae in England and Wales; it was commenced in 2000 [11]. The program consists of 17 sentinel GUM clinics and their supporting laboratories around the country and 9 GUM clinics and 7 laboratories in London, which provides good geographic coverage and, thus, a representative sample of gonorrhea in England and Wales. N. gonorrhoeae isolates are collected from consecutive patients, both men and women, during the months of May, June, and July, and consent is obtained. GRASP collects only 1 gonococcal isolate per patient and, where multiple sites are infected in a patient on diagnosis, the GRASP order of preference is followed (a rectal isolate before a urethral isolate and a urethral isolate before a cervical isolate, followed by any other site). Isolates recovered from the same patient <28 days apart were not included in the present study, to exclude treatment failures.
All N. gonorrhoeae isolates recovered as part of GRASP in London in 2000, 2001, 2002, and 2003 had their MICs to a range of antimicrobials determined, and isolates that were resistant (MIC 1 mg/L) or that had reduced susceptibility (MIC 0.1250.5 mg/L) to ciprofloxacin were identified [11]. All QRNG isolates were genotyped by NG-MAST [10]. Briefly, internal regions of the por and tbpB genes were amplified by polymerase chain reaction, and both strands of DNA were sequenced by use of an ABI 3700 sequencer. Sequences were aligned, edited, and trimmed to a fixed length from conserved positions, as described elsewhere [10]. Allele numbers were assigned to each different por and tbpB sequence, and the corresponding sequence type (ST) was assigned on the basis of the combination of the alleles at the 2 loci by use of the NG-MAST Web site (available at: http://www.ng-mast.net) (e.g., por-53 and tbpB-21 gives ST147). Clusters of isolates, defined as 2 isolates having the same ST, were identified. Alignments of trimmed sequences were performed by use of ClustalX (version 1.8) [12].
Demographic and behavioral data on each patient collected as part of GRASP were used in conjunction with the genotyping data. The data collected included age, sexual orientation, ethnicity, number of sex partners abroad and in the United Kingdom during the previous 3 months, other sexually transmitted infections at the time of clinic attendance, and treatment.
Results.
In 2000, 19 (0.9%) QRNG isolates were identified; the number increased to 21 (1.8%) in 2001, to 76 (7.2%) in 2002, and to 76 (7.9%) in 2003 (table 1). In 2000, QRNG isolates were predominantly recovered from heterosexual men (68%), and, during the following 3 years, the proportion of isolates from MSM increased, from 10% to 38% (table 1). Very few QRNG isolates were recovered from women during the 4 years (e.g., 4 in 2000 and 7 in 2002) (table 1). Each year, QRNG isolates were predominantly recovered from patients of white ethnicity (e.g., 58% in 2000 and 62% in 2003) (table 1). The median age of patients infected with QRNG was between 30 and 34 years (table 1). The percentage of QRNG isolates that were likely to have been acquired abroad (i.e., when the patient reported having sexual contract abroad) decreased from 33% in both 2000 and 2001 to 16.7% in 2002, and then increased slightly, to 21%, in 2004 (table 1). The predominant regions, when known, in which QRNG was acquired abroad were the Far East (43%) and western Europe (31%).
In 2000, 63% of the QRNG isolates had unique STs, and 37% belonged to ST clusters of 2 indistinguishable isolates. The extent of clustering increased in 2001 and 2002, with only 9% of isolates having a unique ST and 91% of isolates belonging to ST clusters of 2 isolates in 2002 (table 1), although, in 2003, the percentage of isolates belonging to a ST cluster of 2 isolates decreased, to 76%. The number of ST clusters increased over time: there were 3 ST clusters in 2000, 6 in 2001, 7 in 2002, and 10 in 2003. The size of the ST clusters also increased with increasing prevalence: the largest ST cluster detected in 2000 consisted of 3 isolates, and the largest ST clusters detected in 2002 and 2003 consisted of 24 and 16 isolates, respectively (table 1).
Six STs (ST147, 202, 203, 205, 206, and 211) were detected in 2 of the 4 years (table 1). ST147 (43 isolates) was present in all 4 years and was found predominantly in MSM (93%) and in patients of white ethnicity (65%). In contrast, the most-prevalent ST (ST203; 44 isolates) appeared first in 2001 and was isolated from heterosexual men only and from women of predominantly black ethnicity (59%). Having multiple sex partners in the United Kingdom and previous gonorrhea were common in patients whose isolates belonged to these persistent STs (table 2). The patients with isolates that belonged to one of the 4 most-prevalent ST clusters (table 2) were significantly less likely to have reported having had sex partners abroad than were the patients with isolates that belonged to the less prevalent ST clusters or that had unique STs (2 = 30.64; P = .001). Additionally, the patients with isolates that belonged to one of the 4 most-prevalent ST clusters were significantly more likely to have had previous gonorrhea than were the patients with a quinolone-sensitive (n = 4241) or reduced-susceptibility (n = 133) isolate (2 = 54.40; P = .001).
In 2000, 47 isolates with reduced susceptibility to ciprofloxacin were recovered; these were also typed by NG-MAST. The STs of the isolates with reduced susceptibility to ciprofloxacin were distinct from those of the QRNG isolates, except for 1 ST (ST84), which included isolates with both reduced susceptibility and resistance to ciprofloxacin. None of the isolates known to be fully susceptible to ciprofloxacin on the NG-MAST Web site (as of 28 April 2005, 710 of the 4245 entered isolates were known to be fully susceptible to ciprofloxacin), the majority of which were from the United Kingdom, belonged to the same ST as did any of the QRNG isolates.
Discussion.
Typing isolates by NG-MAST, in conjunction with the demographic and behavioral data collected as part of GRASP, has shown that the transmission of QRNG in London appears to have changed over the course of 4 years. In 2000, QRNG was predominantly isolated from heterosexuals in London and was associated with foreign travel [5]. From 2001 onward, QRNG was more frequently isolated from MSM and, in 2003, QRNG was no longer significantly associated with heterosexuals [5, 8]. In 2002 and 2003, 3 of the 4 largest ST clusters were associated with MSM (table 2).
Between 2001 and 2002, the number of QRNG isolates increased nearly 4-fold, and this increase was associated with a relative increase in the size of ST clusters. The 2 large ST clusters of 43 (ST147) and 44 (ST203) isolates were associated with MSM of predominantly white ethnicity and with heterosexuals of predominantly black ethnicity, respectively, and with a low percentage of patients reporting having had sex partners abroad. With less importation of QRNG and with its increasing prevalence and endemic transmission, the size of ST clusters increased, from the largest ST cluster (ST147) of 3 isolates in 2000 to an ST cluster (ST203) of 24 isolates in 2002, with the overall percentage of small ST clusters (<4 isolates) decreasing.
It was not possible to confirm linkage between patients with isolates that belonged to any of the ST clusters, because PN data were not available to identify definite sexual contacts, but the patients with isolates belonging to the major ST clusters had similar demographic and behavioral characteristics, suggesting that they may have been linked through the same transmission network. The patients with isolates that belonged to these ST clusters reported having multiple sex partners and previous gonorrhea. Taken together, the genotyping and demographic data suggest that QRNG strains are being transmitted in specific population groups in Londonthe members of which engage in high-risk sexual behavior and can be viewed as a core group [13, 14] and that this contributes to the increasing prevalence of QRNG infection.
Six of the ST clusters were found in several years, such as ST147 in all 4 years and ST203 in 3 years. It would appear that QRNG strains can persist in populations for several years, despite the fact that NG-MAST examines variation in antigen-encoding genes that are likely to be subject to rapid diversifying selection pressure from the host immune system and that rates of recombination are considered to be high in N. gonorrhoeae [15, 16]. Persistence of gonococcal genotypes has been previously observed in other studies [9, 17, 18].
In 2003, QRNG was isolated fairly evenly between heterosexuals (58%) and MSM (42%), and these data contrast with those from the United States. The gonococcal surveillance centers of the Centers for Disease Control and Prevention (CDC) recently reported that, in 2003, the prevalence of QRNG infection in MSM was 4.9%, whereas the prevalence in heterosexuals was very low at 0.4% [3]. The CDC has recommended that changes be made to the US treatment guidelines, such that ciprofloxacin is not used to treat MSM with gonorrhea.
The high prevalence of QRNG infection identified by GRASP [5] initiated a recommendation for a change in treatment guidelines, from ciprofloxacin to either 250 mg of ceftriaxone administered intramuscularly or 400 mg of cefixime administered orally [6]. Genotyping and enhanced surveillance should be used to monitor the effects of a major change in therapy and to track the evolution of QRNG genotypes in a population, and they could be used in a more timely manner to better identify transmission networks that facilitate the spread of antibiotic-resistant gonorrhea.
Acknowledgments
We are grateful to the London laboratories and genitourinary medicine clinics that are members of the Gonococcal Resistance to Antimicrobials Surveillance Programme (GRASP), for providing Neisseria gonorrhoeae strains and demographic and behavioral data on the patients that were collected as part of GRASP from 2000 to 2003. We thank Elizabeth Rudd, for advice.
References
1. The WHO Western Pacific Region Gonococcal Antimicrobial Surveillance Program. Antimicrobial resistance in gonococci, WHO Western Pacific Region, 1996. Commun Dis Intell 1997; 21:34953. First citation in article
2. The WHO Western Pacific Region Gonococcal Surveillance Program. Annual report of the Australian gonococcal surveillance program 2002. Commun Dis Intell 2003; 27:48891. First citation in article
3. Centers for Disease Control and Prevention. Increases in fluoroquinolone-resistant Neisseria gonorrhoeae among men who have sex with menUnited States, 2003, and revised recommendations for gonorrhea treatment, 2004. MMWR Morb Mortal Wkly Rep 2004; 53:3358. First citation in article
4. Clinical Effectiveness Group (Association for Genitourinary Medicine and the Medical Society for the Study of Venereal Diseases). 2001 national guideline on the management of gonorrhoea in adults. Available at: http://www.bashh.org/guidelines/2002/gc_0601.pdf. Accessed 17 February 2005. First citation in article
5. Fenton KA, Ison CA, Johnson AP, et al. Ciprofloxacin resistance in Neisseria gonorrhoeae in England and Wales in 2002. Lancet 2003; 361:18679. First citation in article
6. Clinical Effectiveness Group (British Association of Sexual Health and HIV). National guideline on the management of gonorrhoea in adults. Available at: http://www.bashh.org/guidelines/ceguidelines.htm. Accessed 19 August 2005. First citation in article
7. GRASP Steering Group. The Gonococcal Resistance to Antimicrobials Surveillance Programme (GRASP), year 2002 collection. London: Health Protection Agency, 2003. First citation in article
8. GRASP Steering Group. The Gonococcal Resistance to Antimicrobials Surveillance Programme (GRASP), year 2003 collection. London: Health Protection Agency, 2004. First citation in article
9. Ward H, Ison CA, Day SE, et al. A prospective social and molecular investigation of gonococcal transmission. Lancet 2000; 356:18127. First citation in article
10. Martin IMC, Ison CA, Aanensen DM, Fenton KA, Spratt BG. Rapid sequence-based identification of gonococcal transmission clusters in a large metropolitan area. J Infect Dis 2004; 189:1497505. First citation in article
11. Paine TC, Fenton KA, Herring A, et al. GRASP: a new national sentinel surveillance initiative for monitoring gonococcal antimicrobial resistance in England and Wales. Sex Transm Infect 2001; 77:398401. First citation in article
12. Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Higgins DG. The ClustalX windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res 1997; 25:487682. First citation in article
13. Yorke JA, Hethcote HW, Nold A. Dynamics and control of the transmission of gonorrhoea. Sex Transm Dis 1978; 5:516. First citation in article
14. Gunn RA, Fitzgerald S, Aral SO. Sexually transmitted disease clinic clients at risk for subsequent gonorrhoea and Chlamydia infections. Sex Transm Dis 2000; 27:3439. First citation in article
15. O'Rourke M, Stevens E. Genetic structure of Neisseria gonorrhoeae populations: a non-clonal pathogen. J Gen Microbiol 1993; 139:260311. First citation in article
16. O'Rourke M, Spratt BG. Further evidence for the non-clonal population structure of Neisseria gonorrhoeae: extensive genetic diversity within isolates of the same electrophoretic type. Microbiology 1994; 140:128590. First citation in article
17. Martin IMC, Ghani A, Bell G, Kinghorn G, Ison CA. Persistence of two genotypes of Neisseria gonorrhoeae during transmission. J Clin Microbiol 2003; 41:560914. First citation in article
18. Palmer HM, Leeming JP, Turner A. Investigation of an outbreak of ciprofloxacin-resistant Neisseria gonorrhoeae using a simplified opa-typing method. Epidemiol Infect 2001; 126:21924. First citation in article