Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.
Antimicrobial resistance is a major public health problem. Two recent publications have drawn attention to the growing nature of this problem, and the potential impact on the National Health Service. A report by the House of Lords1 highlighted the alarming increase in antibiotic resistance and concluded that the problem “ought to be considered much more widely than it is at present.” This was further supported by the Standing Medical Advisory Committee's report The Path of Least Resistance,2 which underscored the need to urgently prioritise antimicrobial resistance on the national agenda and recommended the creation of a national strategy for minimising the development of antimicrobial resistance. Increasing antibiotic resistance is likely to be related to the increased use of broad spectrum agents, “over the counter” use of antibiotics, and inappropriate prescribing practices. Antibiotic resistant Neisseria gonorrhoeae may be also imported into England and Wales if infections are acquired in areas where such practices are common.
In the light of these growing concerns, there is a need to evaluate current strategies for monitoring N gonorrhoeae antimicrobial susceptibility. This article takes an evidence based approach to consider the justification for and methodology of a new national surveillance system for monitoring N gonorrhoeae antimicrobial resistance in England and Wales.
N gonorrhoeae is the second most common bacterial sexually transmitted infection (STI) in England and Wales, with over 20 000 new infections being diagnosed in genitourinary medicine (GUM) clinics in 2000.3 This figure does not include infections diagnosed outside of the GUM clinic sector or undiagnosed prevalent infections in the community. Diagnoses of uncomplicated gonorrhoea in England and Wales have continued to rise since 1994,3 with substantial annual increases as large as 20% (1995 to 1996) and 30% (1998 to 1999) being observed over recent years.4,5 The reasons for the increasing incidence are unclear. However, given the correlation between gonococcal infection and high risk sexual behaviours,6 failure to maintain safer sex at the population level may be a contributing factor. Young people are most commonly affected with current rates being highest in men aged 20–24 years and women aged 16–19 years. A recent estimate suggests that of those reattending an STD clinic, 28% of men and 58% of women with gonorrhoea have unrecognised infections.7 This is worrying, given the complications of untreated infection which include pelvic inflammatory disease, infertility, ectopic pregnancy, and chronic pelvic pain.8 Further, gonococcal infection during pregnancy may lead to spontaneous abortion, stillbirth, or neonatal complications.9
Treatment failure due to antibiotic resistance, exacerbates both the transmission and the sequelae of gonococcal infections and may also be driving the increases. Antibiotic resistance in N gonorrhoeae is acquired by two different mechanisms: firstly, by plasmid transfer for penicillinase producing strains (PPNG) and high level tetracycline resistance (TRNG) and, secondly, by chromosomal mutation for penicillin and tetracycline (CMRNG), fluoroquinolones (CRNG), and other antibiotics. Antibiotic treatment should be expected to eradicate 95% of infections of uncomplicated gonorrhoea in the community.10 UK point estimates of the prevalence of PPNG include 2.6% in Scotland11 and 4%12 and 1.8%13 in London. The current UK guidelines for the treatment of uncomplicated gonorrhoea now recommend the first line use of fluoroquinolones (either ciprofloxacin or ofloxacin).14 However, ciprofloxacin resistance was first seen in the United Kingdom in the early 1990s15 and has been increasingly reported over recent years.16,17 Quinolone resistant gonococci are more prevalent in countries in the Western Pacific region. Initially, most such infections seen in the United Kingdom were acquired abroad but recent data show that more than half are now acquired through heterosexual contact within England and Wales.
Thus, a number of key questions remain. Is gonococcal antibiotic resistance a problem in England and Wales? If so, what is the burden of disease, who is affected, and how does it vary across geographic regions? What antibiotic resistance patterns of N gonorrhoeae are most prevalent and how are these changing over time? How can we best advise national prescribing policy for disease control?
Surveillance data on antibiotic resistance in N gonorrhoeae are currently provided through two main systems in England and Wales. Ongoing laboratory surveillance of resistant gonococcal isolates is provided, without charge, by the Public Health Laboratory Service (PHLS) Genitourinary Infections Reference Laboratory (GUIRL) at Bristol. Annually, approximately 3000 isolates are sent for diagnostic confirmation and quantitative antibiotic susceptibility testing from over 200 microbiological laboratories in England, Wales, and Ireland. The GUIRL provides a largely reference service and collects additional behavioural and demographic data only on resistance isolates therefore, detailed comparisons between resistant and susceptible isolates cannot be made. Overall trends observed in the GUIRL data suggest a decline in the number of PPNG isolates during the early 1990s. A large rise, however, was observed between 1997 and 1998. Moreover, the number of PP/TRNG and CMRNG isolates referred to GUIRL have been rising throughout the 1990s. During 1998, the GUIRL received 532 antibiotic resistant isolates of N gonorrhoeae, 68% occurred in men. Men aged 25–34 years and women aged 16–19 years were most likely to be infected (modal groups). Ethnic origin was available for 365 (67%) cases; 66% were in white patients, 16% black Caribbean, 4% black African, and 5% Asian; 62% reported having acquired their infections in the United Kingdom and 23% in the Far East. The distribution of resistant isolates was PPNG (17%), PP/TRNG (28%), CRNG (26%), and CMRNG (29%).18
The Gonococcal Surveillance Project at Imperial College, London, was established in 1996, and by 1999 had 13 of the larger London GUM clinics participating in a sentinel surveillance programme. Annually, over 1500 specimens were obtained during a 3 month collection period and quantitative antibiotic susceptibility testing was performed on all isolates. The Gonococcal Surveillance Project provided robust point and trend prevalence estimates of antibiotic resistance among tested isolates.13 However, this active surveillance was restricted to the London region, and the project collected relatively few behavioural and demographic data on patients. Nevertheless, prevalence estimates indicated that 4.2% of isolates showed plasmid mediated resistance with 2.4% being high level tetracycline resistant (TRNG), 1.3% both penicillinase producing and tetracycline resistant (PP/TRNG), and 0.5% penicillinase producing resistant (PPNG) only. Eight per cent of samples showed chromosomally mediated resistance (CMRNG). Ciprofloxacin resistance was found in less than 1% of samples.13
The true prevalence of gonococcal antibiotic resistance in England and Wales is unknown. However, routine surveillance and ad hoc surveys suggest that there are considerable variations between geographic locations, demographic, and behavioural risk groups. A national, cross sectional study examining the recent rise in N gonorrhoeae suggested that, during 1996, almost one in five homosexual men and about one in 16 heterosexuals presenting with gonorrhoea at GUM clinics in England had a penicillin resistant strain.19 In Scotland, isolates from men accounted for 90% of resistant isolates, with throat and rectal infections more likely than genital infections to be caused by resistant organisms.11 A London survey found an association between ethnicity and the isolation of TRNG and PPNG, highlighting the importance of different population subgroups.20
The contribution of active national surveillance programmes
A number of industrialised countries have developed gonococcal resistance surveillance programmes including the United States, Canada, Australia, and the Netherlands. In 1986 in the United States, the Centre for Disease Control and Prevention (CDC) established the Gonococcal Isolate Surveillance Project (GISP). Data from GISP have documented important patterns of antimicrobial resistance, and have steered CDC national guidelines for gonorrhoea treatment. GISP data have highlighted the decline in PPNG and TRNG since 1991 following the implementation of recommendations for the routine use of ceftriaxone for gonorrhoea treatment21 and later the detection of fluoroquinolone resistance.22 The Australian Gonococcal Surveillance Programme 1998 annual report highlights regional differences in antibiotic susceptibility patterns—for example, the continued suitability of penicillins for use in many parts of rural Australia and high level tetracycline resistance in New South Wales.23 Surveillance initiatives in the Netherlands have highlighted the importance of commercial sex in the establishment and spread of PP/TRNG in the community.24 These existing systems demonstrate the benefits of monitoring not only resistance but also geographical, behavioural, and demographic characteristics.
The advantages of monitoring antimicrobial resistance are not restricted to national surveillance and planning. Knowledge of antimicrobial resistance patterns is important to rationalise prescribing practice and inform strategic policy development. Many centres have used resistance monitoring data to inform their local treatment regimens. Ciprofloxacin is a highly active antimicrobial agent for the treatment of gonorrhoea and its increasing use as first line therapy will produce an additional need to monitor the emergence of antibiotic resistance in this country.
GRASP: establishing national surveillance in England and Wales
In June 2000, the PHLS Communicable Disease Surveillance Centre (CDSC), the Genitourinary Infections Reference Laboratory, and Imperial College launched a new national sentinel surveillance programme: the Gonococcal Resistance to Antimicrobials Surveillance Programme (GRASP). Funded by the Department of Health until 2003, GRASP is supported by the MSSVD and the Scottish N gonorrhoeae Reference Laboratory. It aims to establish a sustainable, cost effective, national sentinel surveillance system for monitoring antimicrobial resistance to N gonorrhoeae in England and Wales. Key objectives include providing prevalence estimates for N gonorrhoeae antimicrobial resistance; to determine drifts in susceptibility to specific antibiotics; and to identify relevant clinical and epidemiological associations with gonococcal resistance in order to inform a rational and cost effective approach to antibiotic therapy for N gonorrhoeae.
The system combines laboratory and clinical data on gonococcal isolates obtained from purposely selected laboratories, and covers two distinct geographical regions: London and outside London. In the first year of collection, 13 London laboratories (previously collaborators in the London Gonococcal Working Group) and 17 outside London were selected to enable good geographical coverage with representation of all NHS regions and to maximise the number of isolates collected. All consecutive gonococcal isolates identified in participating laboratories during the months of June, July, and August were sent to either the GUIRL (outside London) or Imperial College (London) for susceptibility testing. Minimum inhibitory concentrations (MICs) of all isolates were been determined for the following antibiotics and at concentrations including (as a minimum): penicillin (0.03–4.0 mg/l), ciprofloxacin (0.002–0.125 mg/l) extended range 0.125–32 mg/l tested as necessary, spectinomycin (2–64 mg/l), tetracycline (1–32 mg/l), and ceftriaxone (0.12 mg/l, single concentration). GUM clinics provided demographic and behavioural data for each patient included in the GRASP collection. Data were obtained from routinely collected information in the case notes including sex, age, ethnic background, sexual orientation, postal area, travel history, previous infections, symptom presence, concurrent STIs, number of partners, and therapy received. Multicentre (MREC) and local (LREC) research ethics committee approval has been obtained.
CDSC collects, collates, and links the behavioural (from GUM clinics) and susceptibility data (from reference laboratories) in order to determine the distribution and determinants of gonococcal disease and antimicrobial resistance, temporal trends in both clinical and low level antibiotic resistance, and to identify potentially linked clusters (using phenotypic and genotypic methods). Although isolates obtained from sites outside GUM clinics are also collected, detailed behavioural data will only be collected from participating GUM clinics. Results from the first year's collection have now been published25 and disseminated locally (through participating centres) and nationally.
GRASP is a unique collaboration which has brought together key areas of the GUM specialty, clinical laboratories, public health medicine, communicable disease surveillance, and health policy. The initiative has also raised the possibility of creating a UK-wide network for the prospective monitoring of gonococcal antimicrobial resistance, by establishing collaborative networks with colleagues in Scotland. Through this process it is hoped that a more robust understanding of the distribution and transmission dynamics of gonococcal antimicrobial resistance will be possible.
Financial support: GRASP is funded by the Department of Health (England). Conflict of interest: None declared by all authors. Contributors: TP was lead author of this paper and is the coordinator of GRASP; KF, AT, CI, IM, and AH, are co-investigators and collaborators on GRASP and contributed to the writing of this paper; AR and GK are members of the GRASP Steering Group and contributed to the preparation of this paper.