John S. Moran

Chlamydia trachomatis among Patients Infected with and Treated for Neisseria gonorrhoeae in Sexually Transmitted Disease Clinics in the United States

Context On the basis of studies of chlamydia and gonorrhea co-infection in the 1980s, guidelines recommend empirical treatment for chlamydia for patients receiving treatment for gonorrhea. The relevance of these recommendations to current practice is unknown. Contribution These 1995 data suggest that, in U.S. sexually transmitted disease clinics, the prevalence of Chlamydia trachomatis among patients infected with gonorrhea (20% in men, 42% in women) or treated for gonorrhea (19% in men, 35% in women) remains sufficiently high to warrant continued empirical therapy for chlamydia. Cautions These data are nearly a decade old. They may not apply in settings that are not sexually transmitted disease clinics. The Editors In the early 1980s, shortly after techniques for culturing Chlamydia trachomatis became available, program data and expert opinion suggested that C. trachomatis coexisted in up to 45% of patients infected with Neisseria gonorrhoeae (1). On the basis of these estimates, the U.S. Centers for Disease Control and Prevention (CDC) first suggested (1) and then formally recommended (2) that all patients treated for gonorrhea should also be treated for chlamydia. Presumptive chlamydia treatment in patients treated for gonorrhea (that is, co-treatment) has remained the standard of care, a recommendation recently renewed in the CDC Sexually Transmitted Diseases Treatment Guidelines, 2002 (3). Co-treatment was proposed as a strategy for chlamydia control at a time when chlamydia culture was performed only in research settings. Over the past two decades, diagnostic testing for C. trachomatis has dramatically improved (4). Nonetheless, many public clinics do not routinely test for C. trachomatis. Furthermore, even clinics that do test for chlamydia tend to use less sensitive, earlier-generation tests (enzyme immunoassay, DNA hybridization tests) (5) because the newer, more sensitive nucleic acid amplification tests (such as polymerase chain reaction [PCR]) are more expensive. To date, there are no accurate, rapid, point-of-care tests for chlamydia. Consequently, rapid testing (Gram stain, predominantly in men) is used for gonorrhea but not for chlamydia, and sensitive testing for gonorrhea remains more accessible in everyday practice than does chlamydia testing, particularly in public health settings where reimbursement for amplification tests is typically not available. The prevalence of sexually transmitted diseases (STDs) in the United States has changed notably since 1982, when co-treatment was introduced. Reported rates of gonorrhea in the United States have declined steadily from 418 per 100 000 persons in 1982 to 149 per 100 000 persons in 1995 to 129 per 100 000 persons in 2001 (6). National chlamydia trends have been difficult to determine because of increases in screening, test sensitivity, and case reporting. In regions where screening in women has been implemented for several years, rates of chlamydial infection seem to have decreased (6). If the prevalence of chlamydia among patients treated for gonorrhea has substantially decreased, co-treatment may no longer be indicated as a clinical or public health strategy. Despite the changes in testing technology and infection prevalences, few current data are available on the prevalence of chlamydia among patients treated for gonorrhea. However, over the past decade, estimates of chlamydial infection among persons infected with gonorrhea (that is, co-infection) have varied widely, ranging from 9% to 50% among men and 24% to 67% among women (7-21). Most of these studies, however, included small numbers of patients (7-12) or used less sensitive chlamydia tests (13-17). Furthermore, they addressed only the epidemiologic question of co-infection. Since treatment decisions must often be made before test results are available, the prevalence of chlamydia among patients treated for gonorrhea is more relevant than the prevalence among patients ultimately found to be infected with N. gonorrhoeae. Our analysis had two objectives. First, to investigate co-infection, we sought to determine the prevalence of chlamydia among patients at STD clinics who were infected with N. gonorrhoeae. Second, to assess the appropriateness of co-treatment, we sought to determine the prevalence of chlamydia among patients at STD clinics who were treated for gonorrhea, including those treated presumptively as sex partners of gonorrhea-infected persons. Methods Study Design We performed a cross-sectional analysis using baseline (enrollment) data collected from July 1993 through October 1995 for Project RESPECT, a randomized, controlled trial of counseling interventions (22). The trial was conducted among patients from public STD clinics in Baltimore, Maryland; Denver, Colorado; Long Beach, California; Newark, New Jersey; and San Francisco, California. All English-speaking patients at least 14 years of age who came to the clinics for an STD examination (evaluation of symptoms, screening or routine care, or contact with an infected partner) and who reported penilevaginal intercourse during the preceding 3 months were asked to participate. Participants found to be infected with HIV and men who identified themselves as homosexual or reported having a male sex partner during the preceding 12 months were excluded. All participants gave written, informed consent, and the institutional review boards at each site reviewed and approved the protocol. Overall, 43% of eligible patients agreed to participate in the 12-month intervention and follow-up. Study participants were routinely tested for gonorrhea, chlamydia, and several other STDs. Definitions For this analysis, we defined gonococcal infection as a positive culture for N. gonorrhoeae from a urethral or endocervical swab in men or women, respectively. We defined chlamydial infection as a positive PCR result from a urine specimen or endocervical swab in men or women, respectively. We included all participants with conclusive test results from both a gonorrhea culture and a chlamydia PCR. Nongonococcal urethritis was defined as the presence of at least 5 polymorphonuclear leukocytes per high-power field and absence of gram-negative intracellular diplococci on Gram stain. Mucopurulent cervicitis and pelvic inflammatory disease (PID) were defined according to the clinicians presumptive diagnosis at the initial visit. We considered patients to have been exposed to an STD if they had a referral card or reported that their partner was infected. We defined indications for treatment by categorizing men and women into mutually exclusive hierarchical groups. We used this approach to 1) distinguish treatment decisions made at the initial consultation from those based subsequently on laboratory test results and 2) exclude patients who would be treated for chlamydia anyway. This allowed us to determine the prevalence of chlamydia among patients treated for gonorrhea who would not, in the absence of co-treatment, be treated with an antimicrobial regimen effective against C. trachomatis. We considered that the following patients had other indications for chlamydia treatment: men with nongonococcal urethritis (C. trachomatis is the organism most commonly associated with this disorder [23]); women with mucopurulent cervicitis or PID (who are treated for several potential organisms, including C. trachomatis); and patients whose partners were diagnosed with chlamydia, nongonococcal urethritis, mucopurulent cervicitis, or PID. Next, we categorized men into six groups and women into five groups on the basis of treatment indication. We considered men to have treatment indications for chlamydia if they had 1) nongonococcal urethritis by Gram stain or 2) a sex partner with chlamydia, mucopurulent cervicitis, or PID. We considered the remaining men to have treatment indications for gonorrhea if they had 3) gram-negative intracellular diplococci, 4) a sex partner with gonorrhea, or 5) a positive gonorrhea culture. The final group comprised men 6) without treatment indications for either infection. We considered women to have treatment indications for chlamydia if they had 1) a sex partner with chlamydia or nongonococcal urethritis or 2) a clinical diagnosis of mucopurulent cervicitis or PID. We considered the remaining women to have treatment indications for gonorrhea if they had 3) a sex partner with gonorrhea or 4) a positive gonorrhea culture. The final group comprised women 5) without treatment indications for either infection. Statistical Analysis We compared characteristics of patients included in this analysis with those of patients enrolled in the original study. Chlamydia prevalence was determined for patients with laboratory-confirmed gonorrhea and for patients with each of the hierarchical treatment indications. For comparison between groups, we used chi-square tests and relative risks (RRs) with 95% CIs. MantelHaenszel weighted RRs were used to calculate summary RRs for stratified analyses, for which GreenlandRobins confidence limits were calculated. The large-sample population proportion method (24) was used to calculate 95% CIs for prevalences. Role of the Funding Source Project RESPECT was supported by CDC cooperative agreements. Chlamydia PCR kits were donated by Roche Pharmaceuticals, which was not involved in study design, conduct, analysis, or interpretation. Results Study Participants Of the 4328 patients enrolled in Project RESPECT, 443 were excluded from this analysis because they did not have conclusive results from both a gonorrhea culture and a chlamydia PCR. The remaining 3885 patients (90%), 2184 men and 1701 women, were included. Our sample was similar to the sample in the original study and to the clinic populations in that patients were young (median age, 25 years), were members of racial or ethnic minority groups (59% black, 16% Hispanic, 20% white, and 6% other), had low income (<

Gonorrhea in the United States, 1981-1996. Demographic and geographic trends.

5000/y in 43%), and were approximately equally distribute

Treating uncomplicated Neisseria gonorrhoeae infections: is the anatomic site of infection important?

Objective: To describe demographic and geographic trends in gonorrhea incidence in the United States from 1981 through 1996. Study Design: We analyzed aggregate gonorrhea cases reported to the Centers for Disease Control and Prevention by the 50 states, District of Columbia, and 63 large cities. Annual incidence rates (cases/100,000 persons) were calculated. Results: Between 1981 and 1996, the incidence of reported gonorrhea decreased 71.3%, from 431.5 to 124.0 cases/100,000. However, rates among blacks were 35 times higher than rates among whites in 1996 (684.6 versus 19.4) compared with 11 times higher in 1981 (1,894.3 versus 164.3). Among women of all races, 15 to 19 year olds had the highest rates (716.6 in 1996), whereas among men, 20 to 24 year olds had the highest rates (512.9 in 1996). Southern states had higher rates than other regions. Conclusions: Large segments of the population, including adolescents, young adults, and blacks, continue to have high rates of gonococcal infection; prevention programs and health care providers should address the needs of these groups.

Ciprofloxacin for the Treatment of Uncomplicated Gonorrhea Infection in Adolescents: Does the Benefit Outweigh the Risk?

Background The efficacy of new antimicrobial regimens against Neisseria gonorrhoeae infection of sites other than the urethra and cervix is rarely adequately assessed. Goal of This Study To learn whether modern antigonococcal agents eradicate infections at some mucosal sites less reliably than at others. Study Design This was a systematic review of published therapeutic trials of various antimicrobial regimens for the biological cure of uncomplicated mucosal Neisseria gonorrhoeae infections. Data were aggregated by treatment regimen and the cure rates were calculated by site of infection. Results Of 16,737 infections, 96.4% were cured-female urethra, 98.4%; male urethra, 96.4%; cervix, 98.0%; female pharynx, 83.7%; male pharynx, 79.2%; female rectum, 97.9%; and male rectum, 95.3%. The differences between the cure rates at the pharynx and at all other sites were statistically significant in the crude analysis and after stratifying by treatment regimen. Conclusion Modern antigonococcal regimens highly effective against infection of the urethra are highly effective at the cervix and rectum as well, but pharyngeal infections are more difficult to cure.

Guide to Sexually Transmitted Disease Resources on the Internet

The highest rates of reported gonorrhea infections occur among adolescent females aged 15-19 years. Among the Centers for Disease Control and Prevention (CDC)-recommended single-dose gonorrhea treatment regimens, ciprofloxacin, a fluoroquinolone antibiotic, is approximately half the cost of other CDC-recommended oral treatment regimens. Fluoroquinolone use in patients aged <18 years has been limited because of irreversible articular cartilage damage demonstrated in large, weight-bearing joints of young animals. We reviewed the medical literature to assess whether the risks of a single 500-mg dose of ciprofloxacin to treat uncomplicated gonorrhea infection in adolescents appears to outweigh the benefits. We found no reports of irreversible cartilage toxicity or age-associated adverse events in 5236 human children and adolescents (aged 5 days-24 years) treated with a total of 5486 courses of fluoroquinolones.

Fluoroquinolones, gonorrhea, and the CDC STD treatment guidelines.

The Internet provides patients, clinicians, teachers, and researchers with immediate access to reliable information, authoritative recommendations, and the latest research findings and statistics, but quickly finding the best sources while avoiding the unreliable and obsolete can be a problem. We searched the Internet for the most useful English-language Web sites on sexually transmitted diseases (STDs), with annotations, in 4 tables: sites for patients, for clinicians and teachers, and for researchers, and sites dedicated to a single STD. In the process, we found that government-sponsored sites tended to have the most reliable information. This held true regardless of the kind of information we were seeking. Several university-sponsored sites contained information that was outdated or erroneous. Commercial and nonprofit sites sometimes evinced a bias that could mislead some readers. Both health care professionals and laypersons seeking information about STDs on the World Wide Web should generally start their search at government-sponsored sites.

Preventing tetanus, diphtheria, and pertussis among adolescents: Use of tetanus toxoid, reduced diphtheria toxoid and acellular pertussis vaccines : Recommendations of the advisory committee on immunization practices (ACIP)

To the Editor: A recent editorial1 was critical of the approach CDC has used in generating the recommendations included in the 2002 Sexually Transmitted Diseases Treatment Guidelines.2 While we are always striving to improve the process and the quality of the Guidelines, we cannot agree with the editorial. In addition to addressing the issues raised by the editorial, we would like to describe the criteria used to identify regimens recommended in the Guidelines for treatment of uncomplicated gonorrhea and highlight the public health utility of the Gonococcal Isolate Surveillance Project (GISP) in this decision-making process. The publication of a paper by Trees et al,3 which identified a multiclonal origin for ciprofloxacin-resistant Neisseria gonorrhoeae isolated in Bangkok from 1998 to 1999, prompted the editorial authors to suggest that in making treatment recommendations for the STD Guidelines, CDC should consider excluding from consideration those drugs whose “rate of emergence of resistance is high” in general, and that, in particular, the 2002 Guidelines should not have recommended use of fluoroquinolones for treatment of gonorrhea. The authors’ premise is that, given the prevalence of fluoroquinolone-resistant N gonorrhoeae (QRNG) in several parts of Asia, the apparent rapidity of the rise of resistance in Bangkok (which Trees et al3 helped to document), and the increases in resistance occurring, as cited in the editorial, even in some Western locales such as Scotland, world-wide prevalence of fluoroquinolone resistance is inevitable and imminent; therefore, treatment of gonorrhea with these drugs should be abandoned forthwith. However, as early as 1989, when the Guidelines listed ciprofloxacin as an alternative treatment for gonorrhea,4 CDC and its consultants were aware that intermediate resistance to fluoroquinolones was being identified in areas such as the Philippines.5 By the time subsequent treatment recommendations were developed, it was widely known that resistance was well established in parts of the world; the article that provided the rationale for the gonorrhea treatment recommendations in the 1993 Guidelines stated, “the future of quinolones in the treatment of gonococcal infections is in doubt.”6 However, a decision was made to rely upon data from GISP (a sentinel system to monitor antimicrobial susceptibility in N gonorrhoeae, using 25 STD clinics and five regional laboratories throughout the United States), which have demonstrated that, until very recently, fluoroquinolone resistance was not a problem in the continental United States (QRNG was first identified in Hawaii in 1991). The data from GISP indicate that in 2001 overall prevalence of QRNG was 0.7% (38/5472). However, 37 of 38 QRNG isolates were either from Hawaii or California, states where clinicians have been advised by CDC to avoid using fluoroquinolones for treating gonorrhea; QRNG prevalence for the rest of the country, excluding Hawaii and California, was 0.02% (1/4633).7 These data, demonstrating that QRNG was not prevalent in the United States, have permitted clinicians across the country to continue to use this inexpensive, safe, effective oral regimen to treat patients with uncomplicated gonorrhea. QRNG is still quite uncommon across most of the United States, and we do not believe it is possible to predict when or where that will change. In fact, contrary to the editorial’s premise, mere appearance of resistance to an antimicrobial does not imply that the resistant strain will spread or increase in prevalence; this has been noted on multiple occasions, witness the transience of N gonorrhoeae isolates with intermediate resistance to ciprofloxacin in Cleveland in the mid-1990s, and the sporadic appearance of QRNG strains in various parts of the U.S. (Anchorage, AK in 1999-2000, New Orleans, LA in 2000, or Portland, OR in 1997).7 Thus, although the editorial suggests that CDC consider the rate of emergence of resistance, this can only be done retrospectively— after the increase has occurred. We believe that treatment recommendations must be based on local susceptibility data; what is appropriate in one part of the world or country may not be appropriate elsewhere. In fact, it would not seem wise to prematurely discard a useful drug, when faced only with the potential for emergence of resistance, which might never materialize or develop only after years. The editorial also takes exception to the advice offered in the Correspondence to: Stuart Berman, MD, MS EO2, ESB, Division of STD Prevention, CDC, 1600 Clifton Road, Atlanta, GA 30333. Email: smb1@cdc.gov Received for publication February 3, 2002 and accepted February 13, 2003. John Moran is now with the National Immunization Program, CDC, Atlanta, GA Epidemiology and Surveillance Branch, Division of STD Prevention, Centers for Disease Control and Prevention, Atlanta, Georgia