Methods

In 2008, CDC and the Association of Public Health Laboratories (APHL) convened an independent work group to evaluate available information and make suggestions for CDC to consider in the development of recommendations for the laboratory diagnosis of C. trachomatis and N. gonorrhoeae in the United States. Members of the work group* were selected on the basis of published expertise in the field of C. trachomatis and N. gonorrhoeae diagnostics or were public health laboratory directors, sexually transmitted diseases (STD) clinicians, CDC’s Division of STD Prevention, and representatives of the Food and Drug Administration (FDA) or the Centers for Medicaid Services (CMS). Four members of the work group, including three who served as co-authors, previously had published papers in which they acknowledged receiving financial support from diagnostic test manufacturers for test evaluations. These potential conflicts of interest were disclosed and managed in accordance with the editorial standards of the journals that published the scientific reports. In addition, in August 2013, to maintain objectivity and to confirm that the recommendations were evidence based, a second independent panel of microbiologic, statistical, and clinical experts reviewed the draft recommendations. Approximately 6 months before a meeting held at CDC during January 13–15, 2009, work group members were asked by conference call to identify key questions regarding chlamydia and gonorrhea laboratory diagnostics that emerged from literature reviews and discussed the information available to answer those questions. Work group members were assigned key questions to research (see Appendix) and, with the assistance of CDC and APHL staff, conducted an extensive Medline database of peer-reviewed literature published during January 1, 2000– January 1, 2009. The Medline database was searched using the terms “Chlamydia trachomatis” or “chlamydia” “Neisseria gonorrhoeae” or “gonorrhea” or “lymphogranuloma venereum” or “LGV” and “infection” or “reproductive tract” or “specimen” or “urogenital specimen” or “urine” or “rectum” or “pharynx” or “oropharynx” or “culture” or “nucleic acid amplification” or “nucleic acid probe” or “enzyme immunoassay” or “detection” or “performance” or “screening” or “adolescent” or “prevalence” or “confirmation” or “repeat testing” or “pediatric” or “sexual assault” or “sexual abuse” or “point of care” or “serology.” The key questions were categorized into three principal areas of laboratory diagnostics: 1) performance characteristics of tests, 2) screening applications, and 3) laboratory confirmation of test results. Monthly conference calls or e-mail message exchanges were conducted with work group members researching key questions in each principal area to ensure progress and adequate support in obtaining relevant publications. Work group members assigned to address key questions developed tables of evidence from peer-reviewed publications and presented these tables at the in-person meeting held in January 2009. Each key question was introduced, and publications were discussed in terms of strengths, weaknesses, and overall relevance of the data to the key questions. Scientific publications with findings derived from studies with an analytic plan involving a patient’s infected status were included in developing these recommendations. Studies using discrepant analysis were excluded. All work group members agreed with these inclusion and exclusion criteria because they approach design characteristics used by FDA when evaluating diagnostic tests for marketing in the United States. During the meeting, each topic was presented by the assigned work group member, and an open forum followed to allow all work group members and ad hoc attendees to discuss the merits of publications used to address the key questions. At the end of each discussion, a recommendation was proposed and adopted for consideration by CDC if there were no objections from the work group members. Following the in-person meeting, the same database was searched periodically for subsequently published articles for the work group to consider by e-mail and/or teleconference calls. A writing team was formed to draft the recommendations generated from these discussions, and the senior CDC author was responsible for the overall content.

Testing Technologies

Multiple laboratory test options can be used to detect chlamydia and gonorrhea although some might not be recommended for routine use based on performance. Direct detection of the pathogen using culture or nonculture methods is possible. Of the nonculture tests available, only nucleic acid amplification testing (NAAT) is recommended for routine use whereas other tests (e.g., enzyme immunoassays, nucleic acid probe tests, and genetic transformation tests) are not recommended. Serologic tests that detect a systemic immune response to infection are not recommended because of the lack of precision for the detection of an active infection.

Since 2002, improvements in chlamydia and gonorrhea NAAT technologies have enabled significant implementation and expansion of screening programs using less invasive specimen collection. Although these changes have created opportunities for more rapid and accurate chlamydia and gonorrhea diagnosis and a broader understanding of key populations at risk, they also might have created challenges (e.g., increased laboratory costs and physical design constraints requiring unidirectional specimen processing to minimize contamination when laboratories attempt to incorporate new technologies into their existing test repertoire).

The performance of NAATs with respect to overall sensitivity, specificity, and ease of specimen transport is better than that of any of the other tests available for the diagnosis of chlamydial and gonococcal infections (21–30). Culture for C. trachomatis and N. gonorrhoeae was long the reference standard against which all other diagnostic tests were compared. However, better tests have been needed because of difficulties in maintaining the viability of organisms during transport and storage in the diverse settings in which testing is indicated. In addition, the tissue culture methods for C. trachomatis isolation are difficult to standardize, technically demanding, expensive, and relatively insensitive. Thus, diagnostic test manufacturers developed nonculture tests. The first nonculture tests for C. trachomatis and N. gonorrhoeae included enzyme immunoassays (EIAs), which detect specific chlamydial or gonococcal antigens, and direct fluorescent antibody (DFA) tests for C. trachomatis, which use fluorescein-conjugated monoclonal antibodies that bind specifically to bacterial antigen in smears. These antigendetection tests were followed by nucleic acid hybridization tests, which detect C. trachomatis–specific or N. gonorrhoeae– specific deoxyribonucleic acid (DNA) or ribonucleic acid (RNA) sequences. With the availability of these nonculture tests, some of which could be automated, screening programs for C. trachomatis were initiated, and screening programs for N. gonorrhoeae began to change from culture to using the more convenient and, for remote settings, more reliable nonculture methods. The primary drawback of these tests, especially for C. trachomatis, was that they failed to detect a substantial proportion of infections (30–39). This changed with the introduction of NAATs that amplify and detect C. trachomatis–specific or N. gonorrhoeae–specific DNA or RNA sequences. These tests are approximately 20%–35% more sensitive than the earlier nonculture tests (30–39).

This report emphasizes the importance of maintaining the capability to culture for both N. gonorrhoeae and C. trachomatis in laboratories throughout the country because there are insufficient data to recommend nonculture tests in cases of sexual assault in boys and extragenital site exposure in girls. N. gonorrhoeae culture is required as a test of cure to evaluate suspected cases of gonorrhea treatment failure and to monitor developing resistance to current treatment regimens. Test of cure should be done when clinically indicated only (i.e., not part of routine care). Chlamydia culture capability also should be maintained in some laboratories to monitor future changes in antibiotic susceptibility and to support surveillance and research activities such as detection of lymphogranuloma venereum (LGV) or rare infections caused by variant or mutated C. trachomatis such as the type recently described in Sweden (40,41).

* A list of the members of the work group appears in the Appendix