Literature review > Issue_1 > Review Darwin et al. 

The Digene Hybrid Capture 2 (HC2) CT/GC assay is a signal amplification-based chemiluminescent assay for the detection of C. trachomatis and N. gonorrhoeae. According to the manufacturer's testing algorithm, two further specific assays, the HC2 CT-ID and HC2 GC-ID, are performed on positive samples to determine which of the two organisms is present. Presumably the rationale for using a combined screening test for these two organisms is to reduce costs in low prevalence populations, although in this evaluation high prevalence female populations attending STD clinics in two American cities were studied.

The HC2 test was compared initially with conventional culture for CT and NG. The sensitivity and specificity for detecting either organism was 93.8% and 95.9% respectively. Since culture is known to be less than 100% sensitive, especially for the diagnosis of C. trachomatis, two further tests (DFA and an in-house PCR) were performed to "adjudicate" on samples positive by HC2 but negative by culture. Using the "adjudicated" test result as gold standard, the specificity of HC2 increased to 99.8%.

If the HC2 test is to be used for screening low prevalence populations, then it is vital that its specificity is high. Can we be sure, from the results presented in this paper, that the specificity is indeed 99.8% in low prevalence populations? Probably not, for the following reasons.

Firstly, the specificity of the in-house PCR used to test specimens that were positive by HC2 and negative by culture is not stated. In all, there were 23 of these samples, of which 22 were positive by either DFA or PCR. At one site, 13/15 HC2 positive, culture negative samples were positive by PCR, but only one by DFA. We cannot be certain, from the information presented, that those 12 samples were not PCR false-positives. Secondly, it is a cardinal rule of diagnostics' evaluations that they should be performed in the populations for which the test is intended; in this case low, rather than high prevalence populations.

The discrepant analysis used in this study, according to which only those positive by either HC2 or culture were subjected to further testing, may have overestimated the sensitivity of the HC2 test, since there may have been true positives who were negative by both assays; the authors are aware of this, and have therefore referred to the "relative" sensitivity of HC2 vs culture, rather than to its true sensitivity.

Notwithstanding the reservations expressed above, these results suggest that the HC2 CT/NG test could be a useful new screening test, both more sensitive and more reproducible than culture. I would have liked to see some discussion about the cost-effectiveness of a combined screening test such as this, compared with screening for these infections individually, in different populations.

about SDI | newsletters | grants | publications | literature reviews

WHO Home - WHO Search - TDR Home - SDI Home - SDI Contact us
(c) WHO/OMS 2001