The objective of this study was to assess the utility of a supplementary PCR test following a positive Abbott m2000 PCR test result for Neisseria gonorrhoeae (NG) issued from urogenital specimens tested in the Department of Bacteriology at the Bordeaux University Hospital, France in 2011–2013.

All NG-positive PCR specimens either negative with NG culture or without culture result, were retrieved and tested with two CT/NG PCR tests: Cepheid GeneXpert CT/NG and Roche cobas 4800 CT/NG (both targeting two genes). Analytical sensitivity of the three tests, Abbott, Cepheid and Roche for NG detection, was tested on serial dilutions (10–1 to 10–9) from one colony of NG suspended in cobas^® PCR medium (Roche).

Of 11.010 specimens tested on Abbott m2000, the global prevalence of NG was 2.5% (277/11.010) and varied with the origin of the patient, from 0.8% at a free and anonymous STI screening centre (CDAG) to 6% at STI clinics (CIDDIST). Out of 215 specimens analysed, 112 were confirmed by culture or with a second PCR-positive sample, and 103 were tested with the two other PCR assays. A total of 197/215 NG-positive Abbott PCR results were confirmed. The overall positive predictive value (PPV) of the CT/NG Abbott test was 91.6%, ranging from 73.5% for asymptomatic patients consulting at CDAG to 95.8% for symptomatic patients consulting at CIDDIST. Concerning the analytical sensitivity, the Cepheid test was 10 and 100 times more sensitive than the Abbott and the Roche tests, respectively.

In populations where the prevalence was < 1%, the Abbott CT/NG test had a PPV < 90% and therefore required confirmation testing. When NG screening is associated with that of CT in populations with variable prevalence, it should be recommended to either use a NG PCR test with two targets or confirm a positive result by another PCR test with a different target.