Factors That Influence Confirmation of Neisseria gonorrhoeae Positivity by Molecular Methods

Several nucleic acid amplification tests (NAATs) with high sensitivity exist. However, the specificity of NAATs may be suboptimal, particularly for extragenital biospecimens. Consequently, confirmation with a second NAAT is common, although this represents a burden on resources. Furthermore, the rationale for confirmation is contentious. The objective of this work was to assess confirmation in over 13,000 screen-positive samples representing various biospecimens and three separate screening assays, the Abbott RealTime CT/NG (Abbott Molecular, Inc., Des Plaines, IL), the Cobas CT/NG test (Roche Molecular Systems Inc., Alameda, CA), and the BD ProbeTec ET CT/GC amplified DNA assay (BD Diagnostics, Sparks, MD). Factors predictive of confirmation were determined via logistic regression involving sex, year, whether the sample was formally validated, and sample site. Level of confirmation varied according to screening assay (96.2%, 86.0%, and 73.9% for the Abbott, Roche, and BD tests, respectively) in sample types formally included according to the manufacturers' instructions (i.e., validated). Sex did not affect confirmation for 2/3 assays, and the likelihood of confirmation of samples not formally included in manufacturer instructions (i.e., nonvalidated) was 89.1%, 82.1%, and 59.2% for the Abbott, Roche, and BD tests, respectively. Rectal swabs, which are nonvalidated samples, confirmed in 91.5%, 90.1%, and 87.4% of samples initially tested with the respective assays. The requirement to confirm in validated samples is not required for all NAATs, although initial assay-specific evaluation is justified given observed variability. Rectal samples represent robust biospecimens for NAAT testing and may not require confirmation when screened with the assays described.