Background Infection with herpes simplex virus type 2 (HSV-2) is common worldwide and an important risk factor for HIV infection. Aetiological diagnosis of HSV-2 is typically determined with the use of commercially available type-specific enzyme-linked immunosorbent assays (ELISAs). This study aimed to determine the prevalence of HSV-2 among people attending sexual health clinics in the Highlands of Papua New Guinea. The study also aimed to compare the performance of two type-specific ELISA assays, the Kalon and HerpeSelect glycoprotein G2 assays, in this context.

Methods: Participants were recruited as part of a longitudinal sexual health study. Participants attended four appointments over a 12-month period and had blood taken for HSV-2 serology at each time point. Both the Kalon and HerpeSelect assays were performed as per manufacturer's instructions.

Results: A total of 132 participants were tested for HSV-2 using the Kalon and HerpeSelect ELISAs. HSV-2 prevalence was 52% (95% CI, 43-60) and 61% (95% CI, 52-69) with Kalon and HerpeSelect assays respectively. There was high concordance (87%, ?=0.75, P<0.001, n=115) between the two assays at the manufacturer recommended index value cut-offs. For participants with discordant results at baseline, (n=16), three sero-conversions were observed over the 12-month period when sequential sera was tested.

Conclusions: A high HSV-2 prevalence was observed in this clinic-based population. Our longitudinal data indicate the higher prevalence of HSV-2 detected with the HerpeSelect ELISA was likely due to false positives rather than a higher sensitivity in the early stages of infection.