Background Infection with gonorrhoea (GC) or syphilis is postulated to increase the transmission of HIV 2–5 times; however, studies were done before highly effective HIV therapy was available. In Philadelphia, partner services (PS) is performed, regardless of viral load (VL), for HIV+ patients who are newly diagnosed, STI coinfected, or who are reported as contacts to a patient with new STI or HIV. We hypothesised that STI coinfection would not affect HIV transmission among partners of patients with undetectable VL receiving PS.

Methods HIV+ Philadelphia residents receiving PS from January 2012–December 2014 with a VL within +/- 6 months of PS interview date, with or without STI within +/- 90 days of PS, were included. Partners not already known to be HIV+ who tested for HIV as part of PS were categorised into contacts of either 1) HIV+ patient, undetectable VL (<50 c/ml) or 2) HIV+ patient, detectable/unknown VL.

Results PS encounters were initiated 2,463 times; 80.9% of encounters resulted in interview and 2,106 partners were elicited. Of the 1,211 locatable partners not known to be HIV+, 668 (55.1%) were tested after PS. New HIV was diagnosed more often among partners of patients with detectable/unknown VL (57/435, 13.1%) versus those with undetectable VL (17/233, 7.3%) (OR = 1.9, 95% CI 1.1–3.5). When patient VL was undetectable, there was little difference in HIV diagnoses among partners of patients with no STI (8/89, 9.0%), syphilis (7/90, 7.7%), GC (2/49, 4.1%), or syphilis/GC (0/5).

Conclusion Patients with undetectable VL who were coinfected with GC, syphilis or both did not transmit HIV to their named sexual contacts at a higher rate than those with HIV alone. Molecular sequencing data can add insight into actual transmission between partners. HIV/STI PS programs could consider deprioritizing the provision of PS to patients with undetectable VL regardless of STI coinfection.

Disclosure of interest None of the authors have any conflicts of interest to disclose.