While the risk of transmission through oral sex for men who have sex with men (MSM) is low, discrepancies remain between study findings and there is uncertainty about the exact degree of risk.¹

Between July 2001 and September 2003, a total of 4150 MSM were newly diagnosed with HIV infection in England, Wales, and Northern Ireland and reported to the Communicable Disease Surveillance Centre in London. Reports for 1359 cases received during this time included the question “Does the patient believe himself to have been infected through oral sex?” The remaining 2791 cases had only laboratory reports or earlier clinician report forms where this question was not asked.

The oral sex question was answered for 688 (50.6%) of the 1359 cases, of which for 625 (90.8%) the response was no, and yes for 63 (9.2%) cases. For 671 cases this information was not recorded even though the question was included on the form.

All 63 cases where the patient believed himself to have been infected with HIV through oral sex were further investigated by a discussion with the clinician or healthcare provider. From these further discussions during the follow up, 27 (42.8%) cases were believed to have been infected from unprotected anal intercourse. Of the remaining 36 cases, 16 (2.3%) claimed to have had only oral sex as their risk for acquiring HIV, with 20 (2.9%) cases always reporting protected anal sex but unprotected oral sex. Previous negative testing history and HIV status of partners was taken into account when discussing possible HIV risk with clinicians or healthcare providers.

It is difficult to quantify oral sex risks and this could be an obstacle to accuracy^1–3; none of these individuals were re-interviewed for this study and risk was assessed by clinician and note review only. There may be recall difficulties surrounding condom use, including whether they were used, or if used, coming off or splitting, or brief anal-penile contact that was not considered relevant or remembered. In addition, there was limited information about whether ejaculation had occurred or about breaks in the oral mucosa. However, 16 cases reported no anal sex and 20 cases reported only protected anal sex and unprotected oral sex. In total this represents 5.2% of those MSM reports where the question was answered. We are aware that, for half, the question was not answered, and if we classified those reports as not infected through oral sex, then 2.6% (36 of 1359) were probably infected through this route. The indication given by these UK surveillance data is that oral sex carries a small but real risk.