Reader Comments

Post a new comment on this article

Evidence for Non-Sexual HIV Acquisition during a Study of Male Circumcision

Posted by plosmedicine on 30 Mar 2009 at 23:49 GMT

Author: David Gisselquist
Position: independent consultant
Institution: independent
Submitted Date: December 09, 2005
Published Date: December 15, 2005
This comment was originally posted as a “Reader Response” on the publication date indicated above. All Reader Responses are now available as comments.

From a randomized controlled trial of male circumcision for HIV prevention, Auvert and colleagues report a 60% lower rate of HIV incidence in the intervention compared to the control group (0.85 compared to 2.11 infections per 100 person-years [PY])[1]. However, from reported data, it is not clear how much circumcision protected men from sexual acquisition of HIV infection.

Auvert and colleagues report 23 incident infections during 2,076 PYs in men reporting no sexual contacts without a condom during relevant 3 or 9 month observation intervals, and 46 infections during 2,498 PYs in men reporting 1 or more sexual contacts without a condom. From this data, the population attributable fraction of HIV incidence associated with any vs. no sexual contacts without a condom is 27%. This suggests that 73% of observed incidence was from non-sexual exposures (assuming accurate reporting of sexual behavior and that condoms prevent HIV transmission).

To get a better view of the relationship between sexual exposures and HIV incidence in this study, I would like to see the rates of HIV incidence per 1,000 reported coital acts unprotected by condoms for men who have and have not been circumcised. I would like also to see these rates for men reporting various numbers of unprotected contacts in an interval. For example, for men reporting no unprotected contacts, the rate for circumcised and uncircumcised men together (23/0) is undefined. If rates of HIV incidence per 1,000 unprotected contacts are also high for men reporting low numbers of unprotected contacts in an interval, this will support the suspicion that non-sexual exposures were responsible for much of the observed incidence.

Evidence that sexual exposures were responsible for only 27% of HIV incidence appears inconsistent with the observed 60% lower incidence in circumcised than in uncircumcised men. Some of the observed lower incidence in circumcised men may be due to less non-sexual HIV acquisition during their less frequent health care for genital complaints[1,2]. Some men many have misreported their sexual behavior; however, insofar as assumed misreporting is relied on to explain what happened, any conclusion that circumcision protects would not be based on evidence, but rather on assumptions that overcome evidence.

Other puzzles in the study include much higher HIV incidence for men aged more vs. less than 21 years, for men recruited before vs. after 30 December 2002, and for men during the 13th to 21st months of follow-up vs. the 4th to 12th months. Does this evidence provide any clues to blood exposures and non-sexual acquisition?

It may be that non-sexual HIV acquisition in this study cannot be identified and delimited. If so, it's not clear how convincing this research will be, or should be, in promoting circumcision for HIV prevention. Evidence from other studies suggesting a protective effect from circumcision may already be sufficient to persuade uncircumcised men in long-term partnerships with HIV-positive women to seek circumcision. However, without good evidence that circumcision prevents not less than some specified number of HIV infections per 1,000 coital acts with HIV-positive women, is it responsible to recommend circumcision to men who are anticipating possible coitus with unknown non-spousal partners? Data reported from this study do not (yet) provide that evidence.


1. Auvert B, Taljaard D, Lagarde E, Sobngwi-Tambekou J, Sitta R, et al. (2005) Randomized controlled intervention trial of male circumcision for reduction of HIV infection risk: The ANRS 1265 Trial. PLoS Med 2(11): e298.

2. Potterat JJ, Brewer DD, Muth SQ, Brody S (2005) The protective effect of male circumcision as a faith lift for the troubled paradigm of HIV epidemiology in sub-Saharan Africa. Available at: http://medicine.plosjourn.... Accessed 15 November 2005.

No competing interests declared.