Re: Re: Re: HIV in South Africa 27 June 2003
Previous Rapid Response Next Rapid Response Top
Matthew L Grove,
Consultant Rheumatologist

Send response to journal:
Re: Re: Re: Re: HIV in South Africa

In response to Chris Tyler's posting of the 26/6/03

My contention with Dr. Papadopulos-Eleopulos was with regard to her statement that because HIV is most strongly associated with the act of passive anal intercourse, then

"Since the main and absolutely necessary property of sexually transmitted agents is bidirectionality, that is, transmission from the passive to the active partner and vice versa, this means HIV cannot be sexually transmitted."

This is of course true for a heterosexually transmitted STD. But not for one spreading through a homosexual community, where individuals may be either active or passive partners in sexual intercourse.

If bidirectional transmission is to be defined as transmission that can occur in either direction per sexual act, then I have indeed erred in my statement.

I should instead of taken issue with the idea that bidirectionality is "absolute and necessary" for a disease to spread through a male gay population, as multiple unidirectional transmission events could clearly account for spread. It is worth noting that in clearly acknowledged heterosexual bidirectionally transmissible STDs (the Perth group is not going to dispute the existence of Chlamydia or Gonorrhea I hope...?), sex is between one infected and one non-infected partner (who then becomes infected), ie the disease spreads by multiple unidirectional events. The only difference in this case is that they can be between either the active or passive partner, and the roles are fixed; the partner has no choice as to whether be active or passive.

I.e. A virus transmitted most effeciently by unprotected anal intercourse from active to passive partner could spread rapidly through a promiscuous gay population. I can't understand why Dr Papadopulos-Eleopulos would dispute this when it is clearly possible; such an organism would qualify as an STD.

The next issue Chris raises is how such a virus could spread through a heterosexual population when the Padian study (1) shows that heterosexual transmission does not occur.

The objections to the Padian study have been set out in detail above, but to briefly recap:

(a) The couples were selected to be discordant for HIV, ie were self selected to be inefficient infection transmitters. Whether this was due to poor virus secretion, late stage disease, HLA type, safe sex practice - whatever - this is selection bias. They knew that one of them was seropositive, and that they were in a study of HIV transmission, and they had access to information on the theory of HIV transmission and how to avoid it.

(b) 15% of couples chose to be abstinent; 75% used condoms consistently. Ie only 20% of couples were at risk, and these "inconsistently" used condoms. Anal sexual practices declined over the course of the trial. This shows that the couples involved in the study followed the advice available as to how to minimise risk.

The Padian study neither proves or disproves whether HIV can be heterosexually transmitted from female to male; it is insufficiently powered to do this and vulnerable to bias. The only solid conclusion that can be drawn is that if you find out that you are HIV +ve, and have probably been so for some time, and your wife is still HIV -ve (despite the fact that you have been having sex for an indeterminate period whilst you were HIV +ve), and you take on and follow safe sex practice, then you can carry on having sex with your wife because the risk of transmission appears to be very low. This is good news for couples who find themselves in this position ... but not grounds to say heterosexual sex is necessarily safe. Particularly if your partner carries another STD.

I am running short of time so will briefly address the last point, how HIV seropositivity appears to be spreading too rapidly in South Africa for heterosexual transmission to credibly be the only route. I find the arguments that another vector might be involved (dirty needles etc) quite convincing, and certainly worthy of serious study. I don't feel this is inconsistent with HIV positivity being caused by a blood borne (and occasionally sexually transmitted) retrovirus, though.


Matt Grove

(1) Padian NS. Shiboski SC. Glass SO. Vittinghoff E. Heterosexual transmission of human immunodeficiency virus (HIV) in northern California: results from a ten-year study. American Journal of Epidemiology. 146(4):350-7, 1997 Aug 15

Competing interests:   None declared