FOR THE PAST DECADE, HIV prevention efforts in most African countries have been predicated on the assumption that heterosexual intercourse is the predominant mode of HIV transmission among adults, whereas mother-to-child transmission predominates in children. UNAIDS documents have routinely stated that 90% of adult infections are the result of heterosexual transmission, and national AIDS control programs have emphasized behavioral risk reduction as the key to prevention. During the past 3 years, however, a small group of scientists led by David Gisselquist has questioned these assumptions, publishing a series of articles—mostly in the International Journal of STD and AIDS, postulating that unsafe medical injections may be responsible for the majority of HIV infections in sub-Saharan Africa.1,2 If true, this would clearly have major implications for HIV control policies in Africa and beyond.
Since the initial publication of the papers by Gisselquist and colleagues, there has been considerable discussion of the evidence for and against their hypothesis.3–9 These articles have largely concluded that the hypothesis is incompatible with the data, pointing especially to the observed age and sex distribution of HIV infection, which is very typical of sexually transmitted infections (STIs). The very low prevalence of HIV infection in children aged 5 to 14 years in population-based studies10–12 is particularly difficult to explain given that medical injections are more common in children than adults.
A number of modeling studies have been carried out, including the study by French et al in this issue,13 to explore the degree to which iatrogenic transmission or heterosexual transmission can explain the observed HIV epidemics in sub-Saharan Africa, where adult prevalence ranges from 1% to 37%. French and colleagues use deterministic models to project the HIV epidemics that would arise from unsafe injections or heterosexual transmission given a range of assumptions concerning contact rates, mixing patterns, and transmission probabilities. Their key conclusion is that unsafe injections could not produce sustained HIV epidemics in adults without making implausible assumptions, for example, that the transmission probability from an unsafe injection is in the range 30% to 50% rather than below 3% as estimated from empiric studies of accidental needlestick exposures.14 In contrast, heterosexual transmission is able to explain epidemics compatible with those seen in Africa for a range of assumptions lying well within plausible bounds.
There are a number of limitations in the modeling carried out by French et al, most of which they have identified and discussed.13 First, they only model transmission among adults. Given the higher frequency of medical injections among children than adults, it would seem appropriate to model the effects of this, although it seems likely that this would produce an age distribution incompatible with the observed data if a substantial adult epidemic is to be modeled. Second, the modeling of transmission through injections does not seem to take into account the possibility of a higher rate of injections among HIV-infected persons resulting from HIV-related morbidity. Third, the modeling of heterosexual transmission is quite simplistic and ignores variations in infectivity during the course of HIV infection, the enhancing effects of STI cofactors, and the protective effect of male circumcision. The transmission probabilities per sexual contact required by the model to produce adult prevalences of 7.5% or 25% are in the range 0.004 to 0.016, somewhat higher than current estimates, but this may be appropriate to take into account the effects of STI cofactors and high infectivity in early and late infection that are smoothed out by the model.
Even without the use of complex models, the difficulty in explaining HIV epidemic spread on the basis of unsafe injections is evident from simple back-of-the-envelope calculations. Suppose each HIV-negative adult has two medical injections a year, of which 20% are unsafe, in accord with empiric estimates.15 If HIV prevalence in adults in the community has reached 10%, the probability of exposure to an HIV-contaminated injection is 2 × 0.20 × 0.10 or 0.04. Assuming a transmission probability from a contaminated injection on the order of 1% (higher than the most plausible estimates14), this would produce an annual HIV incidence on the order of 0.04%, less than 1/20th of typical incidence rates during the growth phase of African HIV epidemics.
Looking at transmission the other way around, let us consider one HIV-infected adult. If 20% of needles are reused, then each time the index case has an injection, there will be roughly a 20% × 0.01 or 0.002 probability that a subsequent patient will be infected, ignoring the possibility that this patient is already HIV-positive. To produce a basic reproduction number (R0) above one, as required to achieve a sustained HIV epidemic, each HIV-infected adult would therefore need to have at least 500 injections during the course of their infection, which seems highly implausible.
It is much easier to see how sexual transmission can produce a sustained epidemic. Assuming an average of 100 sexual contacts per year and a duration of infection in the era before antiretroviral therapy of around 10 years,16 there would be 1,000 chances to transmit the virus. With transmission probabilities in the absence of cofactors of a few per thousand, and much higher probabilities in the presence of other STIs and in early and late infection, R0 values above 1 are readily achievable, especially in countries with high rates of STIs and low rates of male circumcision.
Contact patterns also have a major effect on transmission dynamics as noted by French and colleagues.13 Sexual spread of HIV is greatly enhanced, especially in the early years of an epidemic, by assortative mixing patterns whereby high rates of sexual partner change are concentrated in a relatively small “core group” of the population.17 These core groups are also likely to have high rates of STI cofactors, further boosting transmission. Iatrogenic transmission seems less likely to be enhanced by assortative mixing. Although injections may be assortative in terms of age and sex groups, for example, if immunization and child health clinics are separate from general adult clinics, it seems less likely that injections would show assortative mixing of the sort assumed by French et al in their “heterogeneous” mixing scenario.
What are the implications of these observations? One of the guiding principles of medical practice is the edict to “do no harm.” There can be no excuse for even a low level of HIV transmission resulting from unscreened blood transfusions or medical injections with nonsterile equipment. In most countries, unscreened transfusions are now uncommon and only a small proportion of HIV infections are attributable to this transmission route. However, everyone would agree that measures to ensure a safe blood supply remain an essential component of national AIDS control programs. The same applies to transmission through unsafe injections. Use of nonsterile equipment for medical injections is an unacceptable practice. Although there is a consensus among the scientific community that such injections play a relatively minor role in HIV spread in most countries, as stated by the World Health Organization and UNAIDS,18 measures to eliminate them should be regarded as an essential component of AIDS control programs.
The empiric evidence combined with the results of modeling studies, including the work of French and colleagues, continues to support the prevailing view that heterosexual intercourse is the main mode of HIV transmission in sub-Saharan Africa. Countries such as Uganda, that have vigorously implemented control measures based on this premise (the so-called ABC prevention messages), have achieved substantial and sustained reductions in HIV incidence and prevalence. In other countries, rates of infection remain at very high levels. Although control programs in such countries must ensure that iatrogenic transmission is reduced to a minimum, greater reductions in HIV spread will be achieved through more vigorous measures to promote safer behavior, which need political leadership at the highest levels.
1. Gisselquist DP. Estimating HIV-1 transmission efficiency through unsafe medical injections. Int J STD AIDS 2002; 13:152–159.
2. Gisselquist D, Rothenberg R, Potterat J, Drucker E. HIV infections in sub-Saharan Africa not explained by sexual or vertical transmission. Int J STD AIDS 2002; 13:657–666.
3. Lopman BA, Garnett GP, Mason PR, Gregson S. Individual level injection history: A lack of association with HIV incidence in rural Zimbabwe. PLoS Medicine 2005; 2:142–146.
4. Schmid GP, Buve A, Mugyenyi P, et al. Transmission of HIV-1 infection in sub-Saharan Africa and effect of elimination of unsafe injections. Lancet 2004; 363:482–488.
5. Boily MC, White RG, Alary M, Lowndes C, Orroth K. Transmission of HIV via unsafe injection or unsafe sex? Anomalies or misunderstanding? Int J STD AIDS 2003; 15:61–62.
6. Kiwanuka N, Gray R, Serwadda D, et al. The incidence of HIV-1 associated with injections and transfusions in a prospective cohort, Rakai, Uganda. AIDS 2004; 18:342–344.
7. Thoma M, Gray RH, Kiwanuka N, Serwadda D, Wawer M. Unsafe injections and transmission of HIV-1 in sub-Saharan Africa. Lancet 2004; 363:1650–1651.
8. Lopman BA, Garnett GP, Gregson S, Mason PR. Authors’ reply: Don’t let the hypothesis slip. PLoS Med, 2005; 2:e147.
9. Garnett GP, Fraser C. Let it be sexual—Selection, aggregation and distortion used to construct a case against sexual transmission. Int J STD AIDS 2003; 14:782–784.
10. Killewo J, Nyamuryekunge K, Sandstrom A, et al. Prevalence of HIV-1 infection in the Kagera region of Tanzania: A population based study. AIDS 1990; 4:1081–1085.
11. Glynn JR, Ponnighaus J, Crampin AC, et al. The development of the HIV epidemic in Karonga District, Malawi. AIDS 2001;15:2025–2029.
12. Fontanet AL, Messele T, Dejene A, et al. Age- and sex-specific HIV-1 prevalence in the urban community setting of Addis Ababa, Ethiopia. AIDS 1998; 12:315–322.
13. French K, Riley S, Garnett G. Simulations of the HIV epidemic in sub-Saharan Africa: Sexual transmission vs transmission through unsafe medical injections. Sex Transm Dis 2006; 33:127–134.
14. Baggaley RF, Boily MC, White RG, Alary M. Risk of HIV-1 transmission for parenteral exposure and blood transfusion—A systematic review and meta-analysis. AIDS. In press.
15. Hutin Y, Hauri A, Armstrong G. Use of injections in healthcare settings worldwide, 2000: Literature review and regional estimates. BMJ 2003; 327:1075.
16. Morgan D, Mahe B, Okongo MJ, Lubega R, Whitworth JA. HIV-1 infection in rural Africa: Is there a difference in median time to AIDS and survival compared with that in industrialized countries? AIDS 2002; 16:597–603.
17. Gupta S, Anderson RM, May RM. Networks of sexual contacts: Implications for the pattern of spread of HIV. AIDS 1989; 3:807–817.