High viral load and symptoms of primary infection mean worse survival, says study conducted in HIV-positive sex workers in Kenya

A high viral load soon after infection with HIV and severe symptoms caused by primary HIV infection are associated with an increased risk of death, according to a study conducted amongst female sex workers in Kenya and published in the May 1^st edition of Clinical Infectious Diseases. The identification of these factors could distinguish, the investigators believe, which individuals would benefit from the timely initiation of antiretroviral therapy.

Although two-thirds of all those with HIV live in southern Africa very few studies looking at the natural history of HIV infection have been conducted in this region of the world. Therefore, in 1993, Kenyan and American investigators established a prospective cohort involving female sex workers in Mombasa, Kenya. Their aim was to gather data on HIV acquisition, biological and clinical events during HIV seroconversion.

HIV-negative sex workers, most of whom worked as barmaids, were recruited to the study and asked to attend monthly follow-up appointments where they provided blood samples and were asked to describe any symptoms suggestive of primary HIV infection.

Between 1993 and 2004 a total of 1579 women were recruited to the study. Of these, 218 became infected with HIV and were eligible for the investigators’ current analysis. Symptoms consistent with primary HIV infection were mentioned by 175 (76%) women, the most common being fever (61%) and headache (49%), with diarrhoea (17%) and rash (8%) being among the other symptoms reported.

A viral load was obtained from 168 (77%) women between four and 24 months of seroconversion, the median viral load being 50,000 copies/ml. CD4 cell counts were only available for 84 (34%) women, the median being 498 cells/mm³.

The median duration of follow-up was 4.6 years. A total of 40 women died, the mortality rate during the first five years of follow-up being two deaths per 100 patient years, increasing to ten deaths per 100 patient years between years five and nine. Cumulative survival at the end of nine years was 51%. The investigators note that this is at the ‘lower end’ of survival seen amongst HIV-positive individuals in industrialised countries in the era before effective anti-HIV therapy became available.

Viral load between months four and 24 was strongly associated with mortality, with each 1 log₁₀ copies/ml increase in viral load resulting in a two-fold increase in the hazard ratio of death (HR = 2.21, p = 0.001). Women with an initial viral load over 50,000 copies/ml had a median survival of seven years, compared to almost nine years survival in women whose viral load was initially between 10,000 – 49,999 copies/ml. The investigators note that there were only three deaths among the 45 women whose first viral load was below 1,000 copies/ml and that 85% were still alive after ten years of follow-up.

Women with a higher CD4 cell count in the period after seroconversion had better survival, but this did not reach statistical significance (p = 0.09).

Each symptoms of acute HIV infection was associated with a 14% increase in the risk of death. Median survival was just under eight years amongst women who had five or more symptoms of primary infection compared to 8.6 years amongst women with two – four symptoms. Among women who had only one symptom or no symptoms suggestive of primary HIV infection the survival rate at ten years was 70%.

The investigators then looked to see if any particular symptoms were associated with an increased risk of death and noticed that patients experiencing diarrhoea as part of their HIV seroconversion illness were statistically more likely to die (p = 0.05) and that there was a non-significant trend for rash to be associated with an increased risk of death (p = 0.1).

In multivariate analysis, a higher initial viral load (p = 0.001), and greater number of symptoms of primary infection (p = 0.05) both remained significantly associated with an increased risk of death.

The regular monitoring which the women in this study received is highlighted as a methodological strength by the investigators, although they do acknowledge weaknesses, not least that the study population were sex workers and it may not be possible to extrapolate these results onto other populations of African women.

“Our results suggest that biologic markers can identify individuals early in the course of HIV infection who are at risk of more rapid disease progression”, conclude the investigators, adding, “this may allow more targeted clinical monitoring and timely initiation of treatment, especially as antiretroviral therapy becomes increasingly available in Africa.”