CROI: Untreated HIV-positive individuals have a higher risk of death even at CD4 counts over 350

This article is more than 16 years old. Click here for more recent articles on this topic

Even with CD4 counts above 350 cells/mm3, untreated HIV-positive individuals have an increased risk of death compared with the general population, according to data presented on Wednesday at the Fifteenth Conference on Retroviruses and Opportunistic Infections in Boston.

Rebecca Lodwick of University College London, assisted by Professor Andrew Phillips, presented findings from the Study Group in Death Rates at High CD4 Counts in Antiretroviral Naive Patients — the largest dataset yet to examine whether HIV-infected individuals with high CD4 counts who have not started antiretroviral therapy have an increased risk of death compared with the population as a whole.

To date, it is unclear whether treatment-naive individuals with a CD4 count above the currently recommended threshold for starting treatment — 350 cells/mm3 according to the most recent U.K. and U.S. guidelines — have an increased risk of death compared with HIV-negative individuals.



When using a diagnostic test, the probability that a person who does have a medical condition will receive the correct test result (i.e. positive). 

person years

In a study “100 person years of follow-up” could mean that information was collected on 100 people for one year, or on 50 people for two years each, or on ten people over ten years. In practice, each person’s duration of follow-up is likely to be different.


In HIV, an individual who is ‘treatment naive’ has never taken anti-HIV treatment before.

sensitivity analysis

An additional analysis of data, also known as a “what–if” analysis, which indicates how robust the study’s results are. Specific assumptions or variables may be changed, to estimate the outcome in a range of scenarios.


A person who has never taken treatment for a condition.

Using data from 23 HIV cohorts and collaborations in industrialised countries (18 in Europe, five in North America), along with national death rates, the investigators calculated country-, age-, and sex-specific standardised mortality ratios (SMR), stratifying by HIV acquisition risk group.

The results reported at the conference were from a sensitivity analysis that considered all pre-treatment CD4 cell counts above 350 cells/mm3 before January 2005. This analysis mitigated any effect of late reporting of deaths.

A total of 46,400 individuals contributing 98,527 person-years of follow-up were included in this analysis. Most (about 75%) were male, just over 50% were gay men or other men who have sex with men (MSM), 21% were heterosexual, 21% were injecting drug users (IDUs), and 5% were part of another or unknown risk group.

Follow-up was counted from each time an individual had a CD4 count above 350 cells/mm3 until the earliest of either their next CD4 cell measurement, one year, commencement of treatment, or death. Where available, the most recent viral load measurement at the time of each CD4 count was noted.

A total of 237,553 CD4 counts were included in the analysis, with a median of three per person. Individuals with a CD4 count of 350-500 cells/mm3 accounted for about 38% of the total follow-up data, followed by 35% with 500-700 cells/mm3 and 28% with more than 700 cells/mm3.

A total of 487 individuals died during the follow-up period, for an overall rate of 4.9 per 1000 person-years. About 16% of these deaths were known to be AIDS-related and 48% were non-AIDS-related, but for more than one-third of deaths the cause was unknown.

Although the death rate overall were found to be higher than that of the general population, the difference was marginal for MSM, for whom the standardised mortality ratio was 1.37 in the sensitivity analysis. However, heterosexual men and women as a group were found to have a risk of death three times higher than that of the general population, with an SMR of 3.04. For IDUs, the risk was ten times greater, with an SMR of 10.21.

As is true of more advanced HIV disease, the risk of death was higher amongst people with lower CD4 cell counts and higher viral loads. Further analysis (not included in the oral presentation) revealed that after adjusting for risk group, for every doubling of CD4 count the risk of death fell by two-thirds. A small number of people died of AIDS-related causes despite having a high CD4 cell count.

Since the death rate in HIV-positive individuals with a CD4 count of 350 cells/mm3 or higher tended to be close to that of the general population amongst gay and other MSM, but was significantly elevated for other risk groups — in particular IDUs — the investigators suggest that the increased mortality seen in these groups was likely due to other factors besides HIV disease status.

However, they noted that whilst it was not surprising to find an elevated death rate amongst IDUs, the higher mortality rate in heterosexual men and women who did not inject drugs, as well as the borderline significant increase amongst MSM, indicated that HIV itself appears to cause some increase in mortality even for people with relatively well preserved immune function.

The data from this study will inform the ongoing debate over the ideal time to begin therapy, a subject examined in great detail by Professor Philips during a Monday plenary session at the conference.


Lodwick R et al. Age- and sex-specific death rates in ART-naïve patients with CD4 Count above 350 cells/mm3 compared with the general population Fifteenth Conference on Retroviruses and Opportunistic Infections, Boston, abstract 141, 2008.