icon-    folder.gif   Conference Reports for NATAP  
  19th Conference on Retroviruses and
Opportunistic Infections
Seattle, WA March 5 - 8, 2012
Back grey_arrow_rt.gif
Higher Death Rate With 350-499 CD4s vs 500+ in Virologic Responders
  19th Conference on Retroviruses and Opportunistic Infections, March 5-8, 2012, Seattle

Mark Mascolini

SMART and ESPRIT study participants who did not inject drugs, took antiretrovirals, and had a low viral load and a CD4 count above 500 did not have a higher death rate than the general population, according to a 3280-patient analysis [1]. But virologic responders with CD4 counts stuck between 350 and 500 did die at a higher rate than people in the general population.

Some observational cohort analyses suggest that life expectancy among low-risk people with a sustained response to antiretroviral therapy (ART) approaches (but usually still lags) that of the general population. Because incomplete ascertainment of death can compromise cohort study findings, SMART and ESPRIT investigators conducted this analysis of people in the control arms of those trials, that is, people receiving standard-of-care treatment for HIV infection.

The researchers focused on non-injection drug users with a concurrent viral load below 400 copies (in SMART) or below 500 copies (in ESPRIT) and a CD4 count at or above 350 in the past 6 months. Everyone was between 20 and 70 years old. The researchers calculated standardized mortality ratios (SMRs) by comparing death rates in SMART and ESPRIT participants with rates in the Human Mortality Database stratified by country, age, and gender.

This mortality study included 1971 patients from SMART and 1309 from ESPRIT, of whom 2615 (80%) were men. While 53% of study participants lived in North America, 45.5% lived in Europe or Australasia, and the rest in South America or Asia. Men who have sex with men made up 61% of the study group, and people infected heterosexually accounted for 33%. Median age at trial randomization was 43 (interquartile range [IQR] 37 to 50) and median CD4 count 535 (IQR 420 to 724).

During follow-up 62 of 3280 trial participants died to yield a mortality of 5.02 per 1000 person-years (95% confidence interval [CI] 3.85 to 6.43). Cardiovascular disease or sudden death accounted for 19 deaths (31%), followed by non-AIDS malignancy (19%), accident, suicide, or violent death (18%), non-AIDS, nonhepatitis infection (10%), and liver disease (8%). AIDS caused only 2 deaths.

People with a most recent CD4 count above 500 accounted for 34 of these deaths, compared with a virtually identical expected death rate of 33.96 according to age, gender, and country in the general population. That meant these virologic responders with a 500-plus CD4 count had essentially the same life expectancy as the general population (SMR 1.0, 95% CI 0.69 to 1.40).

Trial participants with a latest CD4 count between 350 and 499 accounted for 28 deaths, compared with an expected death rate of 15.86 according to age, gender, and country. Those numbers meant people in the 350-499 CD4 bracket had a 77% higher death rate than the general population (SMR 1.77, 95% CI 1.17 to 2.55).

A sensitivity analysis that restricted values to those most recently measured determined that people with a CD4 count above 500 had a lower death rate than would be expected in the general population (SMR 0.89, 95% CI 0.59 to 1.29), while people with 350 to 499 CD4s had a 26% higher death rate (SMR 1.26, 95% CI 0.72 to 2.05). A second sensitivity analysis that expanded follow-up to include concurrent viral suppression and any CD4 count above 350 since trial enrollment calculated an SMR of 1.22 (95% CI 0.89 to 1.64) in people with more than 500 CD4s and 2.26 (95% CI 1.63 to 3.05) in the 350-to-499 CD4 group.

The research team cautioned that potential limitations in their analysis include selection bias due to a focus on optimal responders in clinical trials and data censoring for any loss of viral control or CD4 progression. But they noted that a bias could favor the general population because people with HIV tend to smoke more than HIV-negative people and have other risk behaviors that may imperil their health.

The investigators also warned that "cumulative harm associated with persistently low-level inflammation [in good antiretroviral responders] may only become apparent as people enter their sixth and later decades of life." So HIV-positive people who survive into their 60s and 70s may end up dying at higher rates than people in the general population.


1. Rodger A, Lodwick R, Schechter M, et al. Mortality in Patients with Well-controlled HIV and High CD4 Counts in the cART Arms of the SMART and ESPRIT Randomized Clinical Trials Compared to the General Population. 19th Conference on Retroviruses and Opportunistic Infections. March 5-8, 2012. Seattle. Abstract 638.