141 
Age- and Sex-specific Death Rates in ART-naïve Patients with CD4 Count above 350 cells/mm3 Compared with the General Population
Rebecca Lodwick*1, K Porter2, C Sabin1, B Ledergerber3, A Cozzi-Lepri1, P Khaykin4, A Mocroft1, L Jacobson5, S de Wit6, A Phillips1, and Study Group on Death Rates at High CD4 Count in Antiretroviral Naive Patients
1Univ Coll London, UK; 2Med Res Council Clinical Trials Unit, London, UK; 3Univ Hosp Zurich, Switzerland; 4Johann Wolfgang Goethe Univ, Frankfurt, Germany; 5Johns Hopkins Univ, Baltimore, MD, US; and 6St Pierre Univ Hosp, Brussels, Belgium
Background: It is unclear whether ART-naive patients
with high CD4 count carry a raised risk of death compared with the general
population, and whether there are trends in mortality rate according to CD4
count and viral load in this group.
Methods: Using data from 24 cohorts and
collaborations in industrialised countries, and using national death rates, we
calculated country-, age-, and sex-standardized mortality ratios (SMR),
stratifying by risk group. Included patients had at least 1 CD4 count >350
cells/mm3 while ART naive. Follow-up was counted from the time of
each CD4 count >350 cells/mm3 until the earliest of: next CD4
count; elapse of 1 year; death; or start of ART. Where available, the most
recent viral load value at the time of each CD4 count was identified.
Results: Of 47474 patients contributing 113,643
person-years of follow-up with current CD4 count >350 cells/mm3,
534 (1.1%) died. Most patients were male (75%), 50% were men who have sex with
men (MSM), 25% heterosexual, 20% injecting drug users (IDU), and 5% were in the
other or unknown risk group. The median (IQR) CD4 count under follow-up was 540
(438 to 700) cells/mm3. The SMR (95%CI) for MSM, heterosexual, and
IDU were 1.14 (0.95 to 1.36), 3.07 (2.48 to 3.76), and 9.58 (8.40 to 10.88), respectively.
In a sensitivity analysis considering only CD4 counts before January 2005
(performed to mitigate any effect of late reporting of deaths), the SMR were 1.25
(1.03 to 1.49), 3.47 (2.78 to 4.27), and 10.08 (8.81 to 11.49). After
adjustment for risk group, a higher CD4 count was associated with a decreased
risk of death (per two fold higher CD4 count: incidence rate ratio (IRR) 0.66,
95%CI 0.54 to 0.81, p <0.0001, Poisson regression). Similarly, in 66,665
person-years where a viral load was available, a higher viral load was associated
with an increased risk of death (per log10 increase: IRR 1.17, 95%CI
1.03 to 1.34, p <0.0001).
Conclusions: In HIV-infected ART-naive people with
CD4 count >350 cells/mm3, death rates tended to be raised
compared with the general population. However, for the MSM risk group this was
marginal, suggesting that some of the raised risk in other groups may be due to
confounding by other factors. Even in this high CD4 count range, lower CD4
count and higher viral load were associated with raised mortality. Although
several cohorts are linked with national death records, it is plausible that
under-ascertainment or late reporting of deaths has resulted in
under-estimation of death rates.
|
