M. Galli et al., NO EVIDENCE OF A HIGHER RISK OF PROGRESSION TO AIDS IN PATIENTS WITH HIV-1-RELATED SEVERE THROMBOCYTOPENIA, Journal of acquired immune deficiency syndromes and human retrovirology, 12(3), 1996, pp. 268-275
The prognostic role of platelet (PLT) counts was evaluated in a cohort
of 1,533 HIV-1-infected subjects followed for a median of 21 months.
Thrombocytopenia (TCP), defined as a PLT count less than or equal to 1
00 x 10(9)/L, was present at enrollment in 11.2% of cases, with counts
less than or equal to 50 x 10(9)/L (severe TCP) in 5.3%. With the sub
jects with normal PLT counts (PLT >150 x 10(9)/L) as the reference gro
up, the relative risk of developing acquired immunodeficiency syndrome
(AIDS) was 0.8 [95% confidence interval (CI) 0.5-1.3, p = 0.4] for su
bjects with severe TCP, 2.1 (95% CI 1.4-3.1, p = 0.002) for those with
PLT counts ranging from 51 to 100 x 10(9)/L (moderate TCP), and 1.6 (
95% CI 1.2-2.1, p = 0.0004) for those with borderline PLT values (PLT
ranging from 101 to 150 x 10(9)/L). Most of the risk increase associat
ed with moderate TCP and borderline PLT values was explained by a high
er prevalence of subjects with an older age and lower CD4(+) cell coun
ts. However, at multivariable analysis considering age, sex, risk grou
p, and zidovudine (ZDV) treatment, the risk for subjects with severe T
CP remained significantly lower than that for subjects with moderate T
CP and borderline values. These results suggest the existence of diffe
rent types of HIV-1-associated TCP and also suggest that severe TCP (w
hich often arises in the early phases of infection) is not related to
disease progression.