Predictive value of the HIV paediatric classification system for the long-term course of perinatally infected children

Citation
L. Galli et al., Predictive value of the HIV paediatric classification system for the long-term course of perinatally infected children, INT J EPID, 29(3), 2000, pp. 573-578
Citations number
24
Categorie Soggetti
Envirnomentale Medicine & Public Health","Medical Research General Topics
Journal title
INTERNATIONAL JOURNAL OF EPIDEMIOLOGY
ISSN journal
03005771 → ACNP
Volume
29
Issue
3
Year of publication
2000
Pages
573 - 578
Database
ISI
SICI code
0300-5771(200006)29:3<573:PVOTHP>2.0.ZU;2-V
Abstract
Background To compare the Centers for Disease Control and Prevention (CDC) paediatric classification system with the long-term course of perinatal hum an immunodeficiency virus type 1 (HIV-1) infection. Methods Prospective study on 366 perinatally infected children followed-up from birth and checked at least every 2 months. Survival, smoothed hazard, adjusted hazard ratio of death, and transition probabilities in clinical an d immunological categories were outcome measures. Results Results Survival was 49% (95% CI : 40-58%) at 8 years. The risk of death wa s high before the age of 2, relatively low between ages 2 and 7, and contai ned thereafter. Children did not advance through the categories sequentiall y. Survival at 8 years was 61.7% (95% CI : 49.8-73.6%) in those children wh o had passed through clinical category A; the hazard ratio of death was 2.5 (95% CI : 1.7-3.8) for 175 (47.9%) children who skipped this category. Tra nsition probability in clinical category B was 39.9% (95% CI : 32.3-45.6%) after one year, but 59.1% (95% CI : 51.4-66.8%) after 5 years. Before 2 yea rs of age, the probability of entry into category C (40%; 95% CI: 35-45%) w as higher than that of entry into immunological category 3 (28% 95% CI : 22 -34%). Conclusions The classification system stands comparison with the clinical r eality, but the CD4-positive cell thresholds in infancy should be adjusted and category B indicator diseases better distributed to improve their predi ctive value.