[Skip to Content]
Access to paid content on this site is currently suspended due to excessive activity being detected from your IP address Please contact the publisher to request reinstatement.
[Skip to Content Landing]
December 20, 1995

Percentage of CD4 Lymphocytes and Risk of AIDS-Reply

Author Affiliations

Institut Pasteur Paris, France
National Cancer Institute Rockville, Md

JAMA. 1995;274(23):1836. doi:10.1001/jama.1995.03530230022017

This article is only available in the PDF format. Download the PDF to view the article, as well as its associated figures and tables.


In Reply.  —Log or square root transformation of CD4 values is quite reasonable to assess their relationship to AIDS because the relative hazard increases in a nonlinear way. The effect of such transformation on other predictive markers, however, depends on their mutual relationships. In the three-variable Cox proportional model of our baseline values, the relative hazards of AIDS were 2.59 (95% confidence interval [CI], 1.02 to 6.56) with a decrement of one log in CD4 percentage, 2.46 (95% CI, 1.04 to 5.88) with p24 antigenemia, and 1.77 (95% CI, 1.01 to 3.10) per log10 HIV-1 RNA. Thus, even if there is a marginally better fit of the CD4 values, p24 log10 antigenemia and HIV-1 RNA levels remain independent predictive markers of AIDS risk.

First Page Preview View Large
First page PDF preview
First page PDF preview