Purpose of review: The CD4 lymphocyte count was identified as a key predictor of risk of AIDS defining diseases almost 20 years ago, early in the HIV epidemic. Several issues concerning its use to predict AIDS have arisen since. These include the difference between short and long-term prediction, the use of CD4 percentage compared with absolute counts, the impact of antiretroviral therapy on the predictive value of the CD4 count, the role relative to other markers such as viral load, the derivation of scores to predict AIDS diseases, the use of CD4 count as a surrogate endpoint, the role of the CD4 count nadir compared with the current value and the differential ability to predict different AIDS diseases.
Recent findings: All the above issues have been clarified further and this process has continued through 2004 and 2005. These developments are briefly described in this review.
Summary: The CD4 count remains the strongest short-term predictor of risk of AIDS so far identified in both treated and untreated patients and should continue to be a mainstay of monitoring for both untreated and treated patients.