The purpose of our study was to define the time course and clinical role of monitoring serum cryptococcal antigen titers (sCRAG) in patients with AIDS-related cryptococcal disease.

A retrospective chart review was conducted. The medical records for all HIV-infected patients with positive cryptococcal antigen (CRAG) tests from January 1993 to May 1998 at San Francisco General Hospital (SFGH) were reviewed for sCRAG titer levels and clinical outcomes.

Out of the 314 patients found to have positive antigen tests, 136 met the inclusion criteria. Twelve (8.8%) had no change in titer from baseline, 6 (4.4%) had an increase, and 118 (86.8%) had a decrease. Examining the association of sCRAG with time to relapse using a variety of Cox models produced largely null results. Rate of change in sCRAG over time (slope) was not significantly predictive of time to relapse nor of time to definite relapse/probable relapse/persistent disease.

Although in the majority of patients, the sCRAG titers appeared to decrease over time, we could not detect a significant correlation between sCRAG titer results of patients who had a clinical response to treatment and sCRAG titers in patients who experienced persistent disease, probable relapse, or definitive relapse of cryptococcal disease. We conclude that follow-up monitoring of the sCRAG titer is not useful in the management of patients with AIDS-related cryptococcal disease on treatment.