WEIGHT-LOSS PRIOR TO CLINICAL AIDS AS A PREDICTOR OF SURVIVAL

In this analysis the aim was to determine the independent effect of mo derate to severe weight loss prior to an AIDS diagnosis on survival af ter AIDS. The study was conducted as part of the Multicenter AIDS Coho rt Study (MACS), a longitudinal study of HIV-l-seropositive gay or bis exual men. Measured weight and self-reported weight loss data were col lected semiannually from 1984 through 1993. The study population inclu ded 962 HIV-1-seropositive men who developed clinical AIDS during the follow-up period. Median survival after AIDS was significantly lower f or men with measured weight loss of greater than or equal to 4.5 kg 3- 9 months and 3-15 months prior to AIDS, or who had lost >10% of their baseline body weight compared with men with less weight loss or weight gain. Men with self-reported unintentional weight loss of greater tha n or equal to 4.5 kg 3-9 months prior to AIDS had significantly poorer survival (median = 1.05 years vs. 1.48 years;p = 0.0001) compared wit h men not reporting weight loss. After adjusting for potential confoun ding factors, men in the high measured weight loss group 3-9 months pr ior to AIDS still had significantly poorer survival [relative hazard ( RH) = 1.36; p = 0.02]. Similar trends were seen for the two longer int ervals prior to AIDS (RH = 1.38, p = 0.01; and RH = 1.50, p = 0.02, re spectively). Men who self-reported weight loss greater than or equal t o 4.5 kg 3-9 months prior to AIDS also had significantly poorer surviv al after AIDS (RH = 1.43; p = 0.002) in multivariate analysis. Analyse s were also conducted excluding men who had AIDS diagnoses known to be associated with wasting and poor prognosis, but the associations repo rted above did not materially differ. Moderate to severe weight loss p rior to AIDS is an independent predictor of survival after AIDS.