NEUROPSYCHOLOGICAL CHARACTERIZATION OF THE AIDS DEMENTIA COMPLEX AND RATIONALIZATION OF A TEST BATTERY

Objective: To define the neuropsychological deficits present in mild h uman immunodeficiency virus type 1 (HIV-1) associated with the acquire d immunodeficiency syndrome (AIDS) dementia complex (ADC) and to devel op a rational neuropsychological test battery for its diagnosis. Desig n: Survey. Setting: Subjects were recruited from large metropolitan ho spital outpatient clinics and were all living independently in the gen eral community. Patients: Three volunteer samples of homosexual-bisexu al men: (1) 15 patients who met clinical and research criteria for mil d ADC; (2) 27 HIV-seronegative (HIV-) controls; and (3) 17 patients wi th AIDS who were neurologically intact (NI-AIDS) who were matched with the ADC subjects by CD4 lymphocyte counts for severity of systemic HI V disease. Main Outcome Measures: Neuropsychological test performance; z score comparisons were made with the HIV-control group using 2.25-S D cutoffs for abnormality. Results: Compared with NI-AIDS subjects, pe rformance of patients with mild ADC was markedly worse in the cognitiv e areas of executive function, memory, and complex attention but not i n affect or the cognitive areas of simple motor function, orientation, language, or visuospatial construction. Within the areas of executive function, memory, and complex attention, all of the HIV- controls and 95% of the NI-AIDS subjects had impaired test performance in a maximu m of one area only. In marked contrast, 14 (93%) of the 15 patients wi th mild ADC had abnormal test performances in all three of these cogni tive areas. Using a criterion of abnormal performance in at least two of the cognitive areas of executive function, memory, and complex atte ntion, all patients with mild ADC could be differentiated from HIV- co ntrols with 100% sensitivity and specificity and from NI-AIDS subjects matched for disease severity by CD4 lymphocyte count with 100% sensit ivity and 94% specificity, which increased to 100% with the requiremen t of impairment in all three cognitive areas. Conclusions: If time con straints or patient compliance limit neuropsychometric testing, examin ation to detect mild ADC first should be directed to the areas of exec utive function, memory, and complex attention. This pattern of neurops ychological deficits in patients with mild ADC is suggestive of subcor tical dementia.