PREDICTIVE ABILITY OF ACUTE PHYSIOLOGY AND CHRONIC HEALTH EVALUATION-II SCORING APPLIED TO HUMAN IMMUNODEFICIENCY VIRUS-POSITIVE PATIENTS

Objective: To evaluate the predictive ability of the Acute Physiology and Chronic Health Evaluation II (APACHE II) prognostic scoring system when applied to human immunodeficiency virus (HIV) seropositive patients in the medical intensive care unit (ICU). Design: A retrospective chart review. Setting: An urban university hospital serving the local community population and also functioning as a tertiary care referral center. Patients: All HIV-positive patients who were discharged from the Yale-New Haven Hospital medical ICU between October 1, 1986 and September 30, 1991. Interventions: None. Measurements and Main Results: APACHE II scoring significantly underestimated the mortality rate in our patient population (n = 161) (35.5% estimated vs. 44.1% observed, p < .025). When patients were evaluated according to total lymphocyte count, APACHE II scores accurately predicted the mortality rate of all patients with a total lymphocyte count of greater than or equal to 201 cells/mm(3) (n = 112) (32.6% estimated vs. 33.0% observed). However, APACHE II scoring significantly underestimated the mortality rate in the group of patients with a total lymphocyte count of less than or equal to 200 cells/mm(3) (n = 36) (44.2% expected vs. 61.1% observed, p < .05), particularly those patients with pneumonia or sepsis (n = 14) (50.5% expected vs. 85.7% observed, p < .01). Conclusion: APACHE II scoring significantly underestimates mortality risk in HIV-positive patients admitted to the medical ICU with a total lymphocyte count of less than or equal to 200 cells/mm(3). This finding is particularly true regarding patients admitted due to pneumonia or sepsis.