Symptomatology of HIV-related illness and community-acquired illness in an HIV-infected emergency department population

Study hypothesis: Community-acquired infections and non-AIDS-related illnesses are a significant proportion of the final diagnoses in HIV-infected patients presenting to an emergency department. We hypothesized that emergency physicians overdiagnose opportunistic infections in the HIV-infected patient. We also hypothesized that the absolute CD4 lymphocyte level could be used to stratify patients by likelihood of HIV related disease. Methods: We retrospectively reviewed ED logbooks and medical records to find all ED patients with self-reported HIV seropositivity during a 19-month period. Age, sex, insurance status, chief complaint(s), ED assessment, and disposition were recorded from the ED logs; absolute CD4 lymphocyte counts, risk factors, and final diagnoses were recorded from the medical records. HIV-related disease was evaluated with the use of established Centers for Disease Control and Prevention criteria. Data were evaluated with the use of the chi(2) test, the chi(2) test for trend, and kappa-proportions. Results: Analysis of 344 ED visits demonstrated that decreasing absolute CD4 lymphocyte counts were associated with increasing incidence of HIV-related disease (P <.001), even when noninfectious causes were excluded. Only 34% of visits were related to HIV-associated illness. Emergency physicians exhibited high sensitivity (72.9%) and specificity (95.5%) in diagnosing HIV-related disease and conducted appropriate visit disposition. Conclusion: ED visits by HIV-infected individuals are often not made for reasons of opportunistic infection, and the absolute CD4 lymphocyte count is inversely related to HIV-related disease.