SETTING: Large academic medical center in Tanzania. OBJECTIVES: To determine the etiologies and outcomes of large pericardial effusions in HIV-infected and uninfected patients. DESIGN: Prospective cohort study of patients admitted with new large pericardial effusions, confirmed echocardiographically. Patients had pericardial biopsies and drainage with extensive analysis of tissue and fluid specimens, and were followed with clinical and echocardiographic examinations. RESULTS: Of 28 patients with large pericardial effusions, 19 were infected with HIV-1. 22 had invasive diagnostic procedures: 14 of 14 HIV-infected patients, but only 4 of 8 non-HIV-infected patients, had tuberculous pericarditis (P = 0.01). All but 1 of the HIV-infected patients had strongly positive tuberculin skin tests, and short-term outcomes were similar in the 2 groups. CONCLUSION: TB is the predominant cause of large pericardial effusion in HIV-infected patients in this setting; non-HIV-infected patients are more likely to have other etiologies. These patients were at an early stage of HIV infection and responded well to treatment. In settings where microbiological studies are not routinely available, HIV-infected patients with large pericardial effusions may be treated empirically for tuberculosis and monitored for improvement. If improvement does not follow within 2-4 weeks further studies are indicated. HIV-negative patients should undergo diagnostic evaluation initially.
SETTING: Large academic medical center in Tanzania. OBJECTIVES: To determine the etiologies and outcomes of large pericardial effusions in HIV-infected and uninfected patients. DESIGN: Prospective cohort study of patients admitted with new large pericardial effusions, confirmed echocardiographically. Patients had pericardial biopsies and drainage with extensive analysis of tissue and fluid specimens, and were followed with clinical and echocardiographic examinations. RESULTS: Of 28 patients with large pericardial effusions, 19 were infected with HIV-1. 22 had invasive diagnostic procedures: 14 of 14 HIV-infectedpatients, but only 4 of 8 non-HIV-infectedpatients, had tuberculous pericarditis (P = 0.01). All but 1 of the HIV-infectedpatients had strongly positive tuberculin skin tests, and short-term outcomes were similar in the 2 groups. CONCLUSION:TB is the predominant cause of large pericardial effusion in HIV-infectedpatients in this setting; non-HIV-infectedpatients are more likely to have other etiologies. These patients were at an early stage of HIV infection and responded well to treatment. In settings where microbiological studies are not routinely available, HIV-infectedpatients with large pericardial effusions may be treated empirically for tuberculosis and monitored for improvement. If improvement does not follow within 2-4 weeks further studies are indicated. HIV-negative patients should undergo diagnostic evaluation initially.
Authors: J P Cegielski; B H Devlin; A J Morris; J N Kitinya; U P Pulipaka; L E Lema; J Lwakatare; L B Reller Journal: J Clin Microbiol Date: 1997-12 Impact factor: 5.948