Inability to predict diagnosis in febrile intravenous drug abusers

Although hospitalization is recommended for all febrile intravenous drug abusers, this practice has not been tested and validated. To determine the distribution of disease and the predictive value of clinical information available in the emergency room for diagnosis in these patients, we prospective...

Full description

Saved in:
Bibliographic Details
Published inAnnals of internal medicine Vol. 106; no. 6; p. 823
Main Authors Marantz, P R, Linzer, M, Feiner, C J, Feinstein, S A, Kozin, A M, Friedland, G H
Format Journal Article
LanguageEnglish
Published United States 01.06.1987
Subjects
Online AccessGet more information

Cover

Loading…
More Information
Summary:Although hospitalization is recommended for all febrile intravenous drug abusers, this practice has not been tested and validated. To determine the distribution of disease and the predictive value of clinical information available in the emergency room for diagnosis in these patients, we prospectively evaluated the clinical and laboratory data for 87 consecutive admissions involving 75 intravenous drug abusers with temperature of 38.1 degrees C or more, emergency room physicians' diagnostic predictions, and final diagnosis. Final diagnoses were pneumonia in 38% of the patients, trivial illness (viral syndrome, pharyngitis, or pyrogen reaction) in 26%, infective endocarditis in 13%, and other conditions in 23%. Neither emergency room physicians' diagnostic predictions nor clinical data correlated with a final diagnosis of endocarditis. Although physicians' prediction of trivial illness was associated with a final diagnosis of trivial illness (p less than 0.05), 29% of these patients had a more serious final diagnosis. These data confirm the need to hospitalize all intravenous drug abusers presenting with fever at an emergency room.
ISSN:0003-4819
DOI:10.7326/0003-4819-106-6-823