Outcomes in patients with an emergency department diagnosis of fever of unknown origin

Sharyn Ingarfield, Tony Celenza, Ian Jacobs, Thomas Riley

Research output: Contribution to journalArticlepeer-review

11 Citations (Scopus)

Abstract

Objective: To describe the outcomes in patients given an ED diagnosis of fever of unknown origin (FUO).Methods: A retrospective analysis of ED records linked to hospital morbidity, mortality and microbiology records of patients presenting to Western Australia's teaching hospitals from July 2000 to July 2003.Results: There were 3218 presentations diagnosed with FUO, 2049 (63.7%) children (median age 1.8 years) and 1169 (36.3%) adults (median age 56.0 years). FUO accounted for 0.3% of adult and 1.5% of paediatric ED presentations. Overall, 1997 (62.1%, 95% confidence interval 60.4-63.8%) were admitted (82% adults vs 50.7% children; P < 0.001). Adults had a longer median length of stay than children (4 days vs 2 days; P < 0.001) and a higher proportion of positive blood cultures (admissions 15.1% vs 4.9%; P < 0.001) commonly with Escherichia coli. Streptococcus pneumoniae was the most common organism isolated from children. Of 3053 FUO index presentations, 338 (11.1%, 95% confidence interval 10.0-12.2%) re-presented. Children were more likely to re-present than adults (13.5% of 1959 vs 6.8% of 1094; P < 0.001).Conclusions: Fever of unknown origin is diagnosed less frequently in adults than in children. Adult patients are more likely to be admitted, have longer lengths of stay and have positive blood cultures. Although FUO is diagnosed infrequently in the ED, blood cultures remain useful in the evaluation of unexplained fever, particularly in adults as age increases.
Original languageEnglish
Pages (from-to)105-112
JournalEmergency Medicine Australasia
Volume19
Issue number2
DOIs
Publication statusPublished - 2007

Fingerprint

Dive into the research topics of 'Outcomes in patients with an emergency department diagnosis of fever of unknown origin'. Together they form a unique fingerprint.

Cite this