Infections in patients hospitalized for fever as related to duration and other predictors at admittance.
Infection. 2014 Jun;42(3):485-92
Authors: Naess A, Mo R, Nilssen SS, Eide GE, Sjursen H
PURPOSE: To find the underlying diagnoses of patients admitted to a medical department with symptoms of fever to ascertain whether the duration of fever, temperature readings, and common laboratory tests could give a clue to the final diagnosis.
METHODS: A retrospective study of the records of 299 patients admitted to the Medical Department, Haukeland University Hospital from July 1st, 2001 until June 30th, 2004 for fever without any diagnosis suggesting the cause of the fever.
RESULTS AND CONCLUSIONS: Infections were the most common diagnoses, but the proportion of patients with infections declined with increasing duration of fever before admission (63.7 % when <7 days, 45.8 % when >21 days). A group of patients on immunosuppressive therapy were all hospitalized within 1 week from the debut of fever, and their causes of fever were comparable to those of non-immunocompromised patients with fever of the same duration. With fever <7 days, patients with bacterial or viral infection had higher maximum temperatures (medians 39.2 and 38.9 °C, respectively) than those without infection (median 38.0 °C). Patients with bacterial infection had higher C-reactive protein (CRP) levels than patients in the other groups, whereas viral infection was associated with higher lymphocyte counts. The neutrophil:lymphocyte ratio was higher in patients with fever due to bacterial infections than in those with viral infections. For patients with fever for <1 week, neutrophil [area under the curve (AUC) 0.723], white blood cell (WBC, AUC 0.692), and monocyte (AUC 0.691) counts and CRP levels (AUC 0.684) were the best single indicators of bacterial infection.
PMID: 24363209 [PubMed - indexed for MEDLINE]