April 28, 1997

Fever, C-Reactive Protein, and Other Acute-Phase Reactants During Treatment of Infective Endocarditis

Author Affiliations

From the Department of Infectious Diseases, Göteborg University, Göteborg, Sweden.

Arch Intern Med. 1997;157(8):885-892. doi:10.1001/archinte.1997.00440290069007

Background:  Fever and sustained elevations of levels of C-reactive protein, erythrocyte sedimentation rate, and other inflammatory markers are common problems during treatment of infective endocarditis. We studied the value of these measurements during an 8-year period in all episodes of infective endocarditis treated in 1 university-affiliated institution.

Methods:  A total of 193 consecutive episodes that fulfilled the criteria for infective endocarditis were prospectively enrolled during 2 periods, 1984 through 1988 and 1993 through 1995. Fever and results of serial measurements of C-reactive protein, erythrocyte sedimentation rate, white blood cell counts, and platelet counts were related to the clinical course of infective endocarditis.

Results:  Fever persisted or recurred in 108 episodes (57%) despite appropriate antibiotic treatment. The causes of persistent fever and recurrent fever were different. Persistent fever that lasted 7 days or longer was caused by acomplicating cardiac infection in 56% of these episodes. Recurrent fever, noted in 31% of all episodes and the major cause of fever during the third and fourth treatment weeks, was caused most often by hypersensitivity reactions to β-lactams. Elevations in C-reactive protein levels were significantly prolonged in the episodes with complicated courses compared with the episodes with uncomplicated courses, while mean erythrocyte sedimentation rate remained unchanged during treatment, not differentiating between complicated and uncomplicated episodes.

Conclusions:  Fever during treatment must be analyzed in terms of persistence and recurrence to provide a basis for clinical decisions. Serial measurements of Creactive protein are useful to monitor the reponse to antimicrobial therapy and to detect complications, while serial determinations of erythrocyte sedimentation rate are of no value.Arch Intern Med. 1997;157:885-892