C Chuard1, C M Antley, L B Reller. 1. Clinical Microbiology Laboratory, Duke University Medical Center, Durham, NC 27710, USA.
Abstract
OBJECTIVES: To determine if routine cardiac valve culture is useful in diagnosing clinically unsuspected infective endocarditis in patients undergoing native valve replacement, to see if false-positive culture results have a deleterious effect on patient care, and to determine if microbiology and histopathology can be used to differentiate partially treated and untreated infective endocarditis from valve contamination. DESIGN: Case series. SETTING: Tertiary-care teaching hospital with 1125 beds. PATIENTS: Forty-eight patients with culture-positive cardiac valves after native valve replacement. RESULTS: A single unsuspected case of endocarditis was disclosed by microbiology over a 5-year period. Histopathology in this case was also positive, however, and the diagnosis should have been suspected clinically. Eighty-three percent of positive cultures were the result of contamination over an 18-month period; results were disregarded appropriately by clinicians. Clinical context or histopathology was required to categorize microorganisms correctly as pathogens or contaminants; only the presence of organisms on Gram stain had a good predictive value for endocarditis. CONCLUSION: Routine valve cultures in patients undergoing native valve replacement are not warranted. Although false-positive culture results had no deleterious effects on patient care in this study, misinterpretation of such results could lead to overtreatment. Microbiology results alone are not sufficient to distinguish endocarditis from contamination.
OBJECTIVES: To determine if routine cardiac valve culture is useful in diagnosing clinically unsuspected infective endocarditis in patients undergoing native valve replacement, to see if false-positive culture results have a deleterious effect on patient care, and to determine if microbiology and histopathology can be used to differentiate partially treated and untreated infective endocarditis from valve contamination. DESIGN: Case series. SETTING: Tertiary-care teaching hospital with 1125 beds. PATIENTS: Forty-eight patients with culture-positive cardiac valves after native valve replacement. RESULTS: A single unsuspected case of endocarditis was disclosed by microbiology over a 5-year period. Histopathology in this case was also positive, however, and the diagnosis should have been suspected clinically. Eighty-three percent of positive cultures were the result of contamination over an 18-month period; results were disregarded appropriately by clinicians. Clinical context or histopathology was required to categorize microorganisms correctly as pathogens or contaminants; only the presence of organisms on Gram stain had a good predictive value for endocarditis. CONCLUSION: Routine valve cultures in patients undergoing native valve replacement are not warranted. Although false-positive culture results had no deleterious effects on patient care in this study, misinterpretation of such results could lead to overtreatment. Microbiology results alone are not sufficient to distinguish endocarditis from contamination.