Epidemiology of Endocarditis: A 22-Year Autopsy Review
Yaolin Zhou, Sean R Wilkinson, Matthew D Cain, Stephanie D Reilly. University of Alabama Birmingham, Birmingham, AL
Background: Infective endocarditis (IE) affects 10,000 to 20,000 persons in the United States every year. Epidemiologic factors include community and hospital bacterial flora and patient risk factors such as damaged native valves, prosthetic valves, intravenous drug abuse, and intracardiac catheters and devices. As the incidence of nosocomial bacteremia has increased, we hypothesized a correlative change in the epidemiology of IE at autopsy.
Design: Following IRB approval, a review of 6386 autopsy reports from 01/01/1990 to 09/30/2012 identified 69 cases of IE. We recorded patient demographics, clinical history, and culture results; supplemented missing data using original medical records; and calculated trends with linear regression models and hypothesis testing.
Results: The incidence of IE showed a modest but statistically significant increase over time (P=0.03; r2=0.16). Median age was 54; males were more affected than females (2.5:1); blacks and whites were equally affected. Endocarditis was more likely to be left-sided only (54:8). Seven patients had left- and right-sided involvement, and 18 patients had multiple valves involved. Despite hospitalization, 21 cases (30.4%) were not diagnosed until autopsy. Nearly half of patients had a recent invasive procedure (46.4%), including dialysis (11) and cardiac surgery (6). Risk factors included known infection (34.8%), prosthetic valves (21.7%), damaged or repaired native valves (18.8%), impaired immunity (17.4%), prior endocarditis (14.5%), and intravenous drug abuse (8.7%). Common presentations were fever (44.9%), neurologic symptoms (39.1%), dyspnea (37.7%), and pain (31.9%). Of cultured organisms, Staphylococcus aureus (SA) was most frequent (51.6%), with 21 cases being methicillin resistant (MRSA). The incidence of MRSA increased over time (P<0.01; r2=0.30). Other causative organisms were Enterococcus (15.6%); 2 of 10 were vancomycin resistant (VRE). There were equal numbers of coagulase negative Staphylococcus (CNS) and gram negative bacilli (6 each). Dialysis rates increased over time (P<0.01; r2=0.23). Dialysis patient cultures grew SA (17 with 9 MRSA), Enterococcus (5 with 1 VRE) and CNS (1). SA was the causative organism at a higher rate in dialysis patients than in non-dialysis patients (P<0.01). Black patients were more likely to be dialysis-dependent (P=0.02) and have SA endocarditis (P=0.02).
Conclusions: Our study showed increasing incidences of endocarditis, MRSA endocarditis, and endocarditis in chronic dialysis patients. Specifically, SA was most likely to be the causative organism in black patients as well as in dialysis-dependent patients.
Monday, March 4, 2013 9:30 AM
Poster Session I Stowell-Orbison/Surgical Pathology/Autopsy Awards Poster Session # 5, Monday Morning