Clinical features, microbiology and surgical outcomes of infective endocarditis: a 13-year study from a UK tertiary cardiothoracic referral centre

QJM. 2015 Mar;108(3):219-29. doi: 10.1093/qjmed/hcu188. Epub 2014 Sep 15.

Abstract

Background: Infective endocarditis (IE) causes substantial morbidity and mortality. Patient and pathogen profiles, as well as microbiological and operative strategies, continue to evolve. The impact of these changes requires evaluation to inform optimum management and identify individuals at high risk of early mortality.

Aim: Identification of clinical and microbiological features, and surgical outcomes, among patients presenting to a UK tertiary cardiothoracic centre for surgical management of IE between 1998 and 2010.

Design: Retrospective observational cohort study.

Methods: Clinical, biochemical, microbiological and echocardiographic data were identified from clinical records. Principal outcomes were all-cause 28-day mortality and duration of post-operative admission.

Results: Patients (n = 336) were predominantly male (75.0%); median age 52 years (IQR = 41-67). Most cases involved the aortic (56.0%) or mitral (53.9%) valves. Microbiological diagnoses, obtained in 288 (85.7%) patients, included streptococci (45.2%); staphylococci (34.5%); Haemophilus, Actinobacillus, Cardiobacterium, Eikenella, Kingella (HACEK) organisms (3.0%); and fungi (1.8%); 11.3% had polymicrobial infection. Valve replacement in 308 (91.7%) patients included mechanical prostheses (69.8%), xenografts (24.0%) and homografts (6.2%). Early mortality was 12.2%, but fell progressively during the study (P = 0.02), as did median duration of post-operative admission (33.5 to 10.5 days; P = 0.0003). Multivariable analysis showed previous cardiothoracic surgery (OR = 3.85, P = 0.03), neutrophil count (OR = 2.27, P = 0.05), albumin (OR = 0.94, P = 0.04) and urea (OR = 2.63, P < 0.001) predicted early mortality.

Conclusions: This study demonstrates reduced post-operative early mortality and duration of hospital admission for IE patients over the past 13 years. Biomarkers (previous cardiothoracic surgery, neutrophil count, albumin and urea), predictive of early post-operative mortality, require prospective evaluation to refine algorithms, further improve outcomes and reduce healthcare costs associated with IE.

Publication types

  • Observational Study
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adult
  • Aged
  • Echocardiography
  • Endocarditis, Bacterial / microbiology
  • Endocarditis, Bacterial / mortality
  • Endocarditis, Bacterial / surgery*
  • Female
  • Gram-Negative Bacterial Infections / microbiology
  • Gram-Negative Bacterial Infections / mortality
  • Gram-Negative Bacterial Infections / surgery*
  • Gram-Positive Bacterial Infections / microbiology
  • Gram-Positive Bacterial Infections / mortality
  • Gram-Positive Bacterial Infections / surgery*
  • Heart Valve Diseases / microbiology
  • Heart Valve Diseases / mortality
  • Heart Valve Diseases / surgery*
  • Heart Valve Prosthesis Implantation / statistics & numerical data
  • Humans
  • Length of Stay / statistics & numerical data
  • Male
  • Middle Aged
  • Mycoses / microbiology
  • Mycoses / mortality
  • Mycoses / surgery*
  • Patient Readmission / statistics & numerical data
  • Retrospective Studies
  • Risk Factors
  • Substance Abuse, Intravenous / complications
  • Treatment Outcome