Elevated serum cardiac troponin I in older patients with hip fracture: incidence and prognostic significance.

摘要

INTRODUCTION: Cardiovascular complications are the main causes of morbidity and mortality in patients with osteoporotic hip fracture (HF). The aim of this prospective study was to evaluate the incidence and prognostic significance of elevated cardiac troponin I (cTnI) in the early peri-operative period in older patients with HF. MATERIALS AND METHODS: A blind evaluation of myocardial injury as detected by cTnI elevation in 238 consecutive older patients with low-trauma HF (mean age 81.9 +/- 7.8 (SD) years; 72% females). Data on demographic and clinical characteristics, in-hospital mortality, hospital length of stay and discharge destination were collected prospectively. Serum cTnI level was analysed from blood collected routinely in the first 72 h of hospital admission. RESULTS: Sixty-nine (29%) patients had elevated cTnI (>0.06 microg/l) but myocardial injury was clinically recognised in only 23 (33%) and only 24 (34.8%) had a history of coronary artery disease (CAD). Patients with elevated cTnI were significantly older, more often had American Society of Anaesthesiologist status score >or=3, a history of CAD or stroke and more often were current smokers than the patients without cTnI elevation. In multivariate regression analysis only age was an independent predictor of cTnI elevation. Patients with cTnI release were twice as likely to have a length of stay >or=20 days (P = 0.047) and 2.7 times more likely to be discharged to a long-term residential care facility (RCF) (P = 0.013). cTnI level >or=1 microg/l was a strong independent predictor of all-cause mortality with 98.3% specificity and 89.1% negative predictive value. CONCLUSION: Peri-operative myocardial injury is common in older HF patients but is frequently unrecognised clinically. Elevated blood cTnI level is an independent predictor of prolonged length of hospital stay (>or=20 days), need for long-term RCF and mortality (if cTnI >or=1 microg/l).