Frailty measured using an electronic frailty index and outcomes in older patients with myocardial infarction

Abstract

Abstract Background Frailty is common amongst older patients with myocardial infarction and is an independent predictor of poor clinical outcomes. Despite this, frailty is not systematically or objectively measured nor routinely used in risk stratification. The electronic Frailty Index (eFI) is a frailty assessment tool which uses electronic health records to identify and classify frailty and has been shown to correlate well with in-person frailty assessment. However, the performance of the eFI in disease specific populations and its ability to predict individual patient outcomes is unclear. Purpose To assess the relationship between frailty, measured using the eFI, and the management and outcomes of older patients with myocardial infarction. Methods Consecutive patients admitted to hospital in the South East of Scotland between 01/10/2014 and the 01/03/2021 with a primary diagnosis of myocardial infarction were identified using International Classification of Disease (ICD-10) code I21 or I22. Patients under the age of 65 (n= 3,601) and those with no biomarker evidence of myocardial injury were excluded (n=1,301). The provision of guideline recommended therapy, recurrent hospital admission due to myocardial infarction, heart failure or stroke and 12-month all-cause mortality were compared between groups based on eFI classification (fit, mild, moderate or severely frail). Results Of the 3,930 patients (mean age 78 [SD 8] years, 43% female, 83% Caucasian), 2,324 (59%) were classified as frail with 1,421 (36%), 664 (17%), and 239 (6%) classified as mild, moderate and severely frail, respectively. The eFI was measurable in all patients. Patients with any degree of frailty were less likely to receive anti-platelet therapy, ACE inhibitor or angiotensin receptor blocker or beta-blocker therapy or undergo revascularisation with fewer than 1 in 10 patients with severe frailty undergoing inpatient coronary catheterisation (7.5% vs 57% non-frail, P <0.001 for all). Frailty was independently associated with an increased incidence of recurrent myocardial infarction, heart failure or stroke within 12 months with the odds greatest in those with severe frailty (OR 1.93 (95% CI 1.42-2.61%)). The primary outcome of all-cause death at 12-months occurred in 355 (25%), 226 (34%) and 109 (46%) patients with mild, moderate and severe frailty compared with 176 (11%) in non-frail patients (Figure 1). Frailty status was an independent risk factor for all-cause mortality with the risk of death greatest in those with severe frailty (HR 2.64 (95% CI 1.63 – 3.67, p<0.001)). Conclusion The eFI identified patients at increased risk of in-patient death, major adverse cardiovascular events, and all-cause mortality within 12 months following myocardial infarction. Measurement of the eFI from linked healthcare data is feasible and could be used to aide personalised risk stratification and service utilisation in older patients admitted with myocardial infarction.Cumulative incidence plotForest plot with adjusted hazard ratio