Abstract 4007: Appropriate ICD Utilization in Heart Failure Patients: Can Risk Scores Help?
Background: Guidelines recommend ICD in heart failure (HF) patients (pts) be restricted to those with “reasonable expectation of survival with good functional status for more than 1 year”. While no standardized criteria exist to enable such assessments, studies suggest marked under-utilization of ICD in HF pts. Canadian HF registry data suggest that HF pts with Charlson Comorbidity Index (CCI) ≥3 have high mortality with or without ICD. Alternatively, in the MADIT-II trial, pts with severe renal dysfunction (SRD) or a MADIT Risk Score (MRS) ≥3 had high mortality and did not benefit from ICD.
Objective: To determine whether the CCI, MRS or SRD identify pts with high 0.5 to 1.5 year mortality in consecutive, unselected HF pts who are eligible for or already have ICD.
Methods: Consecutive pts with active HF and an EF<35% were identified (n=734) using natural language processing search of all in- and out-patient electronic records at a single center over 6 mo. Records were manually reviewed for ICD eligibility, CCI, MRS, SRD and survival. SRD was defined as glomerular filtration rate (GFR) ≤30 ml/min.
Results: Of 734 consecutive systolic HF pts (73% male; Median: Age=72 yrs, GFR=56 ml/min, CCI=4, MRS=3), 285 (39%) had existing ICD±CRT and 388 (86%) of the remaining 449 were eligible for ICD±CRT based on current guidelines. By receiver operating characteristic (ROC) analysis, CCI, MRS and GFR all predicted 6 mo and 1.5 yr mortality (p<0.05 for all; area under curve =0.65– 0.80). See table⇓.
Conclusion: Similar to previous studies, SRD identifies a very high risk group and high MRS or CCI identify high risk groups -regardless of ICD status. Over half of ICD eligible pts have risk scores suggesting they may not benefit from ICD. Prospective trials are needed to determine whether these retrospectively derived risk scores identify pts who truly do not benefit from ICD±CRT.