Frailty, comorbidity, quality of life help risk prediction in PCI patients
MedWire News: Frailty, comorbidity, and poor quality of life are common after percutaneous coronary intervention (PCI), and are associated with increased myocardial infarction (MI) and mortality risks, a study suggests.
Frailty, comorbidity, and quality of life variables must be included in future risk stratification models that predict long-term outcomes after PCI, say Mandeep Singh (Mayo Clinic, Rochester, Minnesota) and colleagues.
Although the widely used Mayo Clinic Risk Score (MCRS) accurately predicts in-hospital mortality and cardiovascular complications after PCI, it only has modest discrimination to predict long-term mortality, they explain.
Singh and team assessed further variables in 628 patients aged 65 years or over who had been discharged after PCI.
Multivariate analysis revealed that frailty (defined by three or more criteria from: weight loss >10lb, exhaustion, physical activity metabolic index <383 kcal/week, time taken to walk 15 feet ≥7 sec, and grip strength ≤21 kg), the physical component score of the Short-Form 36 (SF-36) questionnaire, and comorbidity (Charlston index) were all significantly associated with 1-year mortality, at respective hazard ratios (HR) of 2.74, 1.32 (per 10-point decrease), and 1.09.
When these variables were added to the MCRS, they produced "substantial gains" in prognostic information for long-term mortality, the authors report. Indeed, frailty increased the C-statistic of MCRS from 0.628 to 0.675, comorbidity to 0.671, and quality of life to 0.694.
Accordingly, the C-statistic of MCRS for the 3-year combined outcome of mortality and MI was increased, from 0.573, to 0.607, 0.576, and 0.587 by the inclusion of frailty, comorbidity, and quality of life, respectively.
In terms of the combined mortality and MI outcome, the only risk factor which added significant net reclassification improvement (NRI) to the existing MCRS model was frailty (NRI=16%; p=0.038).
Adding all three variables (frailty, comorbidity, SF-36) to the mortality model increased the C-index from 0.097 to 0.724, with an NRI of 43% (p=0.007).
Addition of all three to a mortality/MI model also resulted in modest improvement in the C-statistic and NRI, report the authors.
In a related commentary, editorialists Sarwat Chaudhry and Thomas Gill from Yale University in New Haven, Connecticut, USA, said: "To meet the needs of an aging population, a new clinical paradigm is needed, one that starts with a comprehensive assessment of risk for adverse clinical and patient-centered outcomes and then tailors therapy and surveillance to each patient's risk."
"Transitioning to this new paradigm is essential to ensuring that optimal care is provided to older patients with cardiovascular disease."
The results are published in the journal Circulation: Cardiovascular Quality and Outcomes.
MedWire (www.medwire-news.md) is an independent clinical news service provided by Current Medicine Group, a trading division of Springer Healthcare Limited. © Springer Healthcare Ltd; 2010
By Piriya Mahendra