External validation of established risk adjustment models for procedural complications after percutaneous coronary intervention. Issue 8 (21st November 2007)

• Record Type: Journal Article
• Authors: Kunadian, B; Dunning, J; Das, R; Roberts, A P; Morley, R; Turley, A J; Twomey, D; Hall, J A; Wright, R A; Sutton, A G C; Muir, D F; de Belder, M A
• Is Part Of: Heart. Volume 94:Issue 8(2008)
• Page Start: 1012
• Page End: 1018
• Publication Date: 2007-11-21
• Subjects: Heart -- Diseases -- Treatment -- Periodicals; Cardiology -- Periodicals; 616.12
• DOI: 10.1136/hrt.2007.129197
• Languages: English
• ISSNs: 1355-6037

Abstract

Abstract : Background: Workable risk models for patients undergoing percutaneous coronary intervention (PCI) are needed urgently. Objective: To validate two proposed risk adjustment models (Mayo Clinic Risk Score (MC), USA and North West Quality Improvement Programme (NWQIP), UK models) for in-hospital PCI complications on an independent dataset of relatively high risk patients undergoing PCI. Setting: Tertiary centre in northern England. Methods: Between September 2002 and August 2006, 5034 consecutive PCI procedures (validation set) were performed on a patient group characterised by a high incidence of acute myocardial infarction (MI; 16.1%) and cardiogenic shock (1.7%). Two external models—the NWQIP model and the MC model—were externally validated. Main outcome measure: Major adverse cardiovascular and cerebrovascular events: in-hospital mortality, Q-wave MI, emergency coronary artery bypass grafting and cerebrovascular accidents. Results: An overall in-hospital complication rate of 2% was observed. Multivariate regression analysis identified risk factors for in-hospital complications that were similar to the risk factors identified by the two external models. When fitted to the dataset, both external models had an area under the receiver operating characteristic curve ⩾0.85 (c index (95% CI), NWQIP 0.86 (0.82 to 0.9); MC 0.87(0.84 to 0.9)), indicating overall excellent model discrimination and calibration (Hosmer–Lemeshow test, p>0.05). The NWQIP model was accurate in predicting in-hospital complications in different patient subgroups. Conclusions: Both models were externally validated. Both predictive models yield comparable results that provide excellent model discrimination and calibration when applied to patient groups in a different geographic population other than that in which the original model was developed.