Selection bias of elderly patients with chronic angina referred for catheterization C. Kaiser a , R. Jeger a , S. Wyrsch a , L. Schoeb a , G.M. Kuster a , P. Buser a , S. Osswald a , F. Bernet b , W. Brett b , L. Grize c , M. Pfisterer a, * for the TIME Investigators a Department of Cardiology, University Hospital, CH-4031 Basel, Switzerland b Department of Cardiothoracic Surgery, University Hospital, Basel, Switzerland c Institute of Social and Preventive Medicine, University of Basel, Basel, Switzerland Received 22 April 2005; received in revised form 28 June 2005; accepted 24 July 2005 Available online 12 October 2005 Abstract Background: Registry patients are generally older and more sick than patients enrolled in trials questioning the generalizability of trial results. We assessed whether such a selection bias also exists in elderly patients with chronic angina referred for catheterization. Methods and results: All 119 patients age  75 years with Trial of Invasive versus Medical Therapy in the Elderly (TIME) inclusion but no major exclusion criteria referred for catheterization during the TIME trial inclusion period in four TIME centers were registered and followed-up for one year. Registry patients differed from the 188 trial patients in the same hospitals in that they were younger, somewhat more frequently male, with less antianginal drugs and studied more often after acute chest pain at rest but with more comorbidities than study patients. Left ventricular ejection fraction and vessel disease were similar. One year mortality was 11.4% in registry and 9.6% in invasive TIME patients but differences disappeared after adjustment for baseline differences. Symptomatic status after one year was similar too. Conclusions: In elderly patients with chronic angina, a bias in the selection for invasive management exists which seems different from that reported in younger patient settings. After adjustment for these selection factors, however, one-year outcome was remarkably similar in registry and trial patients. D 2005 Elsevier Ireland Ltd. All rights reserved. Keywords: Chronic angina; Catheterization; Selection bias; Trial of Invasive versus Medical Therapy in the Elderly (TIME) 1. Introduction Patients enrolled in randomized clinical trials are gen- erally younger and healthier than those referred to the same institutions but not included or put into registries [1]. Older patients, females, patients with comorbidities and patients at higher risk for future events are often excluded, limiting extrapolation of outcome data to general patient popula- tions. This has been noted and discussed for studies with acute coronary syndromes [2–4], cardiogenic shock [5], revascularization [6–9] as well as with implantable cardi- overter defibrillators [10] and has led to the finding that elderly patients and females are severely under-represented in randomized clinical trials [11]. Thus, revascularization data on elderly patients stems only from registries [12,13] and not from prospective randomized trials. Since such a selection bias is observed particularly in elderly patients with a high rate of comorbidities, we compared patients referred for catheterization similar to those collected in registries with patients included in the TIME trial for evaluation of coronary artery disease (CAD) based on their clinical presentation [14,15]. TIME was the first prospective randomized multicenter trial specifically addressing the management of patients age 75 years or older 0167-5273/$ - see front matter D 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ijcard.2005.07.024 * Corresponding author. Tel.: +41 61 265 52 14; fax: + 41 61 265 45 98. E-mail address: pfisterer@email.ch (M. Pfisterer). International Journal of Cardiology 110 (2006) 80 – 85 www.elsevier.com/locate/ijcard