Comprehensive Analysis of Cardiopulmonary Exercise Testing and Mortality in Patients With Systolic Heart Failure: The Henry Ford Hospital Cardiopulmonary Exercise Testing (FIT-CPX) Project

• Published in: Journal of cardiac failure 2015-09, Vol.21 (9), p.710-718
• Main Authors: Brawner, Clinton A., PhD, Shafiq, Ali, MD, Aldred, Heather A., PhD, Ehrman, Jonathan K., PhD, Leifer, Eric S., PhD, Selektor, Yelena, MD, Tita, Cristina, MD, Velez, Mauricio, MD, Williams, Celeste T., MD, Schairer, John R., DO, Lanfear, David E., MD, Keteyian, Steven J., PhD
• Format: Article
• Language: English
• Subjects: Cardiomyopathy; Cardiovascular; Cause of Death - trends; Exercise - psychology; exercise test; Exercise Test - methods; Female; Follow-Up Studies; Heart Failure, Systolic - diagnosis; Heart Failure, Systolic - mortality; Heart Failure, Systolic - physiopathology; Humans; Male; Middle Aged; Oxygen Consumption; Prognosis; Proportional Hazards Models; Retrospective Studies; Severity of Illness Index; Stroke Volume; survival; Survival Rate - trends; United States - epidemiology
• ISSN: 1071-9164; 1532-8414
• DOI: 10.1016/j.cardfail.2015.06.001

Abstract Background Many studies have shown a strong association between numerous variables from a cardiopulmonary exercise (CPX) test and prognosis in patients with heart failure with reduced ejection fraction (HFrEF). However, few studies have compared the prognostic value of a majority of these variables simultaneously, so controversy remains regarding optimal interpretation. Methods and Results This was a retrospective analysis of patients with HFrEF (n = 1,201; age = 55 ± 13 y; 33% female) and a CPX test from 1997 to 2010. Thirty variables from a CPX test were considered in separate adjusted Cox regression analyses to describe the strength of the relation of each to a composite end point of all-cause mortality, left ventricular assist device implantation, or heart transplantation. During a median follow-up of 3.8 years, there were 577 (48.0%) events. The majority of variables were highly significant ( P < .001). Among these, percentage of predicted maximum V ˙ O2 (ppM V ˙ O2 ; Wald = 203; P < .001; C-index = 0.73) was similar to VE -VCO2 slope (Wald = 201; P < .001; C = 0.72) and peak V ˙ O2 (Wald = 161; P < .001; C = 0.72). In addition, there was no significant interaction observed for peak respiratory exchange ratio