Determining free-living daily energy expenditure and physical activity in chronic heart failure: questionnaire - doubly labelled water - motion sensors

Abstract Aims To evaluate free-living Total and Physical Activity Energy Expenditures (TEE/PAEE) and to assess the validity of the Daily Activity Questionnaire in Heart Failure (DAQIHF) in chronic heart failure (CHF) patients, against the doubly labelled water (DLW) and motion sensors methods. Methods 29 women/men (12/17) with CHF performed an incremental symptom-limited peak V̇O2 test. Free-living TEE and PAEE were estimated with the DAQIHF (TEEquest), motion sensor (Armband® TEEActi) and measured over 2 weeks using DLW (TEEDLW). Resting metabolic rate (RMR) and body composition were assessed with different methods, and peak V̇O2 with quality of life were correlated to TEE. Bland-Altman and Student T-Test analyses were used to compare methods. Statistical significance was set for P<0.05. Results Mean TEE did not significantly differ between TEEDLW and TEEquest (+352.4 kJ.24h−1; +5.3%; P=NS) for the whole group, nor between women or men, NYHA class, or cardiomyopathy: dilated cardiomyopathy (DCM)/ischemic cardiomyopathy (ICM). Bland-Altman plots revealed no systematic bias for TEE between methods. In a subgroup of women, TEEquest was significantly higher than TEEacti (P<0.05). RMR estimated from bioelectric impedance overestimated measured RMR (16.4%, P=0.0028). Patients spent 9.4% of their TEE in activities ≥3 metabolic equivalents (METs). Measured peak V̇O2 and estimated from the questionnaire were similar (14.1±4.7 vs. 14.8±3.4 ml.min−1.kg−1; P<0.0001) and were correlated to both TEEDLW and TEEquest (R=0.85 and 0.82 respectively, both P<0.0001). Conclusion Free-living TEE and peak V̇O2 can be estimated from the DAQIHF in patients with CHF across all activity domains allowing a complete description/assessment of daily physical activity intensities associated with powerful prognostic risk factors.