Apple-CPET Ted Rogers Understanding Exacerbations of Heart Failure

Heart Failure (HF) has a prevalence of 3.5% suggesting that over one million Canadians are affected by this disease, and more than 50,000 are newly diagnosed each year. This complex disease is associated with the highest burden of cost to the healthcare system, attributable to hospitalizations, missed work, medications, and health care services

Traditionally, clinicians have relied on static snapshots of patients to determine clinical status and estimate prognosis. More advanced cardiac centers rely on cardiopulmonary exercise testing (CPET), where patients are further stratified based on validated exercise parameters. CPET remains underutilized and resource-intensive. It requires expensive equipment, proficient personnel, and clinicians with specialized training. Thus, there is an unmet need for a more widely available, accessible, and longitudinal assessment of clinical status to better monitor and prognosticate patients outside of the ambulatory setting. Wearable devices such as the Apple Watch hold great promise in this regard, as they provide near-continuous monitoring of biometric data. By combining biometric data with demographic, cardiac, and biomarker testing, the investigators will significantly improve our ability to predict heart failure outcomes such as early warning of decompensation, clinical deterioration (symptoms and brain natriuretic peptide (BNP) as a surrogate), hospitalization, mortality (using the Seattle Heart Failure Model (SHFM) as a surrogate), and/or need for advanced heart failure therapies.

Our study has 5 research questions based on 2 primary outcomes and 3 secondary outcomes in clinically diverse adult ambulatory heart failure patients :

Primary Research Question:

1. Can biometric data obtained from the Apple Watch be used to estimate cardiorespiratory fitness, as assessed by CPET?
2. Does the 'predicted' Apple 6 MW estimate correlate with formal 6 MWT? Secondary Research Questions:
3. Is there a relationship between novel biosensors, including oxygen saturation, and markers of poor prognosis specifically as defined by the SHFM, BNP, Quality of life (QOL) indicators, and CPET parameters?
4. Can surrogates of cardiorespiratory fitness obtained from the Apple Watch, including novel biosensors, predict acute decompensation of heart failure as defined rapid clinic visits, need for IV diuretics, ED visits, heart failure hospitalization and unscheduled health care encounters during the 3-month follow-up?
5. Can biometric data be used to improve a risk prediction model that can distinguish between patients at high versus low risk of all-cause hospitalization (primary outcome), all-cause mortality (secondary outcome), and a composite outcome of all-cause mortality, need for ventricular assist device, or heart transplantation (secondary composite outcome) over a 2 year period?