Background: Heart failure (HF) is associated with hemodynamic compensatory mechanisms. The Compensatory Reserve Index (CRI), originally developed for shock identification, incorporates physiological parameters acquired from pulse oximetry via machine learning (1=normal). Seattle Heart Failure Model (SHFM) is a prognostic instrument for heart failure patients. This study evaluates the prognostic value of CRI in HF patients compared with SHFM and actual one -year survival.
Methods: Ambulatory HF patients were recruited prospectively. CRI was measured in both recumbent and upright positions. Clinical data was collected from medical record and SHFM was calculated. The prognostic value of CRI was compared to SHFM using Pearson’s test and to actual survival using Mann-Whitney test.
Results: The study included 65 patients, age (mean±SD) 67±14 years, 25% females, 37% with reduced ejection fraction and 49% with ischemic heart disease. Recumbent CRI (mean±SD) was 0.90±0.09; upright 0.84±0.13 and delta-CRI (recumbent-upright) was 0.06±0.12. CRI was not associated with functional ability (NYHA class or 6-minute walking distance) or with blood pressure values. A weak association was found between the delta-CRI and hemoglobin (R=0.33, P=0.03), and between recumbent CRI and treatment with effective (≥50% target dose) beta blockers and aldosterone receptor antagonist. SHFM predicted 91±11% one year survival and actual one-year survival was 94% (four patients died due to heart failure). CRI was ineffective in predicting mortality but delta-CRI had a better predictive value than recumbent or upright measurements (R=0.25 compared with 0.1 and -0.16).
Conclusions: In this exploratory prospective outcome study CRI was normal in most heart failure patients. The index correlates with hemoglobin levels and heart failure treatment intensity. CRI in its current state was not found to predict mortality. Further research is needed to improve the index’s validity in this population.