Prospective Study Of A New Prediction Equation For VO2 Peak In Elderly Diastolic Heart Failure

Abstract

Recent studies have shown that diastolic dysfunction is the most common form of heart failure and is associated with severe exercise intolerance. However, due to the time and expense required, many clinics and rehabilitation centers are not able to assess metabolic gas exchange parameters including peak oxygen uptake (VO2 peak) and are forced to use a prediction equation based on the peak workload achieved by the patient. While improved from earlier editions, the ACSM (GETP 6TH edition) prediction equation still results in an error of approximately 10% when predicting VO2 peak in an elderly heart failure population. We previously examined the accuracy of the ACSM prediction equation on a cohort of elderly patients in the PARIS trial, and adapted the ACSM equation to better fit an elderly cohort. However, neither the ACSM nor the PARIS algorithm has been validated in an independent population of diastolic heart failure patients. PURPOSE To prospectively test and confirm the previously generated equation for predicting VO2 peak on a separate cohort of patients with diastolic heart failure. METHODS Symptom limited graded exercise tests were conducted on 67 consecutive elderly diastolic heart failure patients using a bicycle ergometer protocol that started at 12 watts and progressed to 25 watts after 3 minutes. The stages continued to increase by 25 Watts every three minutes until maximal effort was achieved. Breath-by-breath expired gas analysis was performed using a commercially available system with on-line computer calculations. Student t-tests and regression analysis were used to compare measured versus predicted VO2 peak values. The PARIS equation was VO2 peak =1.1 (ACSM)-1.4. RESULTS When applied to a separate cohort of 67 patients, the PARIS equation resulted in a significantly higher mean VO2 peak of 15.4 ml/kg/min when compared to the actual measured VO2 peak of 13.6 ml/kg/min (P <0.01). Regression analysis of the PARIS equation and the actual measured VO2 peak revealed a significant correlation (r=0.83) and SEE of 1.66. Likewise, the ACSM equation resulted in significantly greater mean VO2 peak of 15.3 ml/kg/min vs. actual measured VO2 peak of 13.6 ml/kg/ min (P <0.01). Regression analysis of the ACSM vs. measured values also revealed a significant correlation (r=0.81) and SEE 1.69. CONCLUSIONS These data demonstrate that VO2 peak can be predicted in elderly diastolic heart failure patients. However, both the ACSM and the PARIS equations systematically over-predicted VO2 peak in this important group of patients.