OBJECTIVE: To develop a new predictive model of maximal heart rate based on two walking tests at different speeds (comfortable and brisk walking) as an alternative to a cardiopulmonary exercise test during cardiac rehabilitation.
DESIGN: Evaluation of a clinical assessment tool.
SETTING: A Cardiac Rehabilitation Department in France.
SUBJECTS: A total of 148 patients (133 men), mean age of 59 ±9 years, at the end of an outpatient cardiac rehabilitation programme.
MAIN MEASURES: Patients successively performed a 6-minute walk test, a 200 m fast-walk test (200mFWT), and a cardiopulmonary exercise test, with measure of heart rate at the end of each test. An all-possible regression procedure was used to determine the best predictive regression models of maximal heart rate. The best model was compared with the Fox equation in term of predictive error of maximal heart rate using the paired t-test.
RESULTS: Results of the two walking tests correlated significantly with maximal heart rate determined during the cardiopulmonary exercise test, whereas anthropometric parameters and resting heart rate did not. The simplified predictive model with the most acceptable mean error was: maximal heart rate = 130 - 0.6 × age + 0.3 × HR200mFWT (R(2) = 0.24). This model was superior to the Fox formula (R(2) = 0.138). The relationship between training target heart rate calculated from measured reserve heart rate and that established using this predictive model was statistically significant (r = 0.528, p < 10(-6)).
CONCLUSIONS: A formula combining heart rate measured during a safe simple fast walk test and age is more efficient than an equation only including age to predict maximal heart rate and training target heart rate.