Validity of Predicting VO2max Using Linear Regression of HR and VO2 From a Multi-stage, Submaximal Treadmill Test

Abstract

0603 PURPOSE: The purpose of this study was to determine the validity of predicting VO2max using linear regression of heart rate and oxygen consumption obtained during a submaximal treadmill test with multiple stages. METHODS: A total of 40 (20 males, 20 females) Low Risk (ACSM GETP 6th Ed.) subjects (28.9 ± 7.1 years, Height = 170.2 ± 11.3 cm, Weight = 68.4 ± 12.5 kg, Body Fat = 16.6 ± 5.3%, BMI = 23.5 ± 2.6 kg/m2) completed 2 maximal and 1 submaximal treadmill tests. The treadmill test had a 4 minute warmup stage followed by 2 minute stages that increased in workload by approximately 1.5 METs. Oxygen consumption was measured during each test, and measured VO2max was determined using standard criteria from the 2 max tests (CV = 3.8%). Predicted VO2max was determined by linear regression of heart rate from ECG and VO2 using age-predicted maximum heart rate. VO2max was predicted for each individual using all HR and VO2 pairs and by eliminating heart rates below 100, 110, and 120 bpm, respectively. The accuracy of predicting VO2max using measured HRmax was also determined. The relationship of predicted VO2max to measured VO2max was determined with Pearson's Product Moment Correlation. RESULTS: Mean measured VO2max was 2.785 ± 0.680 L/min, and the most accurate prediction of VO2maxoccurred when heart rates below 110 bpm were excluded from the linear regression and measured HRmax was used (Mean VO2max = 2.681 ± 0.691 L/min, R = 0.931, SEE = 0.252 L/min). Use of age-predicted HRmax with heart rates below 110 bpm excluded also resulted in highly accurate predictions of VO2max (Mean VO2max = 2.767 ± 0.744 L/min, R = 0.922, SEE = 0.266 L/min). CONCLUSIONS: The results of this study demonstrate that a very strong positive correlation exists between measured VO2max and VO2max predicted using linear regression of HR and VO2 and age-predicted HR, and support the use of a submaximal treadmill test and linear regression to accurately predict maximal oxygen consumption in low risk adults.