The reliability and validity of the Åstrand nomogram and linear extrapolation for deriving VO_2max from submaximal exercise data

Background. While the accepted measure of aerobic power remains the V̇O_2max this test is extremely demanding even for athletes. There are serious practical and ethical concerns in attempting such testing in non-athletic or patient populations. An alternative method of measuring aerobic power in such populations is required. A limited body of work exists evaluating the accuracy of the Åstrand-Ryhming nomogram and linear extrapolation of the heart rate/oxygen uptake plot. Issues exist in terms of both equipment employed and sample numbers. Methods. Twenty-five normal subjects (mean age 28.6, range 22-50) completed 52 trials (Bruce treadmill protocol) meeting stringent criteria for V̇O_2max performance. Respiratory gases were measured with a portable gas analyser on a five-sec sample period. The data was analysed to allow comparison of the reliability and validity of linear extrapolations to three estimates of heart rate maximum with the Åstrand nomogram prediction. Results. Extrapolation was preferable yielding intraclass correlation co-efficients (ICC) of 0.9433 comparable to that of the observed V̇O_2max at 0.9443 and a bias of -1.1 ml · min^-1 · kg^-1 representing a 2.19% underestimate. Conclusions. This study provides empirical evidence that extrapolation of submaximal data can be employed with confidence for both clinical monitoring and research purposes. With the use of portable equipment and submaximal testing the scope for future research in numerous populations and non-laboratory environments is considerably increased.