A new maximal bicycle test using a prediction algorithm developed from four large COPD studies

Abstract

Background: Maximum exercise workload (WMAX) is today assessed as the first part of Cardiopulmonary Exercise testing. The WMAX test exposes patients with COPD, often having cardiovascular comorbidity, to risks. Our research project was initiated with the final aim to eliminate the WMAX test and replace this test with a predicted value of WMAX, based on a prediction algorithm of WMAX derived from multicentre studies. Methods: Baseline data (WMAX, demography, lung function parameters) from 850 COPD patients from four multicentre studies were collected and standardized. A prediction algorithm was prepared using Random Forest modelling. Predicted values of WMAX were used in a new WMAX test, which used a linear increase in order to reach the predicted WMAX within 8 min. The new WMAX test was compared with the standard stepwise WMAX test in a pilot study including 15 patients with mild/moderate COPD. Results: The best prediction algorithm of WMAX included age, sex, height, weight, and six lung function parameters. FEV1 and DLCO were the most important predictors. The new WMAX test had a better correlation (R2 = 0.84) between predicted and measured WMAX than the standard WMAX test (R2 = 0.66), with slopes of 0.50 and 0.46, respectively. The results from the new WMAX test and the standard WMAX test correlated well. Conclusion: A prediction algorithm based on data from four large multicentre studies was used in a new WMAX test. The prediction algorithm provided reliable values of predicted WMAX. In comparison with the standard WMAX test, the new WMAX test provided similar overall results.

Keywords

• copd
• cardiopulmonary exercise testing
• wmax
• random forest
• prediction