Estimation of peak work load based on 6-min walk distance and general demographics in patients with COPD: A new regression equation

Abstract

Background: Existing regression equations to estimate Wpeak using 6-min walk distance (6MWD) are not accurate enough to target training intensity during pulmonary rehabilitation (PR) in patients with COPD (Sillen et al ERS 2011). We aimed to develop a new regression equation to estimate Wpeak using 6MWD and general demographics in COPD patients entering PR.

Methods: Measurements of lung function, body composition, and peak and functional exercise capacity were obtained in 3000 patients with COPD (53% men, age: 63.0±9.4 yrs; FEV₁: 44±18% pred.), referred to 4 PR centres in the Netherlands (all member of the Vereniging Astmacentra Nederland). A stepwise multiple regression analysis was performed to estimate Wpeak using 6MWD, gender, age, height, body weight, fat free mass and FEV₁.

Results: On average, patients had a normal body composition (BMI: 25.8±5.5 kg/m²; FFMI: 16.9±18.4 kg/m²), and a poor peak (59±33 Watts) and functional exercise capacity (6MWD: 399±120 metres). The VAN regression equation derived from the stepwise multiple regression analysis was as follows:

Wpeak = -52.787+(0.319*gender, women=0 and men=1)-(0.229*age in yrs)+(0.108*height in cm) + (0.267*body weight in kg) + (0.182*FFM in kg)+(0.132*6MWD in m)+(23.528* FEV₁ in litres).

This regression equation explained 67% of the variation. The mean difference between the actual and the predicted Wpeak was 0.3±19 watts. 26% of the COPD patients had a predicted Wpeak which differed less than 5 watts (±) of the actual Wpeak.

Conclusion: The level of accuracy of this newly derived regression equation seems too low to be used in indivuals with COPD to target training intensity during PR.