Age-specific cutoff values for change in gait speed to predict improvement in 6-minute walk distance in patients with cardiovascular disease
K Sato, K Kamiya, N Hamazaki, K Nozaki, K Ogura, T Miki, S Soeda, K Iida, K Hotta, E Maekawa, M Yamaoka-Tojo, A Matsunaga, J AkoAbstract
Background/Introduction
Changes in 6-minute walk distance (6MWD) are established prognostic indicators in patients with cardiovascular disease (CVD). However, repeated assessment of 6MWD may be challenging in daily clinical practice. Gait speed (GS) may serve as a simple surrogate marker for 6MWD, but the magnitude of change in GS that predicts a clinically meaningful improvement in 6MWD remains unclear, particularly across different age groups.
Purpose
The purpose of this study was to determine the cutoff values for changes in GS that predict improvement in 6MWD in patients with CVD.
Methods
We studied 1,872 patients with CVD who were assessed for 6MWD and GS at discharge and 5 months post-discharge, and for whom longitudinal changes were available. 6MWD improvement was defined as an increase of ≥ 30 m from baseline. Using receiver operating characteristic curve analysis and the Youden index, we calculated cutoff values for changes in GS to predict 6MWD improvement for the total cohort and age-stratified groups (non-older: < 65 years; older: ≥ 65 years). The association between 6MWD improvement and improvement in GS meeting or exceeding the age-specific cutoff values was examined using multivariate logistic regression analysis. Furthermore, after defining patients who showed an improvement equal to or greater than the age-specific GS cutoff values as the improvement group, we examined the association between the age-specific GS cutoffs and all-cause mortality using a landmark analysis. In this analysis, patients who died within 150 days after discharge or became untraceable were excluded.
Results
The cutoff values for changes in GS to predict 6MWD improvement were 0.14 m/s for the total cohort [area under the curve (AUC): 0.68, sensitivity: 0.52, specificity: 0.76], 0.14 m/s for the non-older group (AUC: 0.64, sensitivity: 0.50, specificity: 0.72), and 0.08 m/s for the older group (AUC: 0.70, sensitivity: 0.66, specificity: 0.65). In the multivariate logistic regression analysis, an improvement in GS equal to or greater than the age-specific cutoff values was significantly associated with 6MWD improvement. In the landmark analysis, the GS improvement group defined by age-specific cutoff values was associated with a significantly lower all-cause mortality rate compared with the non-improvement group.
Conclusion
In patients with CVD, the cutoff values for changes in GS to predict 6MWD improvement were 0.14 m/s for the total cohort and non-older patients, and 0.08 m/s for older patients. Furthermore, the cutoff values identified in this study were associated with all-cause mortality. Changes in GS may serve as a practical surrogate marker for longitudinal assessment in patients with CVD, suggesting that GS improvement goals may need to be considered according to age.