Comparison of the long-term reproducibility of the walk test and of exercise peak oxygen consumption in patients with preserved exercise capacity

Abstract

Short-term and long-term reproducibility of the cardiopulmonary (CPX) exercise test have been established. Though short and mid-term reproducibility of the walk test has been ascertained, this was not extensively done for the long-term reproducibility. The aim of the study was to examine the long-term reproducibility of distance walked in an allotted time and to check the stability of the relationship between walked distance and exercise peak VO2 (pVO2).Methods: Forty six subjects (33 men; 57 ± 14 years), referred for functional capacity assessment, were studied twice by CPX and walking test. On the same day, CPX was performed on a bicycle or a treadmill and walk test in a corridor as required by specific guidelines. We performed a 12-minute walk test and the distance covered in six minutes was systematically taken down. A free time interval of 1.5 hours was observed between the exercise tests. Distance walked in the allotted time and pVO2 were analysed. Reproducibility was assessed according to Bland and Altman plots and intra-class coefficient correlation (ICC). The relationship between distance ambulated and pVO2 was analysed by the Spearman coefficient correlation.Results: The time interval between the two evaluations was 290 ± 10 days. During this meantime, for those subjects having drug treatment, no change was recorded in their regimen. BMI remained stable for the entire studied population (28 ± 5 kg/m2). Minute walked distance was respectively 522 ± 83 and 527 ± 76 m in six minutes, 1033 ± 182 and 1041 ± 153 m in 12 minutes. pVO2 was 21 ± 7 and 22 ± 7 ml/kg/min (all p = NS). The walk test was reproducible in the long-term, regardless of the modality (6 or 12-minute walk) as shown by the Bland–Altman plots and the high ICC of .89. Spearman’s rho coefficient between distance ambulated and pVO2 was modest and remained stable over time whatever the allotted time: Spearman’s r = .54; p = .0011 (1st evaluation) and Spearman’s r = .51; p = .0019 (2nd evaluation) between 6-minute distance walked and pVO2.Conclusions: The walking distance in an allotted time seems highly reproducible in the long-term. Its relationship with pVO2 remains stable over time. It could be of value for repeated assessment of patients’ exercise capacity in a first step. Further evaluation in a larger population is needed to confirm our result and its usefulness in clinical practice.