Gait Variability Among Breast Cancer Survivors During Forward And Backward Walking

Abstract

Increase in gait variability measured using instrumented techniques has been associated with increased risk of falls in persons with Parkinson’s disease, decreased functional performance and increased fall-risk in older adults. Recently research on breast cancer survivors (BCS) has focused on estimating fall-risk in this population attributed to gait performance. However, little is known about gait variability characteristics among BCS. PURPOSE: To assess the differences in gait variability among BCS compared to healthy controls during forward, backward, and accelerated forward walking. METHODS: 13 postmenopausal BCS (mean age: 58.5 ± 8.5 years) and 8 healthy controls (mean age: 60.8 ± 6.1 years) participated. Participants completed 5 trials each of forward, backward, and accelerated forward walking conditions on a 16x4’ Zeno walkway with a lead and follow-up distance of 1 m to capture steady-state gait. Coefficient of variation (CV) was calculated as % of standard deviation over mean of 5 trials. CV of stride length, stance time, and stride width were used as dependent variables. A Group (BCS vs healthy controls) X Condition (forward, backward, accelerated forward walking) ANOVA was performed. RESULTS: Significant interaction for stance time showed that BCS had greater CV during forward (4.89 ± 0.63%) and accelerated forward (6.49 ± 0.78%) but lesser CV during backward walking (6.28 ± 1.09%) compared to healthy controls (forward: 3.81 ± 0.80%; accelerated forward: 4.23 ± 1.00%; backward: 9.20 ± 1.38%; P = 0.018). Significant group main effect indicated that BCS (7.09 ± 3.59%) had greater stride length variability compared to healthy controls (3.59 ± 1.32%) across all conditions (P = 0.05). Significant condition main effect was observed for stride length and stance time CV (both P = 0.02) but not stride width (P = 0.063). During forward walking, stride length (by 5%) and stance time (by 3.3%) CV were significantly less compared to backward walking (both P = 0.004). There were no other significant differences. CONCLUSION: Increased stance time variability during backward walking and overall greater stride length variability may be indicative of increased fall-risk among BCS. Future studies need to examine other balance tests in conjunction with these measures to determine the level of fall-risk among BCS.