Peripheral artery disease (PAD) manifests from atherosclerosis, which limits blood flow to the legs and causes changes in muscle structure and function, and in gait performance. PAD is underdiagnosed, which delays treatment and worsens clinical outcomes. To overcome this challenge, the purpose of this study is to develop machine learning (ML) models that distinguish individuals with and without PAD. This is the first step to using ML to identify those with PAD risk early. We built ML models based on previously acquired overground walking biomechanics data from patients with PAD and healthy controls. Gait signatures were characterized using ankle, knee, and hip joint angles, torques, and powers, as well as ground reaction forces (GRF). ML was able to classify those with and without PAD using Neural Networks or Random Forest algorithms with 89% accuracy (0.64 Matthew’s Correlation Coefficient) using all laboratory-based gait variables. Moreover, models using only GRF variables provided up to 87% accuracy (0.64 Matthew’s Correlation Coefficient). These results indicate that ML models can classify those with and without PAD using gait signatures with acceptable performance. Results also show that an ML gait signature model that uses GRF features delivers the most informative data for PAD classification.

Background: Patients with peripheral artery disease with intermittent claudication (PAD-IC) have altered gait variability from the first step they take, well before the onset of claudication pain. The mechanisms underlying these gait alterations are poorly understood.Aims: To determine the effect of reduced blood flow on gait variability by comparing healthy older controls and patients with PAD-IC. We also determined the diagnostic value of gait variability parameters to identify the presence of PAD.Methods: A cross-sectional cohort design was used. Thirty healthy older controls and thirty patients with PAD-IC walked on a treadmill at their self-selected speed in pain free walking (normal walking for healthy older controls; prior to claudication onset for PAD) and reduced blood flow (post vascular occlusion with thigh tourniquet for healthy older controls; pain for PAD) conditions. Gait variability was assessed using the largest Lyapunov exponent, approximate entropy, standard deviation, and coefficient of variation of ankle, knee, and hip joints range of motion. Receiver operating characteristics curve analyses of the pain free walking condition were performed to determine the optimal cut-off values for separating individuals with PAD-IC from those without PAD-IC.Results and discussion: Patients with PAD-IC have increased amount of variability for knee and hip ranges of motion compared with the healthy older control group. Regarding the main effect of condition, reduced blood flow demonstrated increased amount of variability compared with pain free walking. Significant interactions between group and condition at the ankle show increased values for temporal structure of variability, but a similar amount of variability in the reduced blood flow condition. This demonstrates subtle interactions in the movement patterns remain distinct between PAD-IC versus healthy older controls during the reduced blood flow condition. A combination of gait variability parameters correctly identifies PAD-IC disease 70% of the time or more.Conclusions: Gait variability is affected both by PAD and by the mechanical induction of reduced blood flow. Gait variability parameters have potential diagnostic ability, as some measures had 90.0% probability of correctly identifying patients with PAD-IC.

The most common symptom of peripheral artery disease (PAD) is intermittent claudication that involves the calf, thigh, and/or buttock muscles. How the specific location of this leg pain is related to altered gait, however, is unknown. We hypothesized that because the location of claudication symptoms uniquely affects different leg muscle groups in people with PAD, this would produce distinctive walking patterns. A total of 105 participants with PAD and 35 age-matched older volunteers without PAD (CTRL) were recruited. Participants completed walking impairment questionnaires (WIQ), Gardner-Skinner progressive treadmill tests, the six-minute walk test, and we performed an advanced evaluation of the biomechanics of their overground walking. Participants with PAD were categorized into 4 groups according to their stated pain location(s): calf only (C, n = 43); thigh and calf (TC, n = 18); buttock and calf (BC, n = 15); or buttock, thigh, and calf (BTC, n = 29). Outcomes were compared between CTRL, C, TC, BC and BTC groups using a one-way ANOVA with post-hoc comparisons to identify and assess statistically significant differences. There were no significant differences between CTRL, C, TC, BC and BTC groups in distances walked or walking speed when either pain-free or experiencing claudication pain. Each participant with PAD had significantly dysfunctional biomechanical gait parameters, even when pain-free, when compared to CTRL (pain-free) walking data. During pain-free walking, out of the 18 gait parameters evaluated, we only identified significant differences in hip power generation during push-off (in C and TC groups) and in knee power absorption during weight acceptance (in TC and BC groups). There were no between-group differences in gait parameters while people with PAD were walking with claudication pain. Our data demonstrate that PAD affects the ischemic lower extremities in a diffuse manner irrespective of the location of claudication symptoms. ClinicalTrials.gov NCT01970332.

Supervised exercise therapy (SET) is a conservative non-operative treatment strategy for improving walking performance in patients with peripheral artery disease (PAD). Gait variability is altered in patients with PAD, but the effect of SET on gait variability is unknown. Forty-three claudicating patients with PAD underwent gait analysis before and immediately after a 6-month SET program. Nonlinear gait variability was assessed using sample entropy, and the largest Lyapunov exponent of the ankle, knee, and hip joint angle time series. Linear mean and variability of the range of motion time series for these three joint angles were also calculated. Two-factor repeated measure analysis of variance determined the effect of the intervention and joint location on linear and nonlinear dependent variables. After SET, walking regularity decreased, while the stability remained unaffected. Ankle nonlinear variability had increased values compared with the knee and hip joints. Linear measures did not change following SET, except for knee angle, in which the magnitude of variations increased after the intervention. A six-month SET program produced changes in gait variability toward the direction of healthy controls, which indicates that in general, SET improved walking performance in individuals with PAD.