Fall simulations help AK amputees
By Jordana Bieze Foster
Thanks to an improved ability to simulate falls in the laboratory, researchers are identifying variables that contribute to falls in above-knee amputees and implementing training programs to improve response to tripping.
Studies from two different groups, presented in August at the annual meeting of the American Society of Biomechanics, utilized a computerized treadmill to simulate falls through unanticipated bursts of belt acceleration.
Researchers from Brown University used this type of technology to assess fall response in five unilateral transfemoral amputees walking at a self-selected pace. Each subject performed two sets of 13 trials, with each set including 10 perturbations and three uninterrupted walking trials in random order. For the perturbation, the treadmill was accelerated 2.2 m/s for 0.16 seconds.
Motion analysis and surface electromyography demonstrated a passive, mechanical knee flexion response to the perturbation when the prosthetic limb was on the treadmill. When the prosthetic limb was in swing phase, however, the early response involved an increase in thigh muscle activation corresponding to knee flexion angle.
Like the Brown group, researchers from the University of Illinois in Chicago also found that response to tripping in unilateral transfemoral amputees depended on which limb was on the ground at the time of the perturbation. The UIC investigators used motion capture to analyze five amputees as they walked across a floor with a concealed obstacle that could be triggered to rise to a 5 cm height. Each amputee was subjected to an obstruction of each limb.
During failed recoveries when the recovery step was taken with the prosthetic limb, knee flexion on that support limb was associated with a low vertical hip height. Obstruction of the non-prosthetic limb was associated with a short step length on the part of the recovering limb and an increased peak trunk flexion angle velocity.
Those variables, which had previously been shown to contribute to falls in older adults, became the outcome measures used to assess a treadmill training intervention completed by the same five study subjects. Each subject attended six training sessions over the course of two to three weeks.
Each session included two series of perturbed treadmill bouts, with one series targeting the prosthetic limb and the other targeting the non-prosthetic limb. Using the computerized treadmill, disturbances were progressively made more challenging within each series and from one session to the next as subjects demonstrated successful recovery strategies. Perturbations involved belt accelerations lasting 0.5 seconds, starting at 1.5 m/s2 and progressing to 6.25 m/s2.
Three of the five subjects fell during the first training session. By the sixth session, all three were able to successfully recover from the same level of disturbance that had initially caused them to fall.
Training was not associated with any significant change in hip height. However, recovery step length for the prosthetic limb significantly increased during the training period, while time to recovery step from the disturbance onset significantly decreased for both limbs. Trunk flexion angle also significantly decreased with training in both limbs, with trunk flexion velocity significantly decreasing for the prosthetic limb. The largest improvements were seen in step length and trunk flexion velocity for the prosthetic limb during the most challenging disturbances.
The authors hypothesized that the training may lead to adaptations involving the plantar flexor muscles of the non-prosthetic limb, which would be more likely to affect recovery step length on the prosthetic side, according to Jeremy Crenshaw, MS, a research assistant in the Musculoskeletal Biomechanics Laboratory at UIC, who presented his group’s findings at the ASB meeting.
Crenshaw also noted that the subject who had the most successful response to training was also the tallest subject, suggesting that height may be a contributing factor.