Clearing the Stumbling Blocks: Management Strategies Associated With Minimum Foot Clearance in Prosthetic Ambulation

Home > Articles > Clearing the Stumbling Blocks: Management Strategies Associated With Minimum Foot Clearance in Prosthetic Ambulation
By Phil Stevens, MEd, CPO, FAAOP

When evaluating new patients for their first lower-limb prostheses, I often warn them that sooner or later they should expect to fall. We do not want these events to occur, but we want our patients to recognize the possibility and be able to get up when they happen.

Fall prevention also presents a challenge in geriatric medicine, and much of what we know about fall prevention among individuals with lower-limb amputations has been learned from studies of this larger, more prevalent population. One of the markers that has been used in geriatric fall prevention studies is the concept of minimum foot clearance (MFC), or the minimum distance observed between the plantar aspect of the swinging limb and the floor. It's a reasonable premise that as MFC reduces, the likelihood of a trip and subsequent fall becomes more likely. A 2010 systematic review on the topic summarized the findings of 12 published articles about MFC in geriatric fall prevention.1 What's more, the concept of MFC appears to be resonating in the academic world as, according to Google Scholar, this review has since been cited 87 times.

Recent years have seen the introduction of MFC as a potential causal agent in fall prevention among people with lower-limb amputations. This article reviews that growing body of literature and introduces this concept to practicing O&P clinicians.


At the 36th Annual Meeting of the American Society of Biomechanics, held in Gainesville, Florida, Shane Wurdeman, PhD, MSPO, CP, FAAOP, introduced the premise of MFC in fall prevention among people with lower-limb amputations.2 He and his collaborating authors examined 12 individuals with unilateral transtibial amputations and a control group of 11 unaffected individuals as they walked for three minutes on a treadmill at their self-selected walking speeds. As they did so, 3D markers were used to track the virtual position of the anterior and inferior aspects of the foot and its distance from the ground in swing phase, defined as minimum toe clearance (MTC).

Their findings confirmed reduced average MTC values on the prosthetic limb side compared to the sound side limb and the limbs of the control subjects.2 These findings are not unexpected. In the absence of active dorsiflexion at a prosthetic ankle, the relatively plantar-flexed ankle alignment reduces the distance between the swing limb and the floor. This reduced MTC, in turn, increases the likelihood of unintentional contact with the ground in the form of a trip.

Additionally, the authors report that while the average MTC of the sound side limb approximated that observed in the control population, its standard deviation was much higher, increasing the likelihood that it, too, appeared to be at an elevated risk of initiating a tripping event.2 Both limbs demonstrated the increased potential of triggering a stumble, but for different reasons. For the prosthetic limb, the lack of additional swing phase dorsiflexion reduced the distance between the limb and the ground. For the sound side limb, for reasons yet undefined, there was greater variation in the amount of swing phase clearance.2 


Until recent years, prosthetic ankle mechanisms presented as one of three variants. Axial ankles and dynamic response feet would deform under weight bearing conditions, returning to neutral alignment in swing, while more basic feet simply had a rigid attachment mechanism, lacking any attempt at ankle restoration. The introduction of hydraulic ankle-feet offers an alternate approach, as stance phase kinetics position the ankle in relative dorsiflexion during terminal stance, which can then be retained through swing phase, theoretically increasing the MTC of the prosthetic limb.

This theory was tested in two recent clinical trials involving hydraulic ankle-feet with and without microprocessor regulations. In the first trial, the authors recruited 21 active individuals with unilateral transtibial amputations.3 These individuals were brought to a gait lab where they performed a series of walking trials with their legacy nonhydraulic prosthetic feet: the Endolite Esprit foot (n = 12), the Endolite Multiflex foot (n = 6), and an Össur Flex-Foot, an Endolite Elite foot, and a Trulife Seattle Litefoot (n = 1 each).

After completing the baseline walking trials, subjects were fitted with a hydraulic ankle-foot, the Endolite Echelon, and given a 45-minute familiarization period. This included walking over slopes, stairs, grass, and other surfaces during which the hydraulic resistance settings of the foot were refined by an experienced prosthetist. Confirming a sense of having become accustomed to the new foot, the subjects returned to the gait lab for a second set of walking trials.3

 Subsequent analysis confirmed that use of the hydraulic foot significantly increased the average MTC of the prosthetic limb, from 1.76cm to 2.07cm. Further, the average MTC of the sound side limb also increased, from 2.12cm to 2.27cm.3 To account for the possible proximal compensations, gait lab data was used to confirm an absence of hip hiking on the affected side and verify that hip and knee flexion angles did not vary significantly between limbs.


In a related clinical trial, a cohort of eight individuals with unilateral transtibial amputations was evaluated while walking on a treadmill in an instrumented gait lab.4 In their baseline conditions, the subjects walked with their legacy nonmicroprocessor ankle-foot mechanisms at three walking speeds across two grades. The two grades were level and inclined, with the latter defined as a 5 percent incline. Walking speeds were the individually preferred self-selected velocity, 80 percent of the self-selected walking velocity, and 120 percent of the self-selected walking velocity.

With baseline gait lab data obtained, subjects transitioned to a microprocessor foot with a hydraulic ankle joint, the Össur Proprio Foot, which they acclimated to over four weeks before returning for a second iteration of laboratory tests. The authors observed that across walking conditions, i.e., velocity and slope, the MTC of the prosthetic limb side was approximately 70 percent larger with the use of the Proprio Foot.4 

By considering both the MTC values and the standard deviations associated with these values, the authors translated these changes to the likelihood of experiencing a tripping event. They observed that with the legacy prosthetic feet, described by the authors as nonactive dorsiflexion prostheses, patients would be expected to contact an unseen obstacle with a height of 5mm approximately once every 166 steps. By contrast, with the Proprio Foot, the likelihood of such incidental contact with a 5mm obstacle plummeted, with a projected occurrence of once every 3,169 steps.4 Predictably, incidental contact with a taller obstacle would occur more frequently. For example, the authors concluded that a 10mm obstacle would be struck once every ten steps with the use of a nonactive dorsiflexion prosthesis but only once every 50 steps with the Proprio Foot.4 

Researchers gained additional insights by examining the effect of velocity and slope on the relative changes to MTC values with the microprocessor foot. For example, when walking at 80 percent of the self-selected velocity on level ground, mean MTC values only increased 28 percent (from 17.8mm to 22.8mm). By contrast, during ambulation at 120 percent of self-selected velocity up a 5 percent incline, mean MTC values increased 65 percent (from 18.7mm to 30.8mm).4 


The effects of walking velocity on MTC values observed in prosthesis users received additional treatment in a recent gait lab evaluation of 12 individuals with unilateral transtibial amputations using the nonarticulating Endolite Esprit foot.5 In this analysis, participants were asked to walk slowly, at their normal walking speed, and as fast as comfortably possible. Once all walking trials were obtained, the authors analyzed MTC values on the prosthetic and sound limbs across variations in velocity.

On the sound side limb, MTC values increased significantly with increases in velocity, going from 2.3cm at 0.93 m/s, to 2.5cm at 1.13 m/s, to 2.6cm at 1.36 m/s. By contrast, the MTC values of the prosthetic limb were relatively unchanged, observed at approximately 1.1cm across all walking speeds.5 This highlights the inability of nonhydraulic prosthetic feet to situationally adapt to walking environments.


The previous studies were laboratory-based trials, with objective data taken from highly controlled walking conditions. In the last study under consideration, the authors attempted to correlate laboratory data with subsequent real-world experience. A cohort of eight subjects with unilateral transtibial amputations were brought into an instrumented gait lab.6 The subjects, who used a range of nonmicroprocessor, nonhydraulic ankles with carbon fiber feet, walked on a treadmill at their self-selected walking velocities to determine their MTC values. Considerable variations were observed in self-selected walking velocities (0.58 m/s-1.05 m/s) and mean MTC values (11.5mm- 33.3mm).6 However, consistent with the previous article, there were no strong correlations between gait speed and prosthetic side MTC values.

These values established, the authors then used electronic surveys to prospectively track the trip-related stumbles and falls of these individuals every two weeks for the next year. Every two weeks subjects were asked whether they fell, i.e., unintentionally came to rest on a lower surface, or stumbled, i.e., experienced a loss of balance that did not result in a fall, within the preceding two-week period. Follow-up questions sought to determine the cause of either event, including a trip or other factors such as slipping, legs giving out, external push or pull, etc.

After a year of completing the electronic surveys, the eight subjects were in one of three groups. Three subjects reported neither falls nor stumbles; two subjects reported no falls but multiple stumbles; and the final three subjects reported one or more falls and multiple stumbles.6 Tripping was the most common cause of stumbles with known etiology. Importantly, the observed mean MTC values of the prosthetic limb were roughly 50 percent lower for the three individuals who reported experiencing trip-related stumbles (12.3mm) than for the remaining five individuals who did not trip over the ensuing year (25.6mm).6 


This last study underlies the importance of its predecessors. While variations in MTC values as observed in the controlled walking conditions of an instrumented gait lab can suggest an increased risk for an unexpected trip during community ambulation, the prospective monitoring of actual tripping events underscores the reality of MTC values as a marker for stumble and fall risk. Given the findings of such pilot data, clinicians should consider fitting patients who have a history of stumbles and falls with hydraulic ankle-feet—with or without microprocessor control—that have been shown to increase MTC values, because they might reasonably decrease the individuals' risk of future stumbles or falls in the community.

Phil Stevens, MEd, CPO, FAAOP, is in clinical practice with Hanger Clinic, Salt Lake City. He can be contacted at 


1.    Barrett, R. S., P. M. Mills, and R. K. Begg. 2010. A systematic review of the effect of ageing and falls history on minimum foot clearance characteristics during level walking. Gait & Posture 32 (4): 429-35.

2.    Wurdeman, S., J. Yentes, S. Myers, A. Jacobson, and N. Stergiou. 2012. Both limbs in unilateral transtibial amputees display increased risk for tripping. In Proceedings of the 36th Annual Meeting of the American Society of Biomechanics (15-8).

3.    Johnson, L. J., A. R.De Asha, R. Munjal, J. Kulkarni, and J. G. Buckley. 2014. Toe clearance when walking in people with unilateral transtibial amputation: Effects of passive hydraulic ankle. Journal of Rehabilitation Research and Development 51 (3):429-38.

4.    Rosenblatt, N. J., A. Bauer, D. Rotter, and M. D. Grabiner. 2014. Active dorsiflexing prostheses may reduce trip-related fall risk in people with transtibial amputation. Journal of Rehabilitation Research and Development 51 (8), 1229-42.

5.    DeAsha, A. R. and J. G. Buckley. 2015. The effects of walking speed on minimum toe clearance and on the temporal relationship between minimum clearance and peak swing-foot velocity in unilateral trans-tibial amputees. Prosthetics and Orthotics International 39 (2):120-5.

6.    Rosenblatt, N. J., A. Bauer, and M. D. Grabiner. 2016. Relating minimum toe clearance to prospective, self-reported, trip-related stumbles in the community. Prosthetics and Orthotics International DOI: 10.1177/0309364616650085,