Heel Rise and Non–Weight-Bearing Ankle Plantar Flexion Tasks to Assess Foot and Ankle Function in People With Diabetes Mellitus and Peripheral Neuropathy

Abstract

Objective

The objective of this study was to examine the effects of diabetes mellitus and peripheral neuropathy (DMPN), limited joint mobility, and weight-bearing on foot and ankle sagittal movements and characterize the foot and ankle position during heel rise.

Methods

Sixty people with DMPN and 22 controls participated. Primary outcomes were foot (forefoot on hindfoot) and ankle (hindfoot on shank) plantar-flexion/dorsiflexion angle during 3 tasks: unilateral heel rise, bilateral heel rise, and non–weight-bearing ankle plantar flexion. A repeated-measures analysis of variance and Fisher exact test were used.

Results

Main effects of task and group were significant, but not the interaction in both foot and ankle plantar flexion. Foot and ankle plantar flexion were less in people with DMPN compared with controls in all tasks. Both DMPN and control groups had significantly less foot and ankle plantar flexion with greater weight-bearing; however, the linear trend across tasks was similar between groups. The DMPN group had a greater percentage of individuals in foot and/or ankle dorsiflexion at peak unilateral heel rise compared with controls, but the foot and ankle position was similar at peak bilateral heel rise between DMPN and control groups.

Conclusion

Foot and ankle plantar flexion is less in people with DMPN. Less plantar flexion in non–weight-bearing suggests that people with DMPN have limited joint mobility. However, peak unilateral and bilateral heel rise is less than the available plantar flexion range of motion measured in non–weight-bearing, indicating that limited joint mobility does not limit heel rise performance. A higher frequency of people with DMPN are in foot and ankle dorsiflexion at peak unilateral heel rise compared with controls, but the position improved with lower weight-bearing.

Impact

Proper resistance should be considered with physical therapist interventions utilizing heel rise because foot and ankle plantar flexion position could be improved by reducing the amount of weight-bearing.

Introduction

Lower extremity musculoskeletal complications are frequently observed in people with diabetes mellitus (DM) and peripheral neuropathy (PN)¹ and include limited joint mobility, impaired strength, and atrophy and fat infiltration of muscles.^2–6 Together, these complications contribute to foot and ankle movement dysfunction, operationally defined as deficits in the magnitude and direction of foot and ankle plantar flexion during weight-bearing tasks. Cross-sectional data support a relationship between foot and ankle movement dysfunction and midfoot deformity.^7,8 Although there are currently no longitudinal data to support a causal relationship between aberrant movement and deformity, the physical stress theory proposes injuries develop when the repetitive stresses exceed the tissues’ tolerance.⁹ We hypothesize a theoretical framework in which foot and ankle movement dysfunction during weight-bearing tasks, in the presence of loss of protective sensation, contributes to the development and progression of foot deformity, thereby increasing the risk of skin breakdown and amputation in people with DM and PN (DMPN).^10,11 Understanding foot and ankle movement dysfunction in DMPN would potentially inform treatment strategies to complete safe exercise programs and daily activities, which might improve foot and ankle function and reduce the risk of developing foot complications.

The heel rise task is a common test that has been used to identify foot and ankle dysfunction across a variety of pathologies.^12–17 Performance of the heel rise task requires foot and ankle plantar flexion range of motion and strength^7,18 as well as simultaneous interplay of foot and ankle plantar flexion motions.¹³ In people with DMPN and midfoot deformity, a unilateral heel rise task was performed with 85% less foot (forefoot relative to hindfoot) and 65% less ankle (hindfoot relative to shank) plantar flexion excursions compared with non-DMPN controls.¹⁴ Although a previous study showed a significant association of calf strength and foot muscle deterioration to unilateral heel rise performance in people with DMPN,⁷ the study design was unable to discern if limited foot and ankle joint mobility, a common complication of DM,⁵ contributed to the aberrant heel rise performance. The use of a bilateral heel rise and non–weight-bearing ankle plantar flexion task to reduce weight-bearing would help discern the role of muscle strength and limited joint mobility on impaired unilateral heel rise performance.

Failure to plantarflex the foot and ankle by the end of the heel rise task is particularly concerning, as the sustained dorsiflexed position of the midfoot during the heel rise task could indicate a movement dysfunction that is contributing to midfoot deformity.¹⁴ Investigating the simultaneous interplay between foot and ankle plantar flexion at peak heel rise and examining the trajectories of the joints together could have substantial clinical implications for defining the magnitude and direction of normal and aberrant movement patterns during the heel rise task. However, foot and ankle position and trajectories in the sagittal plane during heel rise have not been characterized in people with DMPN.

Identifying the foot and ankle movement dysfunction during heel rise tasks could guide physical therapists on treatment strategies to improve foot function and prevent foot injury, deformity progression, and ultimately ulceration and amputation in people with DMPN. Therefore, the primary purpose of this study was to examine the effects of DMPN, limited joint mobility, and weight-bearing on foot and ankle sagittal plane movements. We assessed foot and ankle plantar flexion during 3 tasks (unilateral heel rise, bilateral heel rise, and non–weight-bearing ankle plantar flexion) in 2 groups (DMPN and controls without DMPN). We hypothesized that people with DMPN, compared with controls, would have (1) fewer foot and ankle movements during the heel rise tasks (effect of DMPN on heel rise performance), (2) less foot and ankle movement in the non–weight-bearing ankle plantar flexion task (limited joint mobility associated with DMPN), and (3) greater reductions in foot and ankle movements with greater weight-bearing (unilateral vs bilateral vs non-weight-bearing). The secondary purpose of this study was to characterize the position of the foot and ankle at the peak heel rise height and non–weight-bearing ankle plantar flexion tasks and qualitatively compare the movement trajectories of the foot and ankle during unilateral and bilateral heel rise tasks. We hypothesized that the (1) percentage of people with a dorsiflexed foot and ankle position would be greater in the DMPN group compared with controls, and (2) foot and ankle trajectories would be altered between unilateral and bilateral heel rise tasks in 4 movement pattern subgroups identified by unilateral heel rise task.

Methods

Participants

Sixty people comprised the DMPN group and 22 people comprised the control group. Inclusion criteria for the DMPN group were (1) type 2 DM diagnosed by the participant’s physician, and (2) PN assessed by the research team. Presence of PN was defined as (1) the inability to sense a 5.07 monofilament on at least 1 out of 6 plantar locations,¹⁹ (2) inability to feel vibration perception threshold less than 25 V tested on the plantar surface of the great toe using a biothesiometer (Biomedical Instrument Co, Newbury, OH, USA),¹⁹ or (3) Michigan Neuropathy Screening Instrument score ≥2.²⁰ PN from causes other than DM was excluded (eg, chemo toxic, alcoholic, lumbar radiculopathy). Inclusion criteria for the control group were no DM or PN. The control group had similar age and body mass index (BMI) compared with DMPN group (Tab. 1). Exclusion criteria for both DMPN and control groups were inability to complete the testing for the study, age >75 years old, pregnant, on dialysis, severe arterial disease (ankle-brachial index >1.3 or <0.9), rigid metatarsophalangeal deformity, presence of a foot ulceration, lower extremity amputation, weight >180 kg, and metal implants and/or pacemaker. All participants read and signed the consent form prior to participating in the study. The protocol was approved by the Washington University Institutional Review Board.

Table 1.

Participant Characteristics^a

Characteristic	DMPN	Control	P
Participants, no.	60	22	—
Sex, female/male	34/26	14/8	.621
Age, y	67 (6) [46–75]	62 (8) [46–74]	.003b
Body mass index, kg/m2	35 (7) [22–49]	32 (6) [21–45]	.088
DM duration, y	14 (10) [0.2–49]	—	—
Hemoglobin A1c, %	7.1 (1.3) [5.1–11.4]	—	—
Foot length, cm	27.7 (1.7) [24.2–31.9]	27.3 (1.9) [24.7–32.2]	.381
Truncated foot length, cm	20.0 (1.2) [17.6–22.9]	19.9 (1.5) [17.6–23.2]	.663

^a Values are mean (SD) [range]. DMPN = diabetes mellitus and peripheral neuropathy.

^b Statistically significant (P < .05).

Kinematic and Kinetic Measurements

A modified Oxford multisegmented foot model (3 segments: forefoot, hindfoot, and shank) was used to assess foot motion (forefoot relative to hindfoot) and ankle motion (hindfoot relative to shank). Reflective markers (10 mm) were attached to participants’ anatomical landmarks as defined previously^14,21 and are provided in the Supplementary Appendix. Kinematic data were acquired using a 10-camera Vicon motion analysis system (Vicon MX, Los Angeles, CA, USA; 100 Hz). Kinematic data were computed using a Cardan x-y-z sequence rotations. The marker located at the posteroinferior aspect of the heel was used to estimate heel height during the heel rise tasks. Kinetic data were collected with Bertec force plates (FP4060–10 model, Bertec Corporation, Columbus, OH, USA; 1000 Hz). After data collection, kinematic data were low-pass filtered using a Butterworth filter with a 6-Hz cut-off frequency in Visual3D software (C-Motion Inc. Germantown, MD, USA). The zero position of the foot was defined as the position of the forefoot relative to the hindfoot segment during the static standing trial. The zero position of the ankle was defined as the position of the shank relative to the hindfoot segment using the global coordinate system.²²

The present study is a secondary analysis of the baseline time point of a longitudinal, parent study (Clinicaltrials.gov, NCT02616263). The target limb of the DMPN group was selected based on the criteria of the parent study that measured toe extension movement pattern associated with metatarsophalangeal joint deformity. The examiner chose the foot that was observed to have the most consistent pattern of toe extension movement during active ankle dorsiflexion, the foot with the least number of foot complications (ie, history of surgery or traumatic injury), and the metatarsophalangeal joint hyperextension angle needed to match the parent study treatment groups. The target limb of the control group was randomly selected prior to data collection. The percentages of right and left target feet were matched between DMPN and control groups.

Tasks

Heel Rise Tasks

For the unilateral heel rise, participants placed the target limb on the force plate, and the non-target limb was flexed at the knee so the foot did not touch the force plate. For the bilateral heel rise, the participants placed the target limb on the force plate and non-target limb on another force plate. For both heel rise tasks, participants placed their hands on the examiner’s outstretched forearm for balance. Participants were asked to raise the heel as high as possible without bending the knee of the stance leg. Participants performed 5 repetitions of each heel rise task. We aimed to examine the trials that represented the participant’s best performance; therefore, we selected the 3 trials that had the highest ankle plantar flexion power for analysis.^14,22 The foot and ankle sagittal dorsiflexion (+)/plantar flexion (−) angles, ankle frontal inversion (+)/ eversion (−) angles, and heel height at the peak heel rise height were averaged and analyzed. Normalized peak heel height was calculated to minimize the influence of different foot lengths [peak heel height/ truncated foot length × 100 (%)]. The truncated foot length was defined as the distance between the first metatarsal head to the heel.

The vertical ground reaction force (vGRF) at peak unilateral and bilateral heel rise was analyzed to assess load through the foot as a result of the upper extremity support and less weight-bearing. The vGRF at peak heel rise was normalized to the participant’s static calibration vGRF and multiplied by 100. Because of upper extremity support for balance, the normalized vGRF at peak unilateral heel rise (target-limb) was 91% (SD: 6) for the DMPN and 91% (7) for controls and at peak bilateral heel rise (target + non-target limbs) was 95% (3) for DMPN and 96% (2) for controls. The vGRF of target-limb at peak bilateral heel rise was 48% (3) for DMPN and 48% (3) for controls. There were no group differences in vGRF on the target limb (independent t test: unilateral heel rise P = 1.00, bilateral heel rise P = 1.00), which indicates that the influence of balance support during heel rise was similar across groups.

Non–Weight-bearing Ankle Plantar Flexion Task

Participants were asked to sit on a plinth with the knee extended and the ankle and foot in a relaxed position. They were asked to actively move their foot into maximal plantar flexion followed by maximal dorsiflexion for 5 consecutive repetitions. The 2 highest values of foot and ankle plantar flexion angle were analyzed and averaged.

Position of Foot and Ankle at Peak

The position of the foot and ankle at peak heel rise height for the heel rise tasks and at peak foot and ankle plantar flexion for the non–weight-bearing ankle plantar flexion task was classified in 4 subgroups (Fig. 1): the plantar flexed position was foot and ankle plantar flexed (foot PF and ankle PF); the dorsiflexed positions were foot plantar flexed and ankle dorsiflexed (foot PF and ankle DF), foot dorsiflexed and ankle plantar flexed (foot DF and ankle PF), and foot and ankle dorsiflexed (foot DF and ankle DF).

Figure 1.

Position of foot and ankle at peak heel rise. (A) Foot PF and ankle PF. (B) Foot PF and ankle DF. (C) Foot DF and ankle PF. (D) Foot DF and ankle DF. Black arrows indicate PF motion and red arrows indicate DF motion in foot and ankle joints. (A) Depicts the plantar flexed position, whereas (B)–(D) depict dorsiflexed position of the foot and/or ankle. DF = dorsiflexion; PF = plantar flexion.

Trajectory Analysis

The trajectory of the mean foot and ankle angles during the unilateral and bilateral heel rise was graphed for each of the subgroups identified from the foot and ankle position at the end of unilateral heel rise. We qualitatively compared the foot and ankle movement trajectories between the 4 subgroups.

Statistical Analysis

All statistical analyses were conducted using SAS 9.4 (SAS Institute, Cary, NC, USA). An alpha level of .05 was set for all statistical analyses. An independent t test and a Fisher exact test were used to assess differences in participant characteristics.

Effect of DMPN, Limited Joint Mobility, and Greater Weight-Bearing

The Shapiro–Wilk test was conducted to test the assumptions for data normality and the dependent variables were normally distributed. Thus, the statistical analysis did not require modification. Repeated-measures analysis of variance was used to analyze the association of 3 dependent variables (unilateral heel rise, bilateral heel rise, non–weight-bearing) with group (DMPN and controls) to determine the effect of DMPN, limited joint mobility, and greater weight-bearing on foot and ankle sagittal movements. The within-subjects factor was task (unilateral heel rise, bilateral heel rise, and non–weight-bearing task) and the between-subjects factor was group (DMPN and controls). Due to multiple comparisons, Bonferroni correction was used to adjust the alpha level (α/3 = .017).

There were no differences between groups for covariates except for age. Repeated-measures analysis of covariance with age as covariate was conducted to determine if age influenced the primary outcome.

Additional kinematic variables (unilateral and bilateral heel rise ankle inversion, peak heel height, and normalized peak heel height) were added retrospectively to clarify the biomechanics of the heel rise. An independent t test was used to compare the additional kinematic variables between the DMPN and control groups. An alpha level was adjusted (α/2 = .025) using the Bonferroni correction.

Position of Foot and Ankle

A Fisher exact test was used to examine differences in the percentage of participants in plantar flexed (foot PF and ankle PF) and dorsiflexed positions (including foot PF and ankle DF, foot DF and ankle PF, and foot DF and ankle DF) of the foot and ankle between DMPN and controls within each task (unilateral heel rise, bilateral heel rise, and non–weight-bearing ankle plantar flexion).

Role of the Funding Source

This study was supported by grants from the National Institute of Diabetes and Digestive and Kidney Diseases (R01 DK107809, F32 DK123916) and the Research Division of the Program in Physical Therapy, Washington University School of Medicine, St. Louis, Missouri. The funders played no role in the design, conduct, or reporting of this study.

Results

Effect of DMPN on Heel Rise Performance

There was a significant main effect of group (foot: P < .001 and ankle: P < .001). In our post hoc analysis, the DMPN group exhibited significantly less foot and ankle plantar flexion in the unilateral heel rise task (foot: P < .001 and ankle: P < .001) and bilateral heel rise task (foot: P < .001 and ankle: P = .001) compared with the control group (Tab. 2).

Table 2.

Kinematic Data^a

Variables	Tasks	DMPN	Control	Mean Difference b	P
Foot plantar flexion, degreesc	Unilateral heel rise	1 (6)	−7 (9)	8	<.001e
	Bilateral heel rise	−10 (7)	−17 (8)	7	<.001e
	Non–weight-bearing	−22 (7)	−26 (5)	4	.001e
Ankle plantar flexion, degreesc	Unilateral heel rise	0 (7)	−6 (6)	6	<.001e
	Bilateral heel rise	−10 (6)	−15 (4)	5	.001e
	Non–weight-bearing	−18 (5)	−23 (7)	5	.004e
Ankle inversion, degreesd	Unilateral heel rise	5 (7)	7 (5)	−2	.079
	Bilateral heel rise	12 (7)	14 (6)	−2	.298
Peak heel height, cm	Unilateral heel rise	6.5 (2.0)	9.1 (2.2)	−2.6	<.001e
	Bilateral heel rise	9.0 (1.8)	10.8 (1.6)	−1.8	<.001e
Normalized peak heel height, %f	Unilateral heel rise	32.3 (9.8)	46.0 (10.8)	−13.7	<.001e
	Bilateral heel rise	45.1 (9.1)	54.6 (7.5)	−9.5	<.001e

^a Values are mean (SD). DMPN = diabetes mellitus and peripheral neuropathy.

^b Mean difference = DMPN – control.

^c Positive value indicates dorsiflexion and negative value indicates plantar flexion motion.

^d Positive value indicates inversion motion.

^e Statistically significant (P < .017 for foot and ankle plantar flexion and P < .025 for ankle inversion, peak heel height, and normalized peak heel height).

^f Peak heel height (cm)/truncated foot length (cm) × 100 (%).

Limited Joint Mobility Associated With DMPN

A post hoc analysis showed that the DMPN group had significantly less foot and ankle plantar flexion during the non–weight-bearing task compared with the control group (foot: P = .001 and ankle: P = .004; Tab. 2).

Effect of Greater Weight-Bearing

There was a significant main effect of task (foot: P < .001 and ankle: P < .001). A post hoc analysis showed that foot and ankle plantar flexion were significantly less in the bilateral heel rise compared with the non–weight-bearing task (foot: P < .001 and ankle: P < .001) and unilateral heel rise compared with bilateral heel rise (foot: P < .001 and ankle: P < .001) in both DMPN and control groups. The difference in the linear trend of foot and ankle plantar flexion across tasks between DMPN and control groups was not statistically significant (interaction of task × group, foot: P = .052 and ankle: P = .092).

Repeated-measures analysis of covariance showed that age did not influence the outcome. Ankle inversion was not significantly different between DMPN and control groups in both unilateral and bilateral heel rise (P = .079 and P = .298, respectively; Tab. 2). Peak heel height and normalized peak heel height were significantly less in the DMPN group compared with the control group in the unilateral and bilateral heel rise task (Tab. 2).

Position of Foot and Ankle

The percentage of people with foot and/or ankle in a plantar flexed and dorsiflexed position in the DMPN and control groups is reported in Fig. 2A. In the unilateral heel rise, the DMPN group had a significantly greater percentage of people in the foot and/or ankle dorsiflexed position (40 out of 60, 67%) compared with controls (7 out of 22, 32%; P = .01). In the bilateral heel rise, there were no group differences in percentage of people in the plantar-flexed and dorsiflexed position between DMPN and controls (8 out of 60, 13% vs 1 out of 22, 5%, respectively; P = .43). For the non–weight-bearing task, all participants in both groups were in the plantar flexed foot and ankle position (Fig. 2B).

Figure 2.

(A) The percentage of participants in the 4 foot and ankle positions in the DMPN and control groups. Foot PF and ankle PF (green) is the plantar flexed position of foot and ankle. Foot PF and ankle DF (bronze), foot DF and ankle PF (brown), and foot DF and ankle DF (red) indicate foot and/or ankle are in the opposite position of the plantar flexion. (B) Scatter plot data of peak foot and ankle position (degrees). Quadrant description (starting at upper-right, clockwise): foot DF and ankle DF, foot DF and ankle PF, foot PF and ankle PF, foot PF and ankle DF. Red circles represent performance of the unilateral heel rise, yellow diamonds represent performance of the bilateral heel rise, and blue squares represent performance of the non–weight-bearing task. Shift toward the lower left quadrant indicates greater foot and ankle plantar flexion at each task. DF = dorsiflexion; DMPN = diabetes mellitus and peripheral neuropathy; PF = plantar flexion.

Trajectory Analysis

The foot and ankle trajectory of the DMPN group during the unilateral and bilateral heel rise tasks are in Figure 3. During unilateral heel rise, subgroups in the foot plantar flexed position (foot PF and ankle PF and foot PF and ankle DF) had foot plantar flexion movement trajectories, whereas subgroups in the foot dorsiflexed position (foot DF and ankle PF and foot DF and ankle DF) had foot dorsiflexion movement trajectories. All 4 subgroups during unilateral heel rise had ankle plantar flexion movement trajectories (Fig. 3A). During bilateral heel rise, all 4 subgroups had both foot and ankle plantar flexion movement trajectories (Fig. 3B).

Figure 3.

Foot and ankle movement trajectory depicting (A) unilateral heel rise and (B) bilateral heel rise in DMPN group. Subgroup allocations were based on foot and ankle position at peak unilateral heel rise and maintained for the bilateral heel rise. (A) The unilateral trajectories showed that all subgroups plantar flexed the ankle (moving down the y-axis) during the unilateral task. Subgroups with the foot DF pattern dorsiflexed the foot during unilateral heel rise (moving to the right on the x-axis). Changes in the trajectories from (A) to (B) show that all subgroups had greater excursions and restoration of foot and ankle plantar flexion when weight bearing was reduced from unilateral to bilateral heel rise. Each dot is a normalized time point of heel rise from 0% to 100%. DF = dorsiflexion; DMPN = diabetes mellitus and peripheral neuropathy; PF = plantar flexion.

Discussion

The results of this study determined that (1) people with DMPN have less foot and ankle plantar flexion during the heel rise tasks; (2) people with DMPN have limited foot and ankle plantar flexion mobility measured during the non–weight-bearing task; (3) greater weight-bearing reduces foot and ankle movements; however, the reduction across tasks is similar between people with DMPN and controls; and (4) people with DMPN have a dorsiflexed foot and ankle position and aberrant trajectory that is most apparent in the unilateral heel rise, but reducing weight-bearing helps improve foot and ankle plantar flexion motion.

Presence of DMPN was associated with less unilateral and bilateral heel rise foot and ankle plantar flexion. In a previous study,¹⁴ the mean difference of foot and ankle plantar flexion between those with DMPN and controls during unilateral heel rise was greater (foot: 21 degrees and ankle: 11 degrees) than what was measured in this study (foot: 8 degrees and ankle: 6 degrees). The difference between studies likely reflects progression of foot and ankle movement dysfunction as foot pathology worsens, as the previous study examined a cohort with established medial column foot deformity. Deficits in both unilateral and bilateral heel rise tasks in the DMPN group suggest that people with DMPN have substantial loss of muscle strength that may lead to aberrant foot and ankle biomechanics during daily activities. Examination of muscle strength as well as foot and ankle movements of patients with DM utilizing the heel rise task could assist physical therapists in prescribing appropriately dosed weight-bearing exercise programs. Foot and ankle movements should be observed to prescribe effective and safe exercise programs. For example, bicycling or swimming would be recommended if abnormal foot mechanics are observed and cannot be corrected, theoretically reducing stress and potentially minimizing injury to the foot but promoting physical activity to help glycemic control.

We observed limited ankle and foot plantar flexion mobility in people with DMPN compared with controls during the non–weight-bearing task (reduction of 22% at ankle and 15% at foot plantar flexion). Our ankle findings are similar to Abate and colleagues⁵ who found a 24% reduction in ankle plantar flexion range of motion in people with DM compared with age-matched healthy controls. To our knowledge, limited foot plantar flexion mobility has not been previously reported in people with DMPN. Multiple factors can contribute to reduced range of motion, including aging²³ and glycemic control.²⁴ In general, DM is known to cause an increased formation of advanced glycation end-products^25,26 that results in increased collagen cross-links associated with joint stiffness and decreased range of motion.^5,27 Interventions to address foot and ankle plantar flexion range of motion could improve foot and ankle function and minimize risks associated with movement dysfunction.

All of the participants, regardless of group, were unable to plantar flex through their available full range of motion, measured during the non–weight-bearing ankle plantar flexion task, during the heel rise tasks. Specifically, the reduction in foot and ankle plantar flexion from non–weight-bearing to bilateral heel rise was 55% and 44% for the DMPN group, respectively, and 35% and 35% for the control group, respectively. The loss of foot and ankle motion as weight-bearing increases implies that the impaired heel rise performance in DMPN and control groups is related to muscle weakness, not the limited joint mobility.

A trend of greater foot plantar flexion reduction with greater weight-bearing in people with DMPN compared with controls suggests that the foot is more vulnerable to deform with greater weight-bearing in people with DMPN. PN in people with DM accelerates the muscle degenerative process, which results in greater impairments in muscle.^28–30 Individuals with DMPN, compared with controls, have up to 49% less intrinsic foot muscle volume.²⁹ Activation of intrinsic foot muscle plays an important role in stiffening the foot during weight-bearing tasks.^31,32 In a previous work, midfoot power contributed 36% of variance of unilateral heel rise in healthy individuals,³³ which suggests the importance of foot strength in performing the weight-bearing heel rise task. From the results of our study combined with previous findings, we speculate that improving the foot intrinsic muscle strength could increase foot plantar flexion during weight-bearing tasks.

Sixty-seven percent of people with DMPN did not end the unilateral heel rise task with the foot and/or ankle in plantar flexion. Fifty-three percent of people with DMPN were unable to complete the unilateral task with the foot plantar flexed, and the foot DF pattern subgroups (foot DF and ankle PF and foot DF and ankle DF) moved their foot into dorsiflexion during the unilateral heel rise (Fig. 3A). Importantly, when weight bearing was lower with the bilateral heel rise task, the majority of people with DMPN were able to restore the final foot and ankle position and trajectory to plantar flexion (Figs. 2A and B and 3B). Thus, with lower weight bearing, the magnitude, direction, and end position of the foot and ankle were improved. This finding implies the importance of examining foot and ankle movements during unilateral and bilateral heel rise to understand how weight bearing influences foot and ankle movements. An identification of movement dysfunction in the early disease process would provide important targets for interventions aimed at reducing the risk of foot complications.

It is interesting to note that controls also had a substantial percentage of people who were unable to reach a foot and ankle plantar flexed position at peak unilateral (32%) and bilateral (5%) heel rise. A previous study¹³ has shown that 100% of controls achieved foot plantar flexion (first metatarsal relative to hindfoot) during bilateral heel rise. Our control group was slightly older and heavier (age = 62 years old, BMI = 32.1 kg/m²) compared with the previously reported controls¹³ (age = 56 years old, BMI = 30.6 kg/m²). Aging and obesity have been found to be associated with less plantarflexor strength^34,35; thus, a combination of aging and obesity may have contributed to aberrant foot and ankle position at peak heel rise. A previous study of Flanagan et al³⁶ demonstrated an aging-associated response to body weight during heel rise that showed significant ankle (foot relative to shank) plantar flexion reduction in older adults during unilateral heel rise compared with that of bilateral heel rise task. Future study is needed to investigate the contribution of aging and BMI on altering foot and ankle function during heel rise.

Limitations

There are several limitations to our study. People with DMPN may have leaned forward or slightly bent their knees to raise their heels. Future kinematic examination of the heel rise task could include trunk and femur segments to allow tracking of compensatory strategies. Although practice trials and balance support were given during unilateral and bilateral heel rise, people with DMPN may have balance deficits that were not measured that could contribute to difficulty completing the tasks. The foot mechanics during non–weight bearing do not fully replicate foot motion during a weight-bearing heel rise task. The weight-bearing heel rise tasks include components of shear and compressive stress at the joint that may inherently change joint motion. Lastly, this study was designed as a cross-sectional study that does not explain causal relationships of heel rise task and development of lower extremity musculoskeletal problems. Future study is needed to determine the relationship of foot deformity progression to heel rise performance.

In conclusion, the presence of DMPN and greater weight bearing was associated with reduced foot and ankle movements. People with DMPN have limited joint mobility, but that does not appear to be the primary factor limiting foot and ankle kinematics during a heel rise task. People with DMPN not only performed fewer foot and ankle movements during heel rise task but also had foot and/or ankle dorsiflexion during the unilateral heel rise task. However, reducing the weight bearing helped improve foot and ankle plantar flexion magnitude, joint position, and movement trajectories. These findings provide useful information in the utilization of the heel rise task to quickly identify foot and ankle movement dysfunction and to provide guidance for the appropriate weight bearing during exercise prescription.

Author Contributions

Concept/idea/research design: H. Jeong, M.J. Mueller, M.K. Hastings

Writing: H. Jeong, Y. Yan, M.K. Hastings

Data collection: H. Jeong, M.K. Hastings

Data analysis: H. Jeong, M.J. Mueller, Y. Yan, M.K. Hastings

Project management: H. Jeong, M.K. Hastings

Fund procurement: M.K. Hastings

Providing participants: M.K. Hastings

Providing facilities/equipment: M.K. Hastings

Providing institutional liaisons: M.K. Hastings

Consultation (including review of manuscript before submitting): H. Jeong, M.J. Mueller, J.A. Zellers, Y. Yan, M. K. Hastings

Acknowledgments

The authors acknowledge Kathryn Bohnert, Darrah Snozek, and Christopher Sorensen, who assisted with participant recruitment and data collection; and Jessica Stumpf, Kaitlyn Winter, Jadean Hoff, Hana Bernhardson, Haley Brogan, Nick Youmans, Mary Ellis, Whitney Korgan, and Nick Schroeder, who assisted with data processing.

Ethics Approval

This study was approved by the Washington University Institutional Review Board.

Funding

This study was supported by grants from the National Institute of Diabetes and Digestive and Kidney Diseases (R01 DK107809, F32 DK123916) and the Research Division of the Program in Physical Therapy, Washington University School of Medicine, St Louis, Missouri.

Clinical Trial Registration

This study is a secondary analysis of the baseline time point of a longitudinal parent study and is registered in Clinicaltrials.gov (NCT02616263).

Disclosure and Presentations

The authors completed the ICMJE Form for Disclosure of Potential Conflicts of Interest and reported no conflicts of interest.

This study will be presented virtually as a platform presentation at the International Foot and Ankle Biomechanics Conference, April 2021, Sao Paulo, Brazil.

This study was presented as a platform presentation at the American Physical Therapy Association’s (APTA) Combined Sections Meeting, February 2020, Denver, CO.

This study was presented as a poster presentation at the following events: the DRC Diabetes Day Symposium, November 2019, St Louis, MO; the 14th Research Training Symposium and Poster Session, October 2019, St Louis, MO; the Washington University in St Louis 24th Graduate Research Symposium, March 2019, St Louis, MO; the APTA Combined Sections Meeting, February 2018, New Orleans, LA.