Contribution of Plantar Fascia to the Increased Forefoot Pressures in Diabetic Patients

RESEARCH DESIGN AND METHODS

Patient recruitment

A total of 61 patients were recruited from the outpatient clinic of the Metabolic Diseases Department at the University of Rome “Tor Vergata.” They were divided into three groups: 27 diabetic patients without neuropathy (D group), 19 diabetic patients with neuropathy (DN group), and 15 diabetic patients with previous neuropathic ulcers (DNPU group). A total of 21 healthy volunteers (control [C]) were also examined for comparison. The groups of patients and healthy volunteers were recruited homogeneously as for sex, age, BMI, and occupation. The level of neuropathy was assessed by using the neuropathy disability score (NDS) and the vibration perception threshold (VPT). NDS (14) is the product of scoring ankle reflex (0 = normal, 1 = present with reinforcement, 2 = absent) plus vibration, pin-prick, and temperature sensation (cold turning fork) at the great toe (0 = normal, 1 = abnormal). VPT was assessed by using a biothesiometer on the tip of the great toe. The mean of three readings was considered.

For the purpose of this study, the presence of neuropathy in the participating diabetic patients was defined by NDS >5 V and VPT >25 V (15). All patients in the D group had NDS ≤4 V and VPT <25 V. Those subjects with a history of previous neuropathic foot ulcers up to 3 months before the study but without active ulcers were included in the DNPU group. All previous ulcers were localized at the metatarsal heads. If present, callosities were removed before measurements.

Exclusion criteria.

The exclusion criteria included: 1) age >70 years; 2) history of peripheral vascular, neurological (other than those of diabetic etiology), musculo-skeletal, or rheumatic disease; 3) any major or minor amputation; 4) peripheral vascular diseases with ankle brachial index <0.85; and 5) Charcot neuroarthropathy and hallux rigidus caused by previous traumas.

The study was approved by the local ethical committee, and all patients gave their informed written consent before the screening, in accordance with the principles of the Helsinki Declaration.

Experimental set-up and measurement protocol

For ultrasound examination, we used a high-resolution 8- to 10-MHz linear array (Toshiba Tosbee SSA 240). All exams were performed by rheumatologists experienced in musculo-skeletal ultrasound technique. The thickness of plantar fascia was examined in both feet in a comparative manner. Plantar fascia was measured under slight tension in the axial and sagittal planes. Thickness was examined for the whole length of the plantar fascia. Measurements (expressed in mm) were taken at one-fifth of the length, starting from the calcaneous insertion, on the medial aspect of the foot. This is a point of the plantar region of the foot that is not in contact with the floor, and therefore it is not influenced by GRF.

The foot loading pattern was evaluated through a piezo-dynamometric platform, which is widely described elsewhere (16). Briefly, the piezo-dynamometric platform was obtained by rigidly fixing a dedicated pressure platform (designed and constructed at the Biomedical Engineering Laboratory of the Italian National Institute of Health, Rome, Italy) onto a commercial force platform (Bertec, Worthington, OH) in order to allow the resultant GRF to be transmitted unaltered from one platform to the other. The resulting compound instrument simultaneously measures, for each sample, the resultant GRF (vertical and horizontal components, free moment, and location of center of pressure) and pressure distribution throughout the foot-floor contact area. By applying an ad hoc mathematical model, it also estimates the local GRF components acting on elementary areas. Thus, each subarea of interest, of any shape and dimension, can be wholly characterized from a dynamic point of view.

The platform was inserted at a level in the middle of a wooden walkway that was long enough to guarantee acquisition “at regimen” (17–19). Patients were trained to walk barefoot on the walkway at their natural speed, and to step on the active surface of the platform with one foot only, without looking down at the platform. Six consecutive footsteps were acquired for each foot, and mean values were entered for statistical analysis.

Metatarsal joint motion was assessed by means of a general-purpose, long-arm mechanical goniometer. The total range of motion at the first metatarso-phalangeal joint was measured as follows: with the patient supine and with the subtalar joint in neutral position, the maximal ranges of dorsal flexion and plantar flexion were determined from the tip of the toe to the heel along the medial side of the foot. For the sake of accuracy, the range of motion was recorded for both feet of each patient by the same operator.

Statistical analysis

One-way ANOVA was performed on the raw data of each group to obtain comparisons between couples of variables in multiple groups. For post hoc multiple comparisons of the means, a Bonferroni test was used for equal variance. Linear regressions were computed between plantar fascia thickness and forces, plantar fascia thickness and LJM, and forces and LJM.

RESULTS

There were no significant differences between groups for sex, age, BMI, or diabetes duration. Mean values are reported in Table 1. Patients and control subjects were homogeneous in terms of their occupation.

Diabetic patients showed increased thickness of plantar fascia throughout the length of the structure. Measurements taken next at the calcaneous insertion showed a statistically significant increase in all patient groups in comparison with control subjects, as shown in Fig. 1A) (2.9 ± 1.2, 3.0 ± 0.8, 3.1 ± 1.0, and 2.0 ± 0.5 mm for the D, DN, DNPU, and C groups, respectively; P < 0.05).

The evaluation of the joint mobility at the first metatarso-phalangeal joint showed a reduction of the range of motion in all diabetic groups (54.0 ± 29.4, 54.9 ± 17.2, 46.8 ± 20.7, and 100.0 ± 10.0° for the D, DN, DNPU, and C groups, respectively; P < 0.05) (Fig. 1B). The reduction was mainly due to dorsiflexion (26.1 ± 12.4, 26.9 ± 12.2, 22.8 ± 10.5, and 64.2 ± 6.4°, respectively; P < 0.05).

The study of foot-floor interaction under the metatarsal heads showed increased vertical forces in DN and DNPU patients and increased medio-lateral forces in the DNPU group (Fig. 1C).

Table 2 reports mean values and standard deviations of the integrals of vertical, anterior-posterior, and medio-lateral components of GRF, expressed as percent of body weight (%N) × ms.

An inverse correlation (r = −0.53) between the thickness of plantar fascia and LJM was found when considering the three pathological groups together, as well as a direct correlation between the thickness of plantar fascia and vertical forces under the metatarsal heads (r = 0.52). No significant correlation was found between LJM and forces. Figure 2 shows the regression line and equation for the relationship between plantar fascia thickness and LJM (Fig. 2A) and for the relationship between plantar fascia thickness and forces (Fig. 2B).

CONCLUSIONS

In diabetic patients, peripheral neuropathy and, more specifically, motor neuropathy are responsible for remarkable deformities of the anatomical structure of the foot (e.g., clawtoes, pes cavus, prominent metatarsal heads, etc.) (8) and functional alterations of gait. These structural and functional changes have been recognized as responsible for increased plantar pressures (7,20) .

In a previous study, we observed that significant changes in forefoot GRF components may be present in diabetic patients without neuropathy (21), thus suggesting that other modifications, independent from peripheral neuropathy, may occur.

The present study shows evidence of a significant increase in plantar fascia thickness, which may entail critical changes in the functionality of the foot (22). Plantar fascia, in fact, sustains the longitudinal arch of the foot during propulsion and assists the body in absorbing forces at the midtarsal joints during the heel strike and the load transfer from the rearfoot to the forefoot (23). When the foot is in contact with the ground, a downward force on the talus is counteracted by an upward force on the metatarsal heads, toes, and the calcaneus. The upward force produces moments that tend to flatten the arch. The moments are countered by the horizontal forces tending to tie the forefoot and calcaneus together. These forces are generated in passive structures (plantar fascia) (23) by the active contraction of intrinsic and extrinsic muscles (24,25).

The increased thickness of the plantar fascia renders its section greater than normal, which increases its resistance to stresses in the direction of its longitudinal axis (horizontal stresses). Thus, greater upward forces should be produced at the metatarsal heads and at the calcaneus to normally flatten the foot.

In diabetic patients, hyperglycemia induces several effects in almost all tissues because of the glycosilation of proteins (26). Thus, the abnormal production of collagen renders the plantar fascia more resistant to tensile force (27,28). The stiffness of the foot is also increased by the greater tensile force exerted by the Achilles tendon, which means greater concentric horizontal forces in the plantar fascia. However, not all of the increased forefoot pressure may be related to a shortened Achilles tendon, since its surgical lengthening does not seem to completely overcome the forefoot hyperpressure.

The locomotor pattern is altered by the reduction of mobility at the metatarso-phalangeal joint. This is partly caused by the abnormalities of the joint connective tissues (protein glycosilation), and it is partly the result of the increased tension in the plantar fascia. Because plantar fascia plays a role in foot digital stabilization (29), its intrinsic greater tensile force is confirmed by the positive correlation we found between increased plantar fascia thickness and the limited metatarso-phalangeal range of motion. The final scenario is that of a rigid cavus foot, which remains rigid during the whole walking cycle, thus facilitating the onset of high plantar pressures. In conclusion, we have shown that diabetic patients do develop changes in plantar fascia and joint mobility, even in the absence of peripheral neuropathy. The relationship between plantar fascia abnormalities and the increased vertical forces under the metatarsal heads of the forefoot suggests that these patients already have a different pattern of plantar pressure distribution (19,21). Therefore, the above alterations may have a role in a more rigid foot, which is thus less adaptable to the floor during the foot-floor interaction. It is then reasonable to hypothesize that these changes may be critical in the onset of the mechanical stress that induces plantar hyperpressure.