Forefoot Deformity in Diabetic Neuropathic Individuals and its Role in Pressure Distribution and Gait

Abstract: Background. Foot deformities have been related to diabetic neuropathy progression but their influence on plantar distribution during dynamic tasks is not completely understood. The purpose of the present study was to investigate the influence of metatarsal head prominence and claw toes on regional plantar pressures during gait in patients with diabetic neuropathy. Methods. Seventy-one adults participated in this study categorized into three groups: a control group (CG, n = 32), patients with diabetic neuropathy without any foot deformities (DG, n = 20), and patients with diabetic neuropathy with metatarsal head prominence and/or claw toes (DMHG, n = 19). Plantar pressure variables (contact area, peak pressure, and maximum mean pressure) were evaluated during gait on rearfoot, midfoot, and forefoot using capacitive insoles (Pedar-X System, Novel Inc., Munich, Germany). A general linear model was applied to repeatedly measure and analyze variance relationships between groups and areas. Results. DMHG presented larger contact areas at the forefoot and midfoot along with higher peak pressure at the rearfoot compared to the other two groups. The DG showed higher mean pressure at the midfoot compared to the other two groups. Conclusion. The coexistence of diabetic neuropathy and metatarsal head prominence in addition to claw toes resulted in overloading the rearfoot and enhancing the contact area of forefoot and midfoot while walking. This plantar pressure distribution is a result of a different coordination pattern adopted in order to reduce plantar loads at the anterior parts of the foot that were structurally altered. Patients with diabetic neuropathy without any forefoot deformities presented a different plantar pressure distribution than patients with deformities suggesting that both neuropathy and structural foot alterations can influence foot rollover mechanisms.
Address correspondence to: Isabel C.N. Sacco, PhD Laboratório de Biomecânica do Movimento e Postura Humana, Depto. Fisioterapia, Fonoaudiologia e Terapia Ocupacional Universidade de São Paulo Rua Cipotânea 51, Butantã São Paulo, Brasil 05360-160 Email: icnsacco@usp.br

     A plantar ulcer in patients with diabetic neuropathy is a determinant factor of a poor quality of life, has a significant economic impact on health systems, and is a recurrent topic in scientific investigation. An increased level of plantar pressure has been related to plantar ulcer formation in neuropathic diabetic subjects. However, the factors contributing to this increase in pressure remain a discussion in specialized literature. A number of factors have been identified as possible causes of plantar ulceration in patients with diabetic neuropathy including loss of protective sensation,¹ repetitive stress² due to the influence of mechanical factors such as muscle length, decreased range of motion,^3–5 callus,⁶ foot pad atrophy,⁷ and foot deformities.⁸ It is not known which of the proposed mechanisms for increasing plantar pressures is most significant in leading to plantar ulceration in patients with neuropathy. Some studies have suggested that the most relevant factor of ulcer pathogenesis is the sensibility reduction.^9,10 Nurse and Nigg¹¹ induced reduction of sensorial foot feedback using an ice exposure technique in 10 healthy individuals and observed that the pressure distribution pattern was different in healthy subjects who increased the pressure values in regions where the sensibility was preserved. Nevertheless, another study which induced a reduction of plantar cutaneous sensation using an anesthetic solution administered to 10 healthy subjects did not observe plantar pressure alteration during gait activities.¹² A consensus does not exist regarding how sensibility deficits alter plantar pressure distribution.      Diabetic neuropathy associated with the presence of foot deformities has been an important question discussed throughout the literature as well. There is evidence that patients with diabetic neuropathy and claw/hammer toes have an increased predisposition to foot ulcers, as individuals have demonstrated a decreasing thickness in submetatarsal heads and subphalangeal fat pad cushions, and that these atrophies interfere in the load attenuation capability of the local tissue, increasing the plantar pressure at the sites of the deformities.^13,14 Some authors have suggested that structural foot abnormality is an important predictor of plantar ulceration in diabetic neuropathic subjects.^15,16 In a recent literature review, the evidence of a higher risk of ulcer formation in diabetic neuropathic subjects with orthopedic foot deformities was confirmed, and the trigger factor for ulcer development was attributed to the hyperkeratinization under the foot deformities, although the authors did not discuss the relationship between these structural alterations and plantar pressure distribution changes during gait.¹⁷      The question that motivated the present study was, “What is the influence of sensorial deficit in peripheral neuropathy and forefoot deformities in plantar pressure distribution during gait in patients with diabetes and healthy subjects?” Thus, the aim of this study was to investigate the influence of prominent metatarsal heads and claw toes in regional plantar pressures during gait in patients with diabetic neuropathy.

Methods

     Subjects. This cross-sectional study involved 71 male and female adult volunteers divided into three groups: a diabetic neuropathic group (DG) composed of 20 patients with diabetic neuropathy diagnosed by physicians; a diabetic neuropathic group with metatarsal head prominences and/or claw toes (DMHG) composed of 19 patients classified by a physiotherapist; and a control group (CG) composed of 32 healthy subjects (Table 1). The control group recruitment attempted to match the diabetic groups’ anthropometric characteristics. All subjects were informed about the procedures and signed a written term of free and informed consent that was approved by the local ethics committee (protocol No. 1054/04).      Inclusion criteria for the patients with diabetic neuropathy were: type 2 diabetes with more than 5 years since onset; a score higher than 4 on the Michigan Neuropathy Screening Instrument questionnaire (MNSI-q¹⁸; MNSI median = 7); and a minimum of two areas in tactile perception testing where the subject did not recognize or feel the touch of a 10-g monofilament.^19–21 The MNSI-q is a validated instrument for screening symptoms related to diabetic neuropathy. The arch index²² was also used in order to exclude major arch alterations (equinus planus and extra cavus feet) that could interfere in gait mechanics.      Subjects were excluded if they were older than 65 years; presented with hallux amputation or partial amputation of the foot, except toes; had orthopedic disorders of the lower limbs; had pain during the data acquisition; used any assistive devices for walking such as walking sticks or canes; had Charcot arthropathy confirmed by radiography; or presented with plantar ulcers at the time of the evaluation.      Gait measurement. Subjects walked on a 10-m walkway at a self-selected speed while wearing only Pedar-X insoles (Novel Inc., Munich, Germany) inside anti-skid socks, as described in the literature.²⁷ The insoles were 2.5-mm thick and contained a matrix of 99 capacitive pressure sensors with a spatial resolution of 1.6 cm2–2.2 cm². Before data acquisition, the subjects were instructed to walk freely in the laboratory to reproduce their normal gait and to adapt to the laboratory environment. A digital metronome was used to limit gait cadence to a range of 96 to 112 steps per minute in order to avoid large differences in gait cadence among subjects. A total of 20 steps were recorded for analysis at a sample rate of 100 Hz.      The subjects were barefoot during gait analysis to avoid the influence of other factors on plantar pressure, such as the subject’s own shoes, and because the goal was to study the complex behavior of foot-to-floor interaction^23,24 without interference. Although the EMED system is conventionally used to evaluate barefoot plantar pressure because of its spatial resolution, the plantar areas evaluated were wide enough not to be influenced by the Pedar-X system spatial resolution. The Pedar-X system has the advantage of acquiring multiple steps and does not require a subject to alter his gait²⁵ in order to make contact with any surface—this pressure data averaged from multiple trials is far more reliable than single trial acquisition.¹⁴      Contact area, peak pressure, and maximum mean pressure were evaluated in three plantar areas: rearfoot, midfoot, and forefoot. Following the scheme established by Cavanagh and Ulbrecht,²⁶ the rearfoot corresponded to 30% of foot length, the midfoot corresponded to 30% of foot length, and the forefoot (which included the metatarsal heads and the toes) corresponded to 40% of foot length. These areas were defined using software from Novel Inc. that divides the foot print into pre-determined areas proportional to the insole’s length and width.      The acquisition, processing and data analysis were performed with Novel Inc. system software (Novel Multiprojects).

Statistical Analysis

     A mean for each plantar pressure variable was calculated across steps and trials excluding the first and the last step of each trial per subject. After confirming the normal distribution and the homoscedasticity of all variables by using the Shapiro-Wilk test and Levene test, respectively, groups and plantar areas were compared using a general linear model for repeated measures analysis of variance (ANOVA, [2 x 3] α = 0.05).

Results

     A significant effect on groups and plantar areas was found in the contact area (F = 2.967; P = 0.020), peak pressure (F = 3.17; P = 0.014), and mean pressure (F = 5.039; P < 0.001; [Table 2]). The DMHG presented larger midfoot and forefoot areas in comparison to CG and DG (P < 0.01), higher mean pressure at rearfoot in comparison to CG (P = 0.046) and DG (P = 0.000), and higher peak pressure at the rearfoot compared to CG (P <0.001). DG presented higher peak pressure at the midfoot in comparison to CG (P = 0.001), and higher mean pressure at the midfoot in comparison to CG and DMHG (P < 0.01).

Discussion

     The purpose of this study was to investigate the influence of two common forefoot alterations of patients with diabetic neuropathy through plantar pressure analysis during gait. The main findings showed that prominent metatarsal heads and claw toe deformities in patients with neuropathy lead to a broader contact area over more anterior parts of the foot (midfoot and forefoot) and to higher loads at the rearfoot area. These results suggested that patients with diabetic neuropathy and metatarsal head deformities adopt a changed coordinated motor pattern while walking to reduce plantar loads at the anterior parts of the foot that are structurally altered. These anterior areas of the plantar surface present more risk of ulcer formation than other areas.^20,25,27      Larger anterior contact area, higher posterior loads, and lesser loads at the metatarsal heads may be a result of an alteration of foot rollover mechanics from the heel rocker action at the initial contact to the propulsion by the forefoot rocker.²⁷ Patients with diabetic neuropathy may experience neuropathy pain during walking, and the metatarsal head prominences may worsen symptoms. The alteration of foot rollover mechanics in these patients may relieve such neuropathy pain and result in better plantar pressure distribution, particularly over structurally modified sites.²⁸ In a cross-sectional descriptive study with participation of 326 patients with type 2 diabetes, Davies et al29 report that the risk of developing painful neuropathy increases with the worsening and progression of illness. High incidence of symptoms (higher MNSI scores), years from diabetes onset, and the development of forefoot deformities contribute to neuropathy severity, and therefore, an increased chance of developing painful neuropathy. Patients with severe diabetic neuropathy seemed to adopt a different coordination pattern to protect the involved foot areas.      In contrast with the present findings, Bus and deLange¹⁴ observed higher peak pressures over the metatarsal heads in 14 patients with diabetic neuropathy and deformities in the anterior parts of the foot. The difference between Bus and deLange results and the present study may suggest that, depending on severity, patients with diabetic neuropathy could adopt a new foot rollover pattern (eg, the reduction of forefoot pressure observed in the present study), or a worsening in pressure levels as a result of diminished contact area and metatarsal head exposure.¹⁴      According to Van Schie’s review,²⁷ the expected foot rollover is composed of three stages: heel-strike, mid-stance (full forefoot loading and heel lift), and propulsion, which is the phase in which the greatest horizontal and vertical forces are directed against the foot, and weight bearing is over a relatively small area in the forefoot. The DMHG may have pushed off (the final propulsion phase) before the metatarsal heads completely touched the ground saving the anterior foot regions from overloading and pain, and this could be the reason why the midfoot contact was altered in this group. Nevertheless, this motor pattern of foot to floor interaction causes foot rollover alteration as described by Giacomozzi et al.³⁰      An insufficient recruitment of the lower limb musculature may occur and thereby contribute to the alterations in plantar pressure distribution observed during gait in these individuals. Some authors have discussed that extensor muscles, such as gastrocnemius and vastus lateralis, showed an activity deficit in patients with diabetic neuropathy,^31–33 and this may be related to propulsion phase alteration.      Thus, the adoption of a different rollover motor pattern in subjects with claw or hammer toes and/or metatarsal head prominence seems to be true. The present study did not observe the different stance phases (heelstrike, flatfoot, and propulsion) but the entire stance, which is a limiting factor in deepening the discussion in this direction.      In contrast to the results observed of the DMHG and CG groups, the DG group presented high pressures over the midfoot suggesting that individuals with diabetic neuropathy without foot deformities may show higher loads at this particular plantar region than those with foot deformities, as Sacco et al34 observed that high midfoot pressure could be due to altered foot rollover mechanics in their prospective study of 31 subjects divided into healthy and diabetic neuropathic groups. Patients with diabetic neuropathy without foot deformities who do not present higher anterior contact area and rearfoot overloads, corroborates the theory that the DMHG plantar pressure distribution could be associated with a protection strategy developed because of the presence of deformities and their consequences to osteomioarticular foot structure.      The present study does not take into account the sub-phases of stance, which could have limited the understanding of foot rollover patterns. Further research is needed to measure plantar pressure and lower limb electromyography in heel strike, flatfoot, and propulsion phases of stance in patients with diabetic neuropathy and orthopedic foot deformities. Such research would help with the understanding of the foot rollover biomechanics of this population and help develop therapeutic strategies to aid in the recovery of mechanical integrity.

Conclusion

     Individuals with diabetic neuropathy and forefoot deformities present an increased midfoot and forefoot contact area and higher rearfoot loads. These alterations may be related to the adoption of a changed coordinated motor pattern aiming for the reduction of overloading and pain at the anterior parts of the foot that are already structurally modified by the deformity. Although the existence of this gait pattern is not confirmed by the present data, the results of this study suggest that these subjects present an altered foot rollover mechanism that may be aggravated by the existence of prominent metatarsal heads and claw toes. Diabetic neuropathic subjects without these deformities do not present the same gait pattern. By understanding the foot mechanics of these individuals there is a possibility of offering treatment to aid in the effective and efficient performance of functional activities, and perhaps diminish pressure during the course of daily living activities.

Acknowledgement

     This study was funded by FAPESP (04/09585-2).