The Biomechanics of Diabetic Foot Amputation

According to the International Diabetes Federation, in 2019 approximately 463 million adults aged 20 to 79 years were living with diabetes mellitus (DM).¹ This number is projected to increase to 700 million by 2045. Multiple pathophysiological factors, including diabetic peripheral neuropathy (DPN), foot deformity, trauma, and peripheral vascular disease, contribute to the genesis of diabetic foot ulcers (DFUs).^2,3 Importantly, 85% of all lower limb amputations in patients with diabetes are preceded by DFU.^4,5 The purpose of this review is to examine specific biomechanical and quality of life (QoL) factors to help identify the most functional levels for lower extremity amputation for compromised patients with DFUs. 

Both peak plantar pressure (PPP), which is the vertical ground reaction force (GRF) divided by the surface area, and shear force, represented by medial-lateral and anterior-posterior GRF vectors, are critical to the biomechanical discussion of DFU formation and lower extremity amputation. Most studies, however, focus only on plantar pressure, partly due to the absence of validated shear stress sensing devices.^6-8 Results from various studies highlight the need to take shear into account when assessing the risk for DFU. There is a weak correlation between peak vertical pressure measurements and ulcer location, with one study reporting that only 38% of ulcers occurred in the same location as peak vertical pressure.⁹ Yavuz et al¹⁰ reported that shear forces can occur in a location different from that of the PPPs; distances between locations of more than 2.5 cm were reported in 60% of patients with DM in their study. Further research highlighted the correlation between shear forces and DFU formation.^6,10-12 Vertical GRFs were higher in patients with DPN than in healthy control subjects. However, both vertical and medial-lateral GRFs were demonstrated in patients with DM who had a history of DFU.^6,10-12 It may be that plantar pressure and shear forces are important factors not only in DFU formation, but also in the identification of the most functional levels of lower extremity amputation that best preserve patient QoL. 

Controversy persists concerning whether partial foot amputation (PFA) or major lower extremity amputation provides better patient QoL. Intuitively, it would seem as though preserving as much of the limb as possible would maximize quality of ambulation and gait. Previous studies have noted that advantages of PFA include improved function, cosmesis, body image, lifestyle, sensory input, and weight-bearing surface as well as decreased risk of contralateral foot amputation and energy consumption.¹³ Few biomechanical studies have been performed, though more recent studies indicate that PFA may not be the best treatment. 

For example, historically it was commonly believed that because PFA preserves the ankle joint, the patient is left with a more normal gait and thus experiences reduced energy expenditure compared with major amputation. Per a 2013 study by Dillon and Fatone,¹⁴ however, neither belief is accurate. Preserving the ankle joint did not permit a better push-off or a more normal gait; power generation at the ankle joint was found to be negligible.¹⁴ Furthermore, studies have shown that PFA increases PPP by decreasing the area of the support surface upon which the limb experiences vertical GRF.^6,15,16 Additionally, findings that PFA results in reduced energy expenditure compared with transtibial amputation^17-19 are based on aerobic studies that use oxygen as an operational measure of energy expenditure.^20,21 That basis of determining energy expenditure is not always the most appropriate or specific when trying to decide on whether a PFA or below-knee amputation (BKA) is the best choice for a given patient. Many other factors may need to be considered, such as patient acceptance, balance, and post amputation physical demands. 

The assumption that PFA results in improved QoL has not been confirmed; current methods used to measure QoL (eg, a scoring measure) do not always measure the appropriate parameters for the patient who underwent an amputation.^14,22 Studies have often used non-validated items, such as the “ability to go to the bathroom in the middle of the night,” to show preference for a PFA.^14,23,24 Although a patient and their physician may opt to preserve as much of the foot as possible, it has not been demonstrated that doing so provides a functional benefit.^14,23,25

Finally, it may be better to consider a particular patient’s wound healing potential and chance for recurrence as primary criteria for determining amputation level as opposed to a purely maximum functional level, that may actually favor transtibial amputation over many of the PFAs.¹⁴ This recommendation is echoed by the results of a 2016 study that found QoL need not be a consideration when choosing between PFA and transtibial amputation.²⁶ For patients with vascular disease for whom the relative risk of ulceration and subsequent amputation is high, age, time with diabetes, and presence of retinopathy were more important factors than QoL when determining the level of amputation.  

Perhaps the greatest concern with PFA is the substantial complication rate and need for secondary amputation. If healing of a PFA is quite unlikely, it may be more efficacious to perform a more proximal procedure instead. Several studies cite that 30% to 50% of patients who undergo a PFA will experience complications. This includes skin breakdown, re-ulceration, and wound failure.^{14,17,19,24,27,28} Only 50% of PFAs eventually heal regardless of level, and 15% to 45% of patients who undergo a PFA will require some type of secondary amputation, two-thirds of whom will undergo a higher amputation on the same limb; 15% to 30% of those will die within 12 months.^14,29 

Hallux amputation

Hallux amputation, which may be indicated to manage local infection, osteomyelitis, and focal gangrene, may be an option for the patient who wants to preserve appearance and maximum foot function. However, it is important that both surgeon and patient are aware that most hallux and first ray amputations typically require more proximal amputations later.³⁰ One study reported significantly higher re-amputation rates in patients with radiographic evidence of hallux rigidus (P = .023).³¹ 

A specific concern with hallux amputation is that as gait shifts to one that is less propulsive, a low gear or lateral push-off, the lesser toes contract and drift medially (adductovarus) for improved grip. This can result in transfer metatarsalgia, hammer toe deformity, and distal or dorsal ulcerations that lead to further amputation.^13,31,32 Preserving the base of the proximal phalanx can both prevent drift and preserve sesamoid attachments, such as flexor hallucis brevis and abductor hallucis attachments, to aid in hallux stabilization.¹³ Using an in-shoe pressure measurement system, Motawea et al³³ found that PPPs and pressure-time integral were significantly increased under the first metatarsal head and lesser toes after unilateral hallux amputation compared with those values for the contralateral intact foot. Those increased values for the affected foot placed patients at increased risk for re-ulceration or re-amputation.  

Lesser digital amputation

Lesser digital amputation is the most common type of PFA but the least studied. In patients with diabetes and neuropathy, ulcers often develop at the distal tip of the digit because of gripping associated with biomechanical imbalance.³⁴ As mentioned before with hallux amputations, it is recommended that the surgeon leave a portion of the base of the proximal phalanx to prevent lateral drift with lesser digital amputation. It also prevents hallux valgus deformity if the second digit is the one amputated.³⁵ Compensation for lesser digital loss leads to an apropulsive gait, a shift toward the hallux for balance and push-off, and increased pressure on the metatarsal heads. Prior to amputation, ulcers tend to develop under the first and fifth metatarsal heads in the more rigid, supinated foot with a low gear push-off (anterolateral imbalance), whereas ulcers are more likely to develop under the first and second metatarsal heads in the pronated foot.³⁶ 

Ray resection

Ray resections are usually performed to address an isolated osteomyelitis postoperative dehiscence or ischemic focal gangrene. When considering a ray resection (especially of a central ray), it is important to perform sufficient bony resection to achieve secondary closure. Outcomes tend to be better with primary amputation of the first or fifth ray, whereas central ray resection (especially of the second ray) can narrow the foot, resulting in significant imbalance and ulceration. Therefore, in many cases midfoot amputations are preferable to multiple ray resections.³⁷ 

Foot type biomechanics determine the outcome of first ray resection. In a patient with a rigid cavus foot with lateral imbalance, loss of the medial column removes a major medial force resisting pronation of the subtalar joint. A rigid cavus foot cannot compensate for increased pronation, and excessive pressures develop laterally, resulting in ulceration. Conversely, with first ray resection of a flexible foot, loss of the medial column results in a shift of all forefoot GRF lateral to the subtalar joint axis, thereby unlocking the midtarsal joint that leads to an increase in forefoot abduction and midfoot collapse. As a result of this pronation, digital contractures and subsequent ulceration can occur as a result of flexor stabilization in which flexor tendons fire earlier in the gait cycle and contract longer in an attempt to resupinate the subtalar joint and stabilize the midtarsal.^38,39 Borg et al⁴⁰ demonstrated that patients with DPN who had undergone first ray amputation experienced significantly increased PPP and pressure-time integrals under the second to fourth metatarsophalangeal joints (P = .008). As a result, the authors of that study recommend routine plantar pressure screening in these patients. 

Lateral ray resection of the fourth and fifth rays results in increased pressure at the fifth metatarsal base during midstance. Because the foot maximally pronates in an effort to get as much of the foot on the ground as possible, an orthosis is designed to balance the rearfoot and accommodate for overload at the fifth metatarsal/cuboid articulation. A toe filler where the fifth ray had been can be used to improve stability. If the amputation disrupts peroneal tendon function, overpowering by the posterior tibial tendon may occur and a tendon balancing procedure may be necessary. In almost all forefoot amputations, tendo-Achilles lengthening (TAL) should be considered. 

Midfoot amputation 

Transmetatarsal amputation. Transmetatarsal amputation (TMA) results in a reasonable platform for propulsion.³⁷ The TMA can preserve the tibialis anterior tendon as well as the peroneal tendons provided care is taken to not amputate too far proximal. Creech et al⁴¹ recommend leaving a 2-cm buttress of the first metatarsal to preserve both the tibialis anterior tendon and the deep plantar perforating artery. Preservation of the artery is important to preserve arterial communication between the dorsal and plantar foot. Equinus secondary to an overpowering of the gastro-soleus muscles is the most important complication of a TMA and must be addressed for it to succeed. Tendo-Achilles lengthening should be performed to prevent overload of the distal amputation and re-ulceration of the residual foot. The 2019 International Working Group on the Diabetic Foot guideline recommends that, for the person with diabetes and a neuropathic plantar metatarsal head ulcer, clinicians should consider performing TAL to promote healing.⁴² In a study of patients who had not undergone amputation and who presented with neuropathic plantar ulcers, Mueller et al⁴³ found that although forefoot PPPs were decreased after TAL combined with immobilization with total contact casting, these values regressed to baseline within 7 months after treatment. In contrast to the patient cohort treated with total contact cast immobilization alone, those treated with both TAL and cast immobilization had a lower rate of early ulcer recurrence. A TAL-induced reduction of ankle plantar flexion forces also benefits patients with a TMA who often develop a secondary equinovarus deformity caused equinus plus an overpowering of the posterior tibial muscle and associated increased lateral shear forces.^44,45 Additional investigation of the long-term effects of prophylactic TAL on forefoot PPP reduction leading to ulceration, as well as on the benefits of routinely doing the release when performing a TMA, is required. 

Tarsometatarsal (Lisfranc) amputation. Partial foot amputation performed proximal to a TMA is done to provide additional soft tissue coverage; however, this procedure alone does not result in a biomechanically stable foot. Rather, adjunctive tendon rebalancing procedures are required to compensate for partial or total loss of the transverse arch.⁴⁶ Tarsometatarsal (Lisfranc) amputation results in a severe varus deformity if the peroneal tendons cannot be preserved or transferred. A tibialis anterior tendon transfer may be an option to address development of an equinovarus deformity. 

Chopart amputation. For Chopart amputations in particular, TAL is recommended to address unopposed ankle plantar flexion.⁴⁶ In addition, use of a rocker soled shoe and an in-shoe filler or prosthesis is necessary to provide protection for the residual foot and prevent ulceration for the Lisfranc, Chopart, or Symes amputation.⁴⁶ Ankle fusion is an alternative for patients who are either unable to use or uninterested in using a prosthesis for the remainder of their lives but fusions in this population are risky due to the complications that led to the amputation in the first place.³⁷ Although modern ankle-foot orthoses have demonstrated improved biomechanics, custom-molded silicone prostheses have been found to be a somewhat functional and cosmetic option. In a study of 4 patients who had undergone Chopart amputation, Burger et al⁴⁷ demonstrated that use of a silicone prosthesis (even without a reinforced sole) resulted in more normal pelvic movement and less trunk inclination toward the amputated side compared with use of footwear with a conventional prosthesis. Greater power generation at the ankle during stance/push-off, and improved barefoot walking also were noted with the silicone prosthesis alone. 

Midfoot amputation has a significant effect on biomechanical function. Dillon and Barker⁴⁸ found that when compared with a control group of patients who had not undergone amputation, patients who underwent midfoot amputation experienced major mechanical adaptations after the metatarsal heads were compromised. When using devices such as slipper sockets and insoles, the treated patients adopted a gait pattern that limited the distal excursion of the center of pressure. Those authors surmised that this adaption likely spares the partial foot from extreme force and moderates the forceful plantar flexion of the triceps surae or the demand for compliance of the residual foot. Of note, after midfoot amputation the hip joint is used as a primary source of power, and there is negligible power generation at the affected ankle joint.⁴⁸ 

Hindfoot amputation 

Hindfoot amputation is generally not recommended as it leaves the plantar flexor muscles unopposed, resulting in severe, permanent equinus if surgical rebalancing is not done. If a patient is unable to use a lower limb prosthesis, an adjunct ankle joint arthrodesis can be performed to address the equinus deformity. It is important to note that hindfoot amputation with concurrent ankle fusion is not a functional level of amputation, and gait will be apropulsive postoperatively.³⁷ 

Symes amputation. Symes amputation consists of transections of the foot through the ankle joint, with removal of the tibial and fibular malleoli and preservation of the posterior tibial artery and the plantar fat pad. The limb is suitable for transfers and standing, but this is not an ambulatory amputation and the patient will require a specialized prosthesis. Although this procedure is not common, if successful it results in relatively few late complications.³⁶ The Symes prosthesis is a heavy exoskeletal prosthesis that houses the bulbous distal limb using a window for entry and closure of the window for suspension. The addition of a prosthetic foot to a Symes prosthesis results in substantial elevation of the prosthesis; thus, it is necessary to add a sole raise to the contralateral shoe. 

Pirogoff amputation. The Pirogoff amputation is made through the ankle articulation and includes removal of ankle cartilage as well as preservation of a portion of the calcaneus for a weight-bearing surface.⁴⁹ Drawbacks of this procedure include the requirement for bone fusion, wound healing problems, and resulting substantial disability as well as a residual limb that requires a prosthesis similar to that used with a Symes, and it is difficult to procure footwear. 

Transtibial and transfemoral amputation

Major amputations are performed if the foot is considered unsalvageable because of severity of infection, amount of soft tissue loss, or severity of peripheral vascular disease, among other considerations. In a study comparing transtibial (below-knee) amputation and transfemoral (above-knee) amputation (AKA), bipedal ambulation was achieved by 80% of patients treated with BKA compared with 38% to 50% of patients treated with AKA.⁵⁰ Furthermore, the additional energy expenditure increased only 10% to 40% in patients treated with BKA; in contrast, energy expenditure increased 60% in patients treated with AKA. An even higher energy expenditure was reported with hip and knee disarticulations. The prosthesis or artificial limb is crucial for restoration of appearance and activities of daily living in patients who have undergone transtibial amputation. A poor fit at the residual limb-socket interface may result in increased mean peak pressure over the patellar tendon during ambulation, pistoning within the socket, skin ulceration, and in some cases re-amputation.^51,52 Shear stresses are equally as important as direct pressures, owing to potential reduction of skin blood flow and subsequent lesion formation, which highlights the importance of developing accurate shear stress sensors.^52,53

When determining the most functional level of amputation for the patient with a high-risk DFU, several factors must be taken into consideration. From preoperative examination to postoperative care, both patient and surgeon have the same goal—a functional limb that is clear of infection. When comparing PFA with a major lower extremity amputation (ie, BKA or AKA), it is vital to consider the goals of the patient and properly set patient expectations. Although PFA may seem like a less drastic decision from a cosmetic or psychological standpoint, the potential exists for significant functional drawbacks related to relative gait performance, energy expenditure, wound healing potential, and re-ulceration rate. Patients must be informed that additional future amputation may be necessary after PFA. In addition, validated QoL measures are needed to help inform decision-making in determining the correct amputation level. When PFA is being considered, a specific localizing factor (eg, extent of infection) may be the primary driver for selecting the level of amputation. The patient must be informed of the specific potential post-amputation sequelae. For midfoot amputation in particular, TMA with TAL likely achieves the best functional outcome. Additional research is needed to identify accurate and useful biomechanical measurements beyond PPPs that will assist in the prevention of DFUs, the identification of the most functional amputation levels and associated offloading/prosthetic devices, and the long-term monitoring of patients who have undergone amputation. In 2018 Tavares et al⁷ reported on their development of an insole optical fiber sensor that dynamically discriminates between shear and plantar pressure, thereby allowing continuous remote gait monitoring of patients as an ehealth solution. New technology such as this could provide a wealth of new data to advance the study of the biomechanics of the diabetic foot following amputation.

Authors: James McGuire, DPM, PT, LPed, FAPWHc; Avery Thomson, DPM, MS, BA; and Pamela Genise Kennedy, DPM, BS

Affiliation: Temple University School of Podiatric Medicine, Philadelphia, PA

Correspondence: James McGuire, DPM, PT, LPed, FAPWHc, Professor Clinician Scholar, Department of Medicine, Temple University, Leonard Abrams Center for Advanced Wound Healing, 148 N. 8^th Street, Philadelphia, PA 19107; jimnanmcguire@gmail.com

Disclosure: The authors disclose no financial or other conflicts of interest.