Effectiveness and safety of elective surgical procedures to improve wound healing and reduce re-ulceration in diabetic patients with foot ulcers


Lawrence A. Lavery, Department of Plastic Surgery, University of Texas Southwestern Medical Center, 1801 Inwood Road, Dallas, TX, USA.

E-mail: lklavery@yahoo.com


The objective is to evaluate the effectiveness and safety of surgical off-loading to heal diabetic foot ulcers and prevent ulcer recurrence. Usually, structural foot deformities such as hallux rigidus, hammertoe deformities and equinus of the ankle contribute to abnormal pressure and shear forces and non-healing foot ulcers. Elective surgery to remove the deformity and restore joint mobility has been shown to be safe and effective to improve wound healing of recalcitrant ulcer and to reduce the risk of re-ulceration. Unfortunately, there is very little high-level evidence to help guide patient selection or to compare clinical outcomes. Copyright © 2012 John Wiley & Sons, Ltd.

Aetiology of foot ulcers

Ulceration is a major risk factor for lower extremity amputation. Foot ulcers precede lower extremity diabetic amputations more than 80% of the time [1]. So, healing ulcers and preventing re-ulceration should be a main focus in amputation prevention. The most frequent factors that contribute to ulceration are unrecognized repetitive injury, peripheral sensory neuropathy with loss of protective sensation, limited joint mobility and structural foot deformity [2]. Structural foot deformities such as hallux valgus and hallux rigidus are associated with ulceration of the great toe and first metatarsal. Equinus deformity, limited ankle joint dorsiflexion, is associated with ulcers on the ball of the foot; and hammertoe deformities are associated with ulcers on the tip of the toes, on the dorsum of the toes and between the toes.

Diabetic foot ulcers are difficult to heal, and once healed, there is a high rate of ulcer recurrence. Conventional measures to heal diabetic foot ulcers are often ineffective. Margolis et al. conducted a meta-analysis of the control arms of randomized clinical studies in which patients received ‘standard therapy’. Only 24% and 31% of wounds healed in 12 and 20 week of studies, respectively [3]. Ulcers that do not respond to standard therapies such as debridement, vascular assessment, infection control, pressure reduction and local wound care should be assessed for uncorrected pressure and shear forces at the ulcer site as a result of structural foot deformity or limited joint mobility.

Ulcer recurrence with standard care

After diabetic foot ulcers heal, patients experience a high rate of re-ulceration. The high rate of recurrence is often because deformity and limited joint mobility persist, despite conventional non-surgical measures to protect the foot from re-injury. Traditional therapies to prevent ulcer recurrence are effective, but even when they are provided, recurrence is high. Recurrence ranges from 30% to 87% [4-6]. Different methods to prevent re-ulceration have been proposed including shoes and insoles, education, regular foot care, home temperature monitoring and silicone injections to augment the fat pad on the sole of the foot. Therapeutic shoes and insoles have been shown to be effective in reducing ulcer recurrence, but the rate of re-ulceration is still unacceptably high. Chantelau reported a recurrence rate of 42% in patients wearing cushioned insoles as compared with 87% when patients wore their own shoe [4]. Dargis proposed the use of a multidisciplinary approach in managing patients with previous ulceration. He reported a recurrence rate of 30.4% in patients who received prevention therapies compared with 58.4% in patients who used self-selected shoes [5]. The ulcer recurrence rate is high because the underlying causal factors in the pathway to ulceration are usually inadequately treated with shoes and insoles and education. Without eliminating the underlying risk factors in the ulcer pathway, recurrence at the same site is very high.

Who is a candidate for elective surgery?

Surgical correction of the structural deformity is an option to heal ulcers and reduce the risk of recurrence [7-10]. However, the risk of surgery is a universal concern. The healing potential of persons with diabetes is difficult to assess. These patients have multisystem disease including poorly controlled diabetes, microvascular and macrovascular diseases, chronic kidney disease, nutrition and immunopathy. Patient compliance, tobacco use, mobility and social factors should be considered when selecting potential surgical candidates. Most of the literature in this area is descriptive and predominately composed of retrospective cohort studies. There is only one prospective randomized clinical study that evaluates an elective surgical procedure, Achilles tendon lengthening, compared with conventional ulcer therapy. Despite the higher risk associated with diabetes for poor healing and infection, elective surgery seems to be safe, to improve ulcer healing and perhaps most importantly, to reduce the risk of re-ulceration.

There are no clear criteria to determine wound healing in persons with diabetes.

Studies in this area do not have rigorous vascular, glucose, nutritional or laboratory inclusion or exclusion criteria [7-15]. Perfusion to the foot is obviously an essential parameter to evaluate wound healing in persons with diabetes; however, even in Mueller's prospective study comparing Achilles tendon lengthening and total contact casting to heal foot ulcers, the only vascular inclusion parameter was ‘a palpable ankle pulse’. Likewise, in Holstein's series of percutaneous Achilles tendon lengthening, he only evaluated ankle and toe pressures after peripheral pulses could not be palpated [14]. Other concerns for healing such as glycosylated haemoglobin were not cited as exclusion criteria in any of the published reports we identified. In fact, the average glycated haemoglobin in Mueller's study was 8.8%, and it was 8.4% in Armstrong's study [10]. Better criteria to select surgical candidates are needed, but currently, there is little objective information to guide that selection process. Perhaps the best criterion to determine if a patient will heal elective surgery is a history of healing a peripheral wound regardless of other established risk factors. At present, patient selection seems to rely on the ill-defined intuition of the surgeon.

Hammertoe correction

Resectional arthropathy of the proximal interphalangeal joint and percutaneous tenotomy of the flexor digitorum longus are often used to repair flexor contracture deformities of the toes. These are simple procedures that are usually performed under a local foot block. Kearny et al. [7] reported results from a case series of 58 percutaneous flexor tenotomies to heal recalcitrant or recurrent ulcers on the tip of toes in persons with diabetes and flexor contractor. The patients selected for this procedure have a clawed toe, so the patients are bearing weight on the tip of the toe. The deformity should be manually reducible. When there is a bony deformity and the toe is not reducible, a soft tissue release as described by Kearney will probably not be effective. Fifty-seven ulcers (98.3%) healed in an average of 40 days. One patient had an amputation of the toe because of pre-existing osteomyelitis of the distal phalanx. Recurrent ulcerations were identified in 12.1% of toes that required surgery in an average of 13.9 months after surgery. Post-operative infection occurred in 5.2% of surgeries. All infections were resolved with oral antibiotics without hospitalization.

Rosenblum described a retrospective series of diabetic patients with recalcitrant great toe ulcers treated with resectional arthroplasty of the great toe [9]. After surgery, all wounds healed. None of the patients required amputation. There was a low rate of post-operative re-ulceration (5%), and 9% of patients were treated for infection. Armstrong reported on a similar series that focused on arthroplasty of the base of the proximal phalanx of the great toe, often referred to as Keller arthroplasty, to heal great toe ulcers. Armstrong compared a cohort of patients with diabetes treated with surgery and a cohort treated with conventional off-loading [8]. The time to healing was faster in the surgery group (24.2 vs. 67.0 days) and re-ulceration was lower (4.8% vs. 35.0%) than in the treatment group. Both groups had similar rates of infection (standard 38.1% versus surgery 40.0%). The risks of resectional arthroplasty of the proximal phalanx of the great toe are developing metatarsalgia of the lesser metatarsal heads and hammertoe deformity of the great toe. And although these complications were not identified by Armstrong, there is a risk of transferring the ulcer across the ball of the foot.

Elective toe surgery in persons with diabetes and foot ulcers seems risky at face value. We do not have good techniques to measure perfusion to the digits, and the toes are the most vulnerable area for ‘diabetic microvascular disease’. There is no high-level evidence with randomized control trials to support the effectiveness of elective toe surgery, to compare adverse events or to evaluate long-term risk reduction of recurrent ulcers or amputations. However, the retrospective studies in patients who have failed standard therapies support the notion that the surgery is safe and effective and the risk of recurrence is low.

Achilles tendon lengthening

The Achilles tendon is deleteriously affected by long-term diabetes. Grant [11] identified structural differences between persons with and without diabetes. He found that tendon fibril packing density was significantly altered because of the non-enzymatic cross linking of fibrillar collagen. This creates a functionally shorter tendon that contributes to equines. Equinus pulls the heel into plantarflexion and increases forefoot pressures [15]. Armstrong and Mueller found that by lengthening the Achilles tendon, forefoot pressures were reduced by 27% [12, 15]. Other investigators have associated the lengthening of the Achilles tendon as an adjunctive procedure in diabetic foot wound healing [13-15].

There are several studies that report the effectiveness of percutaneous Achilles tendon lengthening to heal foot ulcers. The only randomized controlled trial to compare elective surgery to heal foot ulcers was reported by Mueller from Washington University. Mueller et al. conducted a randomized clinical trial to compare the effectiveness of lengthening the Achilles tendon with total contact casting to heal ulcerations [15]. The average ankle joint range of motion was 15° after surgery and 1° of dorsiflexion in patient treated with a cast. Among patients who had surgery, 100% healed compared with 88% in the total contact cast group. Recurrent ulceration was lower in the treatment group. After 2 years, 38% of patients in the Achilles tendon lengthening treatment group developed another foot ulcer compared with an 81% re-ulceration rate in the total contact cast group [15]. The average time to heal was 41 days in the group treated in total contact casts and 58 days in the group treated with surgery; however, 13% of patients with Achilles tendon lengthening developed heel ulcers. The proportion of wounds that healed, rate of healing, complications and the low rate of ulcer recurrence were similar in retrospective cohort studies [13, 14].

Holstein reported the results of a retrospective cohort of 68 patients with 75 recalcitrant foot ulcers that were treated with Achilles tendon lengthening [14]. The average duration of ulcer was 48 months. All of the ulcers had failed to heal despite off-loading for at least 3 months. After surgery, 91% of ulcers healed in an average of 6 weeks. However, there was a high rate of complications after surgery. Like Mueller, 14% of patients developed non-healing heel ulcers that were associated with overcorrection of ankle equinus (dorsiflexion > 15°). In addition, 10% of patients experienced complete rupture of the Achilles tendon, and 4% developed Charcot neuro-osteoarthropathy after surgery. Holstein felt that complications were more common in patients with severe neuropathy of the heel.

Lin and colleagues reported similar experiences with Achilles tendon lengthening in recalcitrant foot ulcers [13]. Lin performed Achilles tendon lengthening on 15 patients who failed to heal after treatment in a total contact cast. He compared the time to healing and ulcer recurrence with 21 patients who healed after casting. Both groups healed ulcers in an average of about 6 weeks. Of the surgery group, 93% healed, and all subjects treated with casts healed. The most striking difference, as in other studies, was the low rate of ulcer recurrence. After 17 months, 19% of patients treated in a total contact cast re-ulcerated. None of the patients who had their Achilles tendon lengthened re-ulcerated.

Achilles tendon lengthening is an excellent adjunct to treat patients with equinus that have not healed with aggressive off-loading as demonstrated by Lin and Holstein [13, 14]; however, it is technically demanding, and objective patient selection criteria have not been developed. Holstein provides insights into his clinical outcomes to refine the selection process and the surgical goals. He had a very high rate of heel ulcers (47%) with 14% developing recalcitrant, non-healing heel ulcers. Holstein identified overcorrection of the ankle equinus (>15°) and severe sensory neuropathy of the heel as key risk factors associated with developing heel ulcers. There is a balance between surgical correction and healing forefoot ulcers while avoiding transfer ulceration to the heel. Correction of less than 10° seems to be effective to heal forefoot ulcers with less risk of heel ulcers.


Elective surgery for recalcitrant foot ulcers is effective. Descriptive studies suggest a high rate of healing, a low rate of complications and a low rate of re-ulceration, especially in comparison with the natural history of foot ulcers and the high rate of ulcer recurrence when patients receive standard prevention therapies. Selection criterion to determine the best candidates for elective surgery is not defined in any meaningful, objective way. The current literature uses surgery when an ulcer has failed to heal with aggressive conservative care or when there is recurrence despite standard preventative therapies. There is clearly no evidence that elective surgery will reduce the risk of ‘the first ulceration’.

Many patients who are candidates for elective surgery have failed all other treatment approaches. The reports of complications after surgery should give us pause to reassess the lessons we have learned but not to discard this approach entirely. Previous experience should help us gather better information, refine patient selection criterion, improve surgical procedures and objectives and re-evaluate post-operative care.

Conflict of interest

None declared.