Medially Based Hallux Interphalangeal Joint Arthroplasty in the Management of Hallux Ulceration: Surgical Technique and Result in Six Consecutive Cases

Increased plantar foot pressures and shear forces have a significant role in the formation of the neuropathic diabetic foot wound.¹ Custom insoles and footwear have demonstrated ability to offload plantar foot pressure, but efficacy tends to be related to compliance and appropriate use.¹ Surgical intervention has become a well-tolerated and efficacious means of achieving offloading for neuropathic foot ulcerations.²

In 2008, Bevilacqua et al² classified the various types of diabetic foot surgery. Class III procedures were defined as those that have a curative effect for diabetic foot ulcerations. Various techniques have been described to achieve surgical offloading for the management of recalcitrant wounds.^2-17 Two Cochrane reviews support the healing potential of surgical intervention for the diabetic foot.^4,5 One study evaluated 6 randomized controlled trials to determine effective treatment measures.⁴ They found that there was significant reduction in lower extremity amputations, and thus increased cost-effectiveness. Other studies found limited benefit of providing patients written instructions and education-centered intervention; such intervention also proved to be less cost-effective than complex interventions.⁴ In 2013, Lewis and Lipp⁵ published a review on the effects of different pressure-relieving interventions in the management of diabetic foot ulcerations. Interventions assessed included nonremovable casts, removable pressure-relieving devices, Achilles tendon lengthening combined with use of a nonremovable cast, surgical debridement, foam dressing, and felt fitted into shoes. One study comparing removable devices with nonremovable casts reported significantly better outcomes in the group treated with nonremovable casts. Two studies showed significantly better outcomes in patients treated with Achilles tendon lengthening and a nonremovable cast than in those treated with only a removable cast.⁵ It should be noted that surgical patients require less compliance; nonsurgical patients are required to wear offloading shoe gear and avoid walking barefoot, and they may have missed visits.³ In addition, nonsurgical patients typically are more involved in the management of their lesions.³ These factors should be taken into consideration when evaluating the success of the postoperative course.

In 2015, Lew et al¹² described a curvilinear dorsal approach for performing a hallux IPJ arthroplasty to offload plantar ulcerations. This technique requires retracting or transecting the EHL tendon and resecting the distal portion of the cartilaginous surface of the proximal phalanx. That case-control series compared 13 surgical patients with 13 nonsurgical patients, all with neuropathic diabetic ulceration of the plantar hallux. A higher incidence of complete healing was noted in the surgical group compared with the offloading, nonsurgical group (3.5 weeks vs. 9 weeks; P =.033). The ulcer recurrence rate was much higher in the nonsurgical group than in the surgical group (54% and 8%, respectively; P =.31). The data reported in that study demonstrate the efficacy of hallux IPJ arthroplasty, including the short- and long-term effects of this procedure on plantar hallux wound healing, as well as the low rate of recurrence.¹² Although the dorsal curvilinear incision used in this technique is effective for offloading and healing plantar ulceration, it carries the inherent risk of surgical wound healing complications.

The current study presents a modification of the technique originally described by Lew et al¹² and evaluates the results of a simplified medially based hallux IPJ arthroplasty for offloading hallux ulcerations. The modified technique spares the EHL tendon, and therefore allows for less complicated wound closure.

The records of 5 consecutive patients (1 female, 4 male; 6 wound cases) who underwent medially based hallux IPJ arthroplasty for treatment of nonhealing plantar hallux ulceration were reviewed. One patient had bilateral wounds. The research compliance department at the authors’ institution considered this project a case series, and thus exempt from full institutional review board review.

Data were collected from electronic medical records. Patient demographics, medical comorbidities, hemoglobin A_1C level, initial wound location and size, final wound measurements, total time to wound closure, wound recurrence, time to final follow-up, and evidence of osteomyelitis (ie, bone biopsy culture or histologic findings) were included in the review. All patients had a history of recurrent plantar hallux IPJ ulceration that had healed initially using standard offloading procedures. All patients demonstrated normal osseous integrity on preoperative plain radiographs. All patients demonstrated normal pedal vascularity on clinical examination and had a history of normal noninvasive vascular testing. Each patient underwent a medially based IPJ arthroplasty as described by the senior authors (D.J.E. and M.M.R.) in the Surgical Technique section of this article. All procedures were performed in an outpatient ambulatory surgery setting. Descriptive statistics were performed using Excel (Microsoft).

Surgical technique

All patients were treated in the operating room under monitored sedation. However, the medially based hallux IPJ arthroplasty approach can be performed in the office at the discretion of the surgeon.

The patient is positioned supine, with the surgical extremity elevated for fluoroscopic imaging. The extremity is prepped to the ankle region, where a pneumatic tourniquet may be used based on surgeon preference and/or patient needs. A standard hallux block is performed. The instrumentation used in the procedure is shown in Figure 1.

The planned incision is drawn on the medial aspect of the hallux IPJ. The incision is approximately 2 cm to 3 cm long and is made at the junction of the plantar and dorsal skin (Figure 2). Using a no. 10 blade, a full-thickness incision is performed down to the level of the bone. Subperiosteal and capsular dissection is performed dorsally and plantarly on the proximal phalanx with a freer elevator. It is critically important to release all periosteal tissue from the IPJ in order to easily remove all resected bone. Dissection of the lateral subcapsular structures can be performed with a small McGlamry elevator (Figure 3). A powered, side-cutting Lindemann burr or Shannon burr is used to perform the arthroplasty (Figure 4). Mini C-arm fluoroscopy is used to confirm arthroplasty orientation (Figure 5). After orientation is confirmed, the surgeon inserts the burr from the medial cortex, through the medullary bone, and into the lateral cortex of the proximal phalanx. This technique is analogous to using a drill bit. Once through both cortices, the surgeon will “window-wash” to cut the dorsal and plantar cortices. A high-torque, low-speed power setting is preferred to enhance osseous resection and minimize soft tissue entrapment. After complete transection, the loose head of the proximal phalanx is easily removed through the incision (Figure 6), and the arthroplasty is confirmed by fluoroscopy (Figure 7). The wound is lavaged with copious normal sterile saline, and the void is then carefully backfilled with small vancomycin-impregnated calcium sulfate beads (Figure 8). The soft tissue envelope is then closed in a full-thickness fashion using 2-0 or 3-0 nonabsorbable monofilament suture. A combination of vertical mattress and simple interrupted suturing technique is used. The extremity is then cleansed and dried. Bacitracin ointment, nonadherent gauze, and a cohesive bandage are used to dress the surgical site.

Full weight-bearing is allowed as tolerated in a protective surgical shoe or controlled ankle motion boot. Sutures are left in place for 2 to 4 weeks, depending on individual healing status. For diagnosis of bone infection, the resected bone is sent for microbiologic and pathologic analysis. Figures 9 and 10 illustrate the clinical progression of a patient who underwent this technique.

The mean follow-up period was 83.17 weeks ± 14.9 standard deviation (range, 54–95 weeks). The mean age at the time of the procedure was 66.5 years ± 12.5 (range, 43–68 years). Four of the 5 patients had diabetic neuropathy. The average glycated hemoglobin level was 7.82% ± 3.6 (range, 5.4%–14%). Two of the 5 patients had chronic kidney disease. Vascular status was intact, as indicated by palpable pedal pulses in all 5 patients, and all patients had a history of standard noninvasive vascular testing (ankle-brachial index and pulse volume recording). The mean preoperative plantar hallux ulceration surface area was 1.6 cm × 1.5 cm. The mean time to total wound closure was 15.5 days ± 4 (range, 10–22 days). Two patients had histologic and/or microbiologic evidence of osteomyelitis. No patient had any evidence of ulcer recurrence at the final follow-up visit. The results of this series are summarized in the Table.

This case series suggests that medially based hallux IPJ arthroplasty is an effective option for internally offloading plantar hallux wounds. This technique has no direct imbalance of hallucal tendon structures because the medial approach preserves the flexor hallucis longus, flexor hallucis brevis, and EHL tendons. Shortening of the hallux may occur; however, this technique preserves periarticular tendon balance and hallux malleus has not been observed.

As demonstrated in the Surgical Technique section of this article, antibiotic-impregnated beads can be used as an adjunct to treat residual infection. However, use of such beads is beneficial only if thorough osseous resection has been performed and the soft tissue envelope is adequate for wound closure after placement of antibiotic beads in the wound. Vancomycin, tobramycin, and gentamicin have demonstrated efficacy.¹⁵ In the current study, vancomycin was used because it was the most easily available antibiotic powder at the authors’ facility. Qin et al¹⁵ sought to determine whether antibiotic-impregnated calcium sulfate with bone resection was superior to bone resection alone. Forty-eight limbs with diabetic foot osteomyelitis were treated with either bone resection alone (28 limbs) or bone resection with calcium sulfate impregnated with vancomycin or gentamicin (20 limbs). Postoperative care, including wound care, offloading, and systemic antibiotics, was consistent between both groups. Healing and recurrence rates as well as amputation rates were compared between the 2 study groups. The study authors concluded that use of antibiotic-impregnated calcium sulfate along with bone resection does decrease the recurrence of diabetic foot osteomyelitis but does not improve the healing rate or the amputation rate. No severe side effects were noted. Prolonged postoperative leakage was the main complication in the study.¹⁵ The authors of the current study have found that routine patient follow-up is important to avoid this complication. Use of absorptive dressings such as calcium alginate has been effective in mitigating wound maceration.

The medially based hallux IPJ arthroplasty approach is a practical technique to internally offload plantar hallux ulcers. Dorsal and curvilinear approaches have also been described.^6,12 The dorsal approach invades the EHL tendon and usually requires direct repair of the resected tendon. This often requires use of a temporary Kirschner wire to protect the tendon repair.¹² First metatarsophalangeal joint arthroplasty (ie, Keller procedure) is another effective technique, although this procedure sacrifices the flexor hallucis brevis tendon attachment and frequently leads to hallux malleus deformity.^7,8,16,17 The medial approach does not have these consequences. Drawbacks to the medially based approach include the need for power instrumentation and intraoperative fluoroscopy. Although the authors of the current study believe there is a reasonable learning curve with this approach, surgeons should be aware of the need for increased instrumentation and possibly surgical time due to the relatively unorthodox approach to the IPJ. It is also important to note that the cohort of patients in the current study demonstrated normal vascularity upon presentation. The authors of this study cannot comment on the effectiveness of this procedure in a patient with clinical evidence of peripheral vascular disease and would recommend a thorough investigation and referral as appropriate prior to surgical intervention.

This report does have limitations. The number of wound cases is relatively small and does not include any control group to make a direct comparison to previously published techniques. Standard offloading modalities or inclusion of a dorsal arthroplasty approach would provide the ability to compare outcome measures directly.

The medially based hallux IPJ arthroplasty approach can provide a stable, plantigrade, wound-free, and infection-free digit, while providing full ambulatory use of the surgical extremity. This technique internally offloads plantar hallux wounds while avoiding intrinsic soft tissue structures, thereby providing long-term offloading and helping maintain a wound-free forefoot.