The Journal of Foot & Ankle Surgery 54 (2015) 541–548
Contents lists available at ScienceDirect
The Journal of Foot & Ankle Surgery journal homepage: www.jfas.org
Evaluation of Hallux Interphalangeal Joint Arthroplasty Compared With Nonoperative Treatment of Recalcitrant Hallux Ulceration Eric Lew, DPM, AACFAS 1, Nicole Nicolosi, DPM 1, Patrick McKee, DPM, FACFAS 2 1 2
Resident, Podiatric Medicine and Surgery, Healthspan/Cleveland Clinic, Cleveland, OH Staff, Orthopaedic and Rheumatologic Institute, Cleveland Clinic, Cleveland, OH
a r t i c l e i n f o
a b s t r a c t
Level of Clinical Evidence: 3
Patients with chronic diabetes can develop plantar hallux ulcerations secondary to neuropathy, increased pressure, and deformity. The present retrospective study evaluated the efficacy of hallux interphalangeal joint (HIPJ) arthroplasty to address recalcitrant ulceration. Two groups of patients with diabetes were compared: a surgical group of 13 patients and a nonsurgical standard therapy group of 13 patients. The patients in the surgical group underwent HIPJ arthroplasty. All the patients in the standard therapy group received local wound care and offloading. The mean duration of follow-up was 19.5 (range 1.2 to 47.9) months, and the mean age was 55 13.0 years. Statistical significance was found in the surgical group for faster time to healing (3.5 weeks [2.5, 4.25] vs 9 weeks [2, 17.29], p ¼ .033) and lower incidence of ulcer recurrence (8% 7.69 vs 54% 53.85, p ¼ .031). There were also fewer amputations in the surgical group (0% 0 vs 38% 38.6, p ¼ .063). To our knowledge, only 1 other published study has evaluated HIPJ arthroplasty as a treatment of recalcitrant hallux ulceration. The present study adds comparison data from a nonoperative standard therapy group and found that HIPJ arthroplasty is an effective curative treatment option to address chronic plantar hallux ulcerations in diabetic patients with neuropathy. Ó 2015 by the American College of Foot and Ankle Surgeons. All rights reserved.
Keywords: diabetes hallux wound limb salvage neuropathy offloading toe preservation
Ulcerations are a frequent antecedent of elective lower extremity amputations in patients with diabetes-related peripheral neuropathy. Hallux ulcerations often develop secondary to increased plantar pressures caused by a combination of nonreducible pedal deformities, limited joint mobility, and neuropathy (1,2). Traditionally, neuropathic ulcerations are treated nonoperatively, using a combination of offloading techniques and local wound care. Offloading techniques are imperative in the reduction of pressure required for healing and in neuropathic ulcer prevention after healing. Despite external offloading, such as custom molded insoles and shoes, ulcer recurrence frequently develops secondary to structural deformity. Ulceration recurrence has ranged from 30% to 87% (2). Operative intervention is considered when ulcers have been unresponsive to conservative treatment (1–6). We present a retrospective study evaluating the efficacy of hallux interphalangeal joint (HIPJ) arthroplasty as a treatment of recalcitrant hallux diabetic neuropathic foot ulcer. Indications, technical pearls, and case examples are also presented. We hypothesized that patients who had undergone HIPJ arthroplasty
Financial Disclosure: None reported. Conflict of Interest: None reported. Address correspondence to: Eric Lew, DPM, AACFAS, Podiatric Medicine and Surgery, HealthSpan/Cleveland Clinic, 9500 Euclid Avenue, NA-40, Cleveland, OH 44195. E-mail address: [email protected] (E. Lew).
Fig. 1. A chronic plantar hallux interphalangeal joint ulceration in a 53-year-old male with type 2 diabetes mellitus and peripheral neuropathy.
1067-2516/$ - see front matter Ó 2015 by the American College of Foot and Ankle Surgeons. All rights reserved. http://dx.doi.org/10.1053/j.jfas.2014.08.014
542
E. Lew et al. / The Journal of Foot & Ankle Surgery 54 (2015) 541–548
Fig. 2. A dorsal incision was placed, overlying the hallux interphalangeal joint.
Fig. 3. Dissection exposed the extensor hallucis longus tendon, which was either retracted or transected for joint visualization and later repaired.
would yield favorable outcomes in the immediate and long-term periods compared with standard care consisting of local wound care and offloading alone. To assess this hypothesis, we reviewed the data from a series of patients who had undergone either HIPJ arthroplasty or standard, nonoperative therapy for the treatment of recalcitrant neuropathic HIPJ ulceration.
or plantar-medial hallux lesions. Patients with diabetic neuropathy or idiopathic peripheral neuropathy and plantar hallux ulcerations, of any size or depth, that had been present for longer than 6 weeks were included for review. Patients were excluded if the duration of the ulcer was less than 6 weeks or if the ulceration was not located on the weightbearing surface of the hallux. Patients with active soft tissue or osseous infection were included. Patients with incomplete documentation and those who had been lost to follow-up were excluded. The patients who underwent HIPJ arthroplasty were identified using the aforementioned electronic medical record database. A search using Current Procedural Terminology code 28160 (Current Procedural Terminology, American Medical Association, Chicago, IL) was performed on the list of subjects with hallux ulcerations after applying the exclusion criteria. The patients who underwent the procedure were included in the HIPJ arthroplasty surgical group. Patients treated nonoperatively were placed into the control group. Selection of this group was made on the basis of propensity score matching with the goal of achieving similarity on age, diabetes, peripheral neuropathy, body mass index, and ulcer volume. Matching was performed using the R package Matching. The patients
Patients and Methods The patients were treated within the Podiatric Section of the Department of Orthopaedics from January 2008 to May 2013 (5 years, 3 months) at the Cleveland Clinic (Cleveland, OH). The present investigation was conducted with institutional review board approval. Patients with chronic plantar or plantar-medial ulcerations were identified using an electronic medical record database. Specifically, an institutionally designed Smartform filter in the Epic Electronic Health Records system (Epic, Verona, WI) was applied to generate a list of patients with a history of plantar
Fig. 4. Partial proximal phalangeal head resection. (A) A sagittal saw blade was used to resect the cartilaginous surface of the proximal phalanx. (B) A minimal amount of cartilage and bone was removed to preserve the hallux length.
E. Lew et al. / The Journal of Foot & Ankle Surgery 54 (2015) 541–548
543
Fig. 5. Hallux interphalangeal joint arthroplasty. (A and B) A reciprocating rasp or rongeur was used to contour any prominent bone on the resected surface of the proximal phalanx and the plantar medial condyle of the distal phalanx.
in the surgical and conservative treatment groups were treated in a standardized fashion consistent with diabetic foot wound care at the Cleveland Clinic. Treatment was based on 6 principles of wound healing: 1. A thorough evaluation of lower extremity perfusion status with noninvasive vascular studies and referrals to a vascular specialist when necessary 2. Eradication of infectious processes 3. Use of proper off-loading techniques
Fig. 6. An anteroposterior radiograph demonstrating fixation with a 0.062-in. Kirschner wire, which was used for percutaneous stabilization of the hallux in some cases in which a larger amount of bone resection was necessary.
4. Ensuring adequate wound moisture balance, along with appropriate wound debridement 5. Edema control 6. Encouraging adequate nutrition and optimizing overall medical condition with internal medicine or endocrinology consultations We explored the number of weeks until healing after the initiation of treatment and evaluated variables for an association with nonhealing. Other outcomes evaluated in both treatment groups included recurrence of ulceration and amputation. The demographic data for age, sex, and body mass index were recorded. The hallux ulceration duration before the initiation of treatment and the size before treatment initiation was documented in weeks and cubic centimeters, respectively. The ankle brachial index, toe brachial index, and incidence of revascularization procedures were also recorded. The
Fig. 7. The skin was closed with nonabsorbable suture in a simple interrupted fashion.
544
E. Lew et al. / The Journal of Foot & Ankle Surgery 54 (2015) 541–548
Fig. 8. Evaluation of hallux interphalangeal joint range of motion. (A) The preoperative range of motion was assessed. (B) An increased range of motion was observed intraoperatively after hallux interphalangeal joint arthroplasty. other comorbidities included for review were diabetes mellitus, chronic kidney disease, peripheral arterial disease (PAD), hemoglobin A1c, and tobacco use. Postoperative complications, including infection, dehiscence, neuritis, new ulceration development, and deep venous thrombosis, were evaluated in the surgical group only. Data collection and a review of the electronic medical records was conducted by 2 investigators (E.L. and N.N.). The HIPJ arthroplasty surgical procedure was performed by 1 podiatric surgeon (P.M.). Nonoperative management of the hallux ulcerations in the standard therapy group was performed by members of the Podiatric Section of the Department of Orthopaedics at the Cleveland Clinic. The outcomes were determined from a review of the medical records. The results of the comparative data were interpreted by 2 investigators (E.L. and N.N.), who were aware of the treatment groups because, although they had participated in the surgical treatment of 2 of the patients in the case series, they had not participated in the nonoperative patient care. The Quantitative Health Sciences Department at the Cleveland Clinic computed the statistics. Procedure The patients in the surgical group underwent HIPJ arthroplasty (Fig. 1) using a dorsal approach over the HIPJ (Fig. 2). Dissection was carried through the subcutaneous tissue, and the extensor hallucis longus was identified and retracted (Fig. 3). However, if necessary for exposure, the extensor hallucis longus was transected or lengthened and later repaired. The HIPJ was opened longitudinally, and the cartilaginous surface of the head of the proximal phalanx was removed using a saw blade (Fig. 4). A reciprocating rasp or rongeur was used to smooth and contour prominent bone, including the resected surface of the proximal phalanx and the plantar medial condyle of the distal phalanx (Fig. 5). In some cases, fixation with a 0.062-in. Kirschner wire was placed
percutaneously to stabilize the hallux during the healing period. The decision to pin the joint was made intraoperatively according to the amount of bone resection and stability of the hallux (Fig. 6). No fixation was applied in the presence of any suspected chronic infection or deep ulcer contamination. The subcutaneous layer was re-approximated with absorbable suture in subdermal inverted interrupted fashion while the skin layer was closed with nonabsorbable suture in simple interrupted fashion (Fig. 7). The range of motion of the hallux interphalangeal and metatarsophalangeal joints was assessed intraoperatively to determine whether it was increased (Fig. 8). The postoperative course consisted of protected weightbearing in forefoot offloading devices such as pressure relief shoes with modifiable inserts or forefoot relief shoes. At the 2-week postoperative visit, the skin sutures were typically removed. All the patients in the standard therapy group underwent offloading with either total contact casting, custom molded inserts, forefoot relief shoes, or drill-and-filled modified shoes (Fig. 9). The drill-and-fill technique involves removal of the dense rubber outsole that underlies the ulcer and replacing with softer material. The patients in both groups received appropriate wound debridement and local wound care throughout the treatment periods. Preventative treatment after ulcer healing was maintained in both groups with frequent monitoring and the use of standard offloading inserts and footgear. Shoe gear consisted of prescriptive depth inlay shoes, which included sufficient room in the toebox to accommodate the insert. Statistical Analysis Continuous variables are presented as the mean standard deviation or median and interquartile range. Both ordinal and nominal categorical variables are described using counts and percentages. Differences in continuous variables between the surgical
Fig. 9. Patients in both the conservative and the surgical groups underwent appropriate wound care and ambulated with various forms of forefoot offloading, including total contact casting, custom molded inserts, forefoot relief shoes, or custom shoes and inserts, with a drill and fill technique. (A) The ulceration location was identified on the insole. (B) The corresponding area on the shoe was drilled. (C) This drill-and-fill technique offloads any weightbearing pressure off the ulcerations.
E. Lew et al. / The Journal of Foot & Ankle Surgery 54 (2015) 541–548
545
Table 1 Comparison of baseline patient characteristics by treatment group (N ¼ 26 patients)
Table 2 Comparison of outcomes by treatment group (N ¼ 26 patients)
Risk Factor
Outcome Variable
Patient age (y) Gender Female Male Body mass index >25.0 kg/m2 Diabetes mellitus neuropathy Chronic kidney disease Peripheral arterial disease Hemoglobin A1c (>6.5%) History of tobacco use Osteomyelitis Ulcer volume (>1.0 cm3) Ulcer duration (>6 wk)
Treatment Group
p Value
HIPJ Arthroplasty (n ¼ 13)
Standard Therapy (n ¼ 13)
55.38 13.02
54.77 12.28
5 8 12 13 4 1 6 8 1 13 13
2 11 12 13 7 4 11 4 1 13 13
(92.3) (100) (30.77) (7.69) (46.15) (61.54) (7.69) (100) (100)
(92.3) (100) (53.85) (30.77) (84.6) (30.77) (7.69) (100) (100)
.87* .38y
NA NA .45y .13y .81z .25y > .99y > .99z .001z
Abbreviations: HIPJ, hallux interphalangeal joint; NA, not applicable. Data presented as mean standard deviation, n, or n (%). * Student’s paired t test. y Exact McNemar’s test. z Wilcoxon signed ranks test. and conservative treatment groups were tested using either Student’s paired t test or paired Wilcoxon signed ranks test. Differences in nominal categorical variables were tested using the exact McNemar test. Differences in ordinal categorical variables were tested using paired Wilcoxon signed ranks tests. The time-to-heal was analyzed using Cox proportional hazards models, with clusters defined by the matched group number, with the comparison between groups performed using the Wald test. The median healing times were estimated using the Kaplan-Meier method. All analyses were performed using R software, version 3.0.1 (R Project for Statistical Computing, Vienna, Austria). A 5% level of significance was used for all testing.
Results A total of 158 patients with chronic plantar hallux ulcerations were treated from January 2008 to May 2013 at the Podiatric Centers of the Cleveland Clinic (Cleveland, OH). After applying the exclusion criteria, 55 patients were included in our retrospective cohort study. Of the 55 patients, 13 (24%) underwent HIPJ arthroplasty and 42 (76%) underwent standard therapy. Thus, 13 age-matched patients were selected from the standard therapy group for comparison with the 13 patients in the surgery group. Therefore, 13 patients each were present in the surgical and standard therapy groups, and 26 patients were reviewed. A statistical comparison of the baseline risk factors for each treatment group is presented in Table 1. The overall mean patient age was 55 13.0 years, and 73% of the patients were male. The mean followup duration was 19.5 (range 1.2 to 47.9) months. The median time to healing was 9 [2, 17.29] weeks in the standard therapy group and 3.5 [2.5, 4.25] weeks in the surgical group (p ¼ .033). The incidence of recurrent hallux ulceration was 7.7% (1 of 13) in the surgical group and 53.9% (7 of 13) in the standard therapy group (p ¼ .041). The incidence of amputation was 0% (0 of 13) in the surgical group and 38.5% (5 of 13) in the standard therapy group (p ¼ .063). Comparisons of the healing time, ulcer recurrence, and amputations between the treatment groups are presented in Table 2. A total of 4 postoperative complications (30.8%) developed in the surgical group. These included incisional wound dehiscence in 2 (15.4%), infection in 1 (7.7%), and new (transfer) ulcer formation in 2 (15.4%). A new ulcer formed at the plantar aspect of the second toe in 1 patient, and another formed on the plantar aspect of the second metatarsophalangeal joint. Both of these transfer ulcerations healed with local supportive care. Case 1 A 50-year-old male presented with a left foot, plantar-medial hallux interphalangeal ulceration. He stated that the ulcer had been recurrent
Interval to heal (wk) Patients with recurrence Patients with amputation
Treatment Group
p Value
HIPJ Arthroplasty (n ¼ 13)
Standard Therapy (n ¼ 13)
3.5 (2, 17.29) 1 (7.69) 0 (0)
9 (4.25, 25) 7 (53.85) 5 (38.46)
.033* .031y .063y
Abbreviation: HIPJ, hallux interphalangeal joint. Data presented as median (minimum, maximum) or n (%). * Wald test from Cox proportional hazards regression. y Exact McNemar test.
for longer than 15 years. Previous treatment had included debridement, iodine cleansing, and the use of silver impregnated dressings. The patient’s medical history included type 2 diabetes mellitus with neuropathy, hypertension, and hyperlipidemia. The patient’s most recent hemoglobin A1c was 7.0 mg/dL, his body mass index was 38 kg/ m2, and he was a current smoker with a 30-pack-year history. On physical examination, the patient’s neurovascular status was intact, including the presence of protective touch-pressure sensation. A full-thickness ulcer was noted, localized to the plantar aspect of his left HIPJ and measuring 2.0 cm in length 2.0 cm in width 0.4 cm in depth. The ulceration did not probe to the bone, and no evidence of purulence, cellulitis, or signs of infection were noted. Radiographs revealed intact hallux phalangeal cortices without signs of lysis or proliferation. The patient spent a lot of time on his feet daily as a heating, ventilation, and air conditioning technician and was unable to take time off from work. Conservative treatment was attempted, consisting of weekly debridement of nonviable tissue and daily home dressing changes. The patient ambulated in new custom orthotics in regular shoes but refused to wear pressure relief shoes. Four weeks later, the patient’s course was complicated by forefoot cellulitis that resolved with oral antibiotics. The blood culture results were negative, no ascending lymphangitis was present, and no evidence of osseous lysis or proliferation was found on radiographic inspection. During the next several months, the ulceration waxed and waned in size, ranging from 0.5 cm 0.4 cm 0.5 cm to 2.0 cm 0.8 cm 0.3 cm. After approximately 9 months of conservative treatment, the wound measured 0.5 cm 0.5 cm 0.2 cm (Fig. 10). At that point, the patient was frustrated with the chronicity of the ulceration and opted to pursue surgical options. Subsequently, the patient underwent left HIPJ arthroplasty. The patient continued frequent monitoring, offloading, and local wound care postoperatively. The ulceration progressively decreased in size and had fully healed at 4 weeks postoperatively (Fig. 11), and the patient resumed his work duties and regular shoe gear. The patient had neither complications nor recurrence at 12 months of follow-up. Case 2 A 59-year-old male with a medical history of uncontrolled diabetes and peripheral neuropathy presented with a chronic and recurrent right hallux ulceration that had been present for more than 5 years. Other significant elements of the patient’s medical history included hypertension, chronic kidney disease, and hyperlipidemia. The patient’s hemoglobin A1c was 8.8 mg/dL, and his body mass index was 37.3 kg/m2. He was also a half-a-pack per day smoker. On physical examination, the patient’s neurovascular status was intact, including protective touch-pressure sensation. A full-thickness skin ulceration was located at the plantar aspect of the left HIPJ. It measured 1.5 cm 0.7 cm 0.5 cm. The ulceration probed to the bone through a sinus tract. Radiographs revealed cortical disruption and lytic changes at the medial condyle of the base of the distal phalanx
546
E. Lew et al. / The Journal of Foot & Ankle Surgery 54 (2015) 541–548
Fig. 10. A 50-year-old diabetic male with a chronic neuropathic plantar hallux ulceration underwent hallux interphalangeal joint arthroplasty.
Fig. 12. An anteroposterior radiograph demonstrating a periosteal reaction and cortical lysis along the base of the distal phalanx and medial aspect of the hallux interphalangeal joint.
(Fig. 12). A white blood cell–labeled bone scan was indicative of chronic osteomyelitis of the distal phalanx. Thereafter, the patient underwent operative debridement and resection of the affected HIPJ. The intraoperative bone culture results were positive for Streptococcus agalactiae. Bone pathologic examination also revealed findings consistent with chronic osteomyelitis. He received appropriate intravenous antibiotic therapy on consultation
with an infectious disease specialist. The patient was followed up on a weekly basis with local wound care and offloading. His postoperative course was complicated by dorsal wound dehiscence and cellulitis at 7 days postoperatively. The wound eventually closed by secondary intention after operative debridement, antibiotic therapy, and local wound care. At 6 weeks postoperatively, the plantar wound had completely healed. He was maintained with custom pressurerelieving insoles and prescriptive depth-inlay shoe gear. He had no ulcer recurrence through 5 years of follow-up (Fig. 13). Discussion
Fig. 11. The patient’s wound had completely healed at 4 weeks postoperatively.
The concept of internal offloading of chronic ulcerations has been described previously and applied to other areas of the foot (1–5). Several surgical treatments have been proposed for healing digital ulcerations, including metatarsal head resection, arthrodesis, and amputation (3). A study by Armstrong et al (4), in 2003, concluded that first proximal phalangeal base resection, or Keller osteotomy, resulted in a faster healing time and fewer repeat ulcerations than standard conservative therapy. A study by Stephens (1), in 2000, reported using flexor hallucis longus contracture release to reduce plantar pressure and heal chronic plantar hallux ulcerations. Kim et al (5), in 2008, revealed that removal of a lesser digital phalanx can result in ulceration healing. Quebedeaux et al (7), in 1996, concluded that hallux amputation contributes to lesser digital deformities and new ulcer formation. Lavery et al (8) postulated that the reason for this was because, after a hallux amputation, the pedal pressure distribution is altered and increases the reamputation risk. In lieu of hallux amputation, toe-preserving surgical intervention such as HIPJ arthroplasty could serve as an alternative to address chronic HIPJ ulcerations and decrease the risk of altered pedal pressure distribution, thus decreasing the risk of lesser digital amputation.
E. Lew et al. / The Journal of Foot & Ankle Surgery 54 (2015) 541–548
547
Fig. 13. (A and B) A 59-year-old male with a chronic plantar medial hallux ulceration and osteomyelitis underwent debridement and resection interphalangeal joint arthroplasty. The plantar wound was healed at 6 weeks postoperatively. No ulcer recurrences were reported at the final follow-up visit at 5 years.
To our knowledge, only 1 study has evaluated the efficacy of HIPJ arthroplasty for the treatment of recalcitrant hallux ulceration (6). In 1994, Rosenblum et al (6) reviewed 46 HIPJ arthroplasties in 40 patients treated for chronic neuropathic ulcer of the hallux. They found that 91% of the patients had healed with no evidence of recurrence in the followup period, which ranged from 4 to 44 months. They also concluded that HIPJ arthroplasty is a valuable procedure for chronic ulcerations. The results of the present study revealed that HIPJ arthroplasty is an effective treatment option to aid in the healing of chronic plantar hallux nonischemic ulcerations in diabetic patients with neuropathy. These results correlate with those from the study by Rosenblum et al (6) performed more than 20 years earlier. An osseous prominence of the HIPJ predisposes the plantar aspect of the hallux to ulcer formation, along with plantar contractures and restricted motion, often observed in neuropathic patients. Resection of this joint prominence internally relieves pressure on the soft tissue and increases the range of motion, thereby eliminating key factors in ulcer development and recurrence. In the surgical group, the ulcerations healed within a shorter period, with no amputations required postoperatively. All patients continued the use of custom inserts and footgear after ulcer resolution, which we believe played a key role in preventing ulcer recurrence. One ulcer recurrence was observed in the HIPJ arthroplasty group at 2 years postoperatively. Additional evaluation of this patient revealed comorbidities characteristic of advanced disease. In addition to a diagnosis of diabetes mellitus, chronic kidney disease, and PAD, the patient had a hemoglobin A1c greater than 10%, an ankle brachial index greater than 1.3, and an ulcer volume greater than 5 cm3. This patient also required the longest to heal postoperatively (12 weeks). Given these findings, we expect that patients with vasculopathy, longstanding disease, and uncontrolled blood glucose levels will have poorer outcomes. The complications in the surgical group included 2 cases (15.4%) of minor incisional wound dehiscence, including 1 patient with associated infection and cellulitis that healed uneventfully. No cases of neuritis or deep vein thrombosis occurred. Another observed outcome was ulcer development localized to the adjacent digits, which occurred in 2 of the patients (15.4%) in the HIPJ arthroplasty group. A new ulcer formed at the plantar aspect of the second toe in 1 patient. In yet another patient, an ulcer developed on the plantar aspect of the second metatarsophalangeal joint. Both of these transfer ulcers healed uneventfully without surgical intervention and did not result in chronic wounds or amputation after addressing the lesions with accommodative offloading inserts. It is believed that excessive resection of the proximal phalangeal head can render the hallux unstable and shortened. This could lead to the potential development of transfer lesions to the second ray. An unstable hallux can also result in a cosmetically
displeasing joint and the possible lack of hallux purchase. Neither of these complications was encountered in the present cohort of patients. In contrast, reduction of the hallux purchase with this procedure proved therapeutic and aided in pressure relief and the prevention of ulcer recurrence. The patients were generally satisfied with appearance and function of their great toe. Ultimately, the foot and ankle surgeon will need to determine how much bone resection will be necessary, taking into account the lesser digital length, to avoid transfer lesions, ulcer recurrence, hallux instability, and excessive shortening. Removing the least amount of bone as possible will usually result in a more relatively stable joint that does not necessitate pin fixation. Another potential complication is hallux extensus, which can be avoided by not overshortening the extensor halluces longus tendon after tenotomy. The relative contraindications to this procedure include moderate to severe PAD. Patients with advanced PAD have a greater risk of surgical wound ischemia, infection, and secondary amputation. There is also an associated risk with surgery in diabetics with multiple comorbidities and advanced multisystem disease. The decision to perform surgery on these patients must be carefully considered, balancing the risks and benefits of this elective procedure. Foot and ankle surgeons are strongly encouraged to obtain appropriate medical, infectious disease, and vascular surgical consultations to optimize their patients’ success for healing. Lavery et al (2) in 2012 stated that despite the greater 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 ulceration. Osteomyelitic infection is not an absolute contraindication. One patient in the surgical group had osteomyelitis of the HIPJ. After surgical resection and prolonged antibiotic therapy, this patient’s ulceration had healed uneventfully at 6 weeks postoperatively. This patient did experience dorsal incisional wound dehiscence and infection that resolved with operative debridement and wound care. The patient had had no recurrence at 5 years of follow-up. Additional fixation with a 0.062-in. Kirschner wire should be avoided in the presence of deeper ulceration or suspicion of infection to avoid potential seeding of any pathogen proximally. The present study had several limitations, including the nature of a retrospective analysis, which can make it difficult to procure meaningful, reliable data in some cases. Also, the number of patients in the surgical group was small, limiting stronger statistical analysis measures pertaining to time-to-event outcomes. There was also the possibility of selection bias related to patient compliance in the surgical group. More advanced, higher powered, prospective or randomized controlled trials are necessary to further evaluate this procedure as a curative measure for chronic plantar hallux wounds.
548
E. Lew et al. / The Journal of Foot & Ankle Surgery 54 (2015) 541–548
In conclusion, HIPJ arthroplasty for the treatment of recalcitrant hallux ulcerations appears to yield favorable outcomes in the immediate and long-term periods, including a shorter healing time compared with a nonoperative control group. This procedure is a viable option for foot and ankle surgeons in preventing hallux ulcer recurrence and amputations in patients with diabetes-related neuropathy. The results of the present investigation could be used in the development of future prospective cohort studies and randomized controlled trials focusing on the treatment of diabetic foot ulceration localized to the HIPJ. Acknowledgments We express our appreciation to Colin O’Rourke, MS (Quantitative Health Sciences Department), for conducting the statistical analysis, and Tammy Owings, DEng (Department of Biomedical Engineering), for her kind assistance with patient data extraction.
References 1. Stephens HM. The diabetic plantar hallux ulcer: a curative soft tissue procedure. Foot Ankle Int 21:954–955, 2000. 2. Lavery LA. Effectiveness and safety of elective surgical procedures to improve wound healing and reduce re-ulceration in diabetic patients with foot ulcers. Diabetes Metab Res Rev 28(suppl 1):60–63, 2012. 3. Resch S. Corrective surgery in diabetic foot deformity. Diabetes Metab Res Rev 20(suppl 1):S34–S36, 2004. 4. Armstrong DG, Lavery LA, Vasquez JR, Short B, Kimbriel HR, Nixon BP, Boulton AJ. Clinical efficacy of the first metatarsophalangeal joint arthroplasty as a curative procedure for hallux interphalangeal joint wounds in patients with diabetes. Diabetes Care 26:3284–3287, 2003. 5. Kim JY, Kim TW, Park YE, Lee YJ. Modified resection arthroplasty for infected non-healing ulcers with toe deformity in diabetic patients. Foot Ankle Int 29: 493–497, 2008. 6. Rosenblum BI, Giurini JM, Chrzan JS, Habershaw GM. Preventing loss of the great toe with the hallux interphalangeal joint arthroplasty. J Foot and Ankle Surg 33:557–560, 1994. 7. Quebedeaux TL, Lavery LA, Lavery DC. The development of foot deformities and ulcers after great toe amputation in diabetes. Diabetes Care 19:165–167, 1996. 8. Lavery LA, Lavery DC, Quebedeax-Farnham TL. Increased foot pressures after great toe amputation in diabetes. Diabetes Care 18:1460–1462, 1995.
Contents lists available at ScienceDirect
The Journal of Foot & Ankle Surgery journal homepage: www.jfas.org
Evaluation of Hallux Interphalangeal Joint Arthroplasty Compared With Nonoperative Treatment of Recalcitrant Hallux Ulceration Eric Lew, DPM, AACFAS 1, Nicole Nicolosi, DPM 1, Patrick McKee, DPM, FACFAS 2 1 2
Resident, Podiatric Medicine and Surgery, Healthspan/Cleveland Clinic, Cleveland, OH Staff, Orthopaedic and Rheumatologic Institute, Cleveland Clinic, Cleveland, OH
a r t i c l e i n f o
a b s t r a c t
Level of Clinical Evidence: 3
Patients with chronic diabetes can develop plantar hallux ulcerations secondary to neuropathy, increased pressure, and deformity. The present retrospective study evaluated the efficacy of hallux interphalangeal joint (HIPJ) arthroplasty to address recalcitrant ulceration. Two groups of patients with diabetes were compared: a surgical group of 13 patients and a nonsurgical standard therapy group of 13 patients. The patients in the surgical group underwent HIPJ arthroplasty. All the patients in the standard therapy group received local wound care and offloading. The mean duration of follow-up was 19.5 (range 1.2 to 47.9) months, and the mean age was 55 13.0 years. Statistical significance was found in the surgical group for faster time to healing (3.5 weeks [2.5, 4.25] vs 9 weeks [2, 17.29], p ¼ .033) and lower incidence of ulcer recurrence (8% 7.69 vs 54% 53.85, p ¼ .031). There were also fewer amputations in the surgical group (0% 0 vs 38% 38.6, p ¼ .063). To our knowledge, only 1 other published study has evaluated HIPJ arthroplasty as a treatment of recalcitrant hallux ulceration. The present study adds comparison data from a nonoperative standard therapy group and found that HIPJ arthroplasty is an effective curative treatment option to address chronic plantar hallux ulcerations in diabetic patients with neuropathy. Ó 2015 by the American College of Foot and Ankle Surgeons. All rights reserved.
Keywords: diabetes hallux wound limb salvage neuropathy offloading toe preservation
Ulcerations are a frequent antecedent of elective lower extremity amputations in patients with diabetes-related peripheral neuropathy. Hallux ulcerations often develop secondary to increased plantar pressures caused by a combination of nonreducible pedal deformities, limited joint mobility, and neuropathy (1,2). Traditionally, neuropathic ulcerations are treated nonoperatively, using a combination of offloading techniques and local wound care. Offloading techniques are imperative in the reduction of pressure required for healing and in neuropathic ulcer prevention after healing. Despite external offloading, such as custom molded insoles and shoes, ulcer recurrence frequently develops secondary to structural deformity. Ulceration recurrence has ranged from 30% to 87% (2). Operative intervention is considered when ulcers have been unresponsive to conservative treatment (1–6). We present a retrospective study evaluating the efficacy of hallux interphalangeal joint (HIPJ) arthroplasty as a treatment of recalcitrant hallux diabetic neuropathic foot ulcer. Indications, technical pearls, and case examples are also presented. We hypothesized that patients who had undergone HIPJ arthroplasty
Financial Disclosure: None reported. Conflict of Interest: None reported. Address correspondence to: Eric Lew, DPM, AACFAS, Podiatric Medicine and Surgery, HealthSpan/Cleveland Clinic, 9500 Euclid Avenue, NA-40, Cleveland, OH 44195. E-mail address: [email protected] (E. Lew).
Fig. 1. A chronic plantar hallux interphalangeal joint ulceration in a 53-year-old male with type 2 diabetes mellitus and peripheral neuropathy.
1067-2516/$ - see front matter Ó 2015 by the American College of Foot and Ankle Surgeons. All rights reserved. http://dx.doi.org/10.1053/j.jfas.2014.08.014
542
E. Lew et al. / The Journal of Foot & Ankle Surgery 54 (2015) 541–548
Fig. 2. A dorsal incision was placed, overlying the hallux interphalangeal joint.
Fig. 3. Dissection exposed the extensor hallucis longus tendon, which was either retracted or transected for joint visualization and later repaired.
would yield favorable outcomes in the immediate and long-term periods compared with standard care consisting of local wound care and offloading alone. To assess this hypothesis, we reviewed the data from a series of patients who had undergone either HIPJ arthroplasty or standard, nonoperative therapy for the treatment of recalcitrant neuropathic HIPJ ulceration.
or plantar-medial hallux lesions. Patients with diabetic neuropathy or idiopathic peripheral neuropathy and plantar hallux ulcerations, of any size or depth, that had been present for longer than 6 weeks were included for review. Patients were excluded if the duration of the ulcer was less than 6 weeks or if the ulceration was not located on the weightbearing surface of the hallux. Patients with active soft tissue or osseous infection were included. Patients with incomplete documentation and those who had been lost to follow-up were excluded. The patients who underwent HIPJ arthroplasty were identified using the aforementioned electronic medical record database. A search using Current Procedural Terminology code 28160 (Current Procedural Terminology, American Medical Association, Chicago, IL) was performed on the list of subjects with hallux ulcerations after applying the exclusion criteria. The patients who underwent the procedure were included in the HIPJ arthroplasty surgical group. Patients treated nonoperatively were placed into the control group. Selection of this group was made on the basis of propensity score matching with the goal of achieving similarity on age, diabetes, peripheral neuropathy, body mass index, and ulcer volume. Matching was performed using the R package Matching. The patients
Patients and Methods The patients were treated within the Podiatric Section of the Department of Orthopaedics from January 2008 to May 2013 (5 years, 3 months) at the Cleveland Clinic (Cleveland, OH). The present investigation was conducted with institutional review board approval. Patients with chronic plantar or plantar-medial ulcerations were identified using an electronic medical record database. Specifically, an institutionally designed Smartform filter in the Epic Electronic Health Records system (Epic, Verona, WI) was applied to generate a list of patients with a history of plantar
Fig. 4. Partial proximal phalangeal head resection. (A) A sagittal saw blade was used to resect the cartilaginous surface of the proximal phalanx. (B) A minimal amount of cartilage and bone was removed to preserve the hallux length.
E. Lew et al. / The Journal of Foot & Ankle Surgery 54 (2015) 541–548
543
Fig. 5. Hallux interphalangeal joint arthroplasty. (A and B) A reciprocating rasp or rongeur was used to contour any prominent bone on the resected surface of the proximal phalanx and the plantar medial condyle of the distal phalanx.
in the surgical and conservative treatment groups were treated in a standardized fashion consistent with diabetic foot wound care at the Cleveland Clinic. Treatment was based on 6 principles of wound healing: 1. A thorough evaluation of lower extremity perfusion status with noninvasive vascular studies and referrals to a vascular specialist when necessary 2. Eradication of infectious processes 3. Use of proper off-loading techniques
Fig. 6. An anteroposterior radiograph demonstrating fixation with a 0.062-in. Kirschner wire, which was used for percutaneous stabilization of the hallux in some cases in which a larger amount of bone resection was necessary.
4. Ensuring adequate wound moisture balance, along with appropriate wound debridement 5. Edema control 6. Encouraging adequate nutrition and optimizing overall medical condition with internal medicine or endocrinology consultations We explored the number of weeks until healing after the initiation of treatment and evaluated variables for an association with nonhealing. Other outcomes evaluated in both treatment groups included recurrence of ulceration and amputation. The demographic data for age, sex, and body mass index were recorded. The hallux ulceration duration before the initiation of treatment and the size before treatment initiation was documented in weeks and cubic centimeters, respectively. The ankle brachial index, toe brachial index, and incidence of revascularization procedures were also recorded. The
Fig. 7. The skin was closed with nonabsorbable suture in a simple interrupted fashion.
544
E. Lew et al. / The Journal of Foot & Ankle Surgery 54 (2015) 541–548
Fig. 8. Evaluation of hallux interphalangeal joint range of motion. (A) The preoperative range of motion was assessed. (B) An increased range of motion was observed intraoperatively after hallux interphalangeal joint arthroplasty. other comorbidities included for review were diabetes mellitus, chronic kidney disease, peripheral arterial disease (PAD), hemoglobin A1c, and tobacco use. Postoperative complications, including infection, dehiscence, neuritis, new ulceration development, and deep venous thrombosis, were evaluated in the surgical group only. Data collection and a review of the electronic medical records was conducted by 2 investigators (E.L. and N.N.). The HIPJ arthroplasty surgical procedure was performed by 1 podiatric surgeon (P.M.). Nonoperative management of the hallux ulcerations in the standard therapy group was performed by members of the Podiatric Section of the Department of Orthopaedics at the Cleveland Clinic. The outcomes were determined from a review of the medical records. The results of the comparative data were interpreted by 2 investigators (E.L. and N.N.), who were aware of the treatment groups because, although they had participated in the surgical treatment of 2 of the patients in the case series, they had not participated in the nonoperative patient care. The Quantitative Health Sciences Department at the Cleveland Clinic computed the statistics. Procedure The patients in the surgical group underwent HIPJ arthroplasty (Fig. 1) using a dorsal approach over the HIPJ (Fig. 2). Dissection was carried through the subcutaneous tissue, and the extensor hallucis longus was identified and retracted (Fig. 3). However, if necessary for exposure, the extensor hallucis longus was transected or lengthened and later repaired. The HIPJ was opened longitudinally, and the cartilaginous surface of the head of the proximal phalanx was removed using a saw blade (Fig. 4). A reciprocating rasp or rongeur was used to smooth and contour prominent bone, including the resected surface of the proximal phalanx and the plantar medial condyle of the distal phalanx (Fig. 5). In some cases, fixation with a 0.062-in. Kirschner wire was placed
percutaneously to stabilize the hallux during the healing period. The decision to pin the joint was made intraoperatively according to the amount of bone resection and stability of the hallux (Fig. 6). No fixation was applied in the presence of any suspected chronic infection or deep ulcer contamination. The subcutaneous layer was re-approximated with absorbable suture in subdermal inverted interrupted fashion while the skin layer was closed with nonabsorbable suture in simple interrupted fashion (Fig. 7). The range of motion of the hallux interphalangeal and metatarsophalangeal joints was assessed intraoperatively to determine whether it was increased (Fig. 8). The postoperative course consisted of protected weightbearing in forefoot offloading devices such as pressure relief shoes with modifiable inserts or forefoot relief shoes. At the 2-week postoperative visit, the skin sutures were typically removed. All the patients in the standard therapy group underwent offloading with either total contact casting, custom molded inserts, forefoot relief shoes, or drill-and-filled modified shoes (Fig. 9). The drill-and-fill technique involves removal of the dense rubber outsole that underlies the ulcer and replacing with softer material. The patients in both groups received appropriate wound debridement and local wound care throughout the treatment periods. Preventative treatment after ulcer healing was maintained in both groups with frequent monitoring and the use of standard offloading inserts and footgear. Shoe gear consisted of prescriptive depth inlay shoes, which included sufficient room in the toebox to accommodate the insert. Statistical Analysis Continuous variables are presented as the mean standard deviation or median and interquartile range. Both ordinal and nominal categorical variables are described using counts and percentages. Differences in continuous variables between the surgical
Fig. 9. Patients in both the conservative and the surgical groups underwent appropriate wound care and ambulated with various forms of forefoot offloading, including total contact casting, custom molded inserts, forefoot relief shoes, or custom shoes and inserts, with a drill and fill technique. (A) The ulceration location was identified on the insole. (B) The corresponding area on the shoe was drilled. (C) This drill-and-fill technique offloads any weightbearing pressure off the ulcerations.
E. Lew et al. / The Journal of Foot & Ankle Surgery 54 (2015) 541–548
545
Table 1 Comparison of baseline patient characteristics by treatment group (N ¼ 26 patients)
Table 2 Comparison of outcomes by treatment group (N ¼ 26 patients)
Risk Factor
Outcome Variable
Patient age (y) Gender Female Male Body mass index >25.0 kg/m2 Diabetes mellitus neuropathy Chronic kidney disease Peripheral arterial disease Hemoglobin A1c (>6.5%) History of tobacco use Osteomyelitis Ulcer volume (>1.0 cm3) Ulcer duration (>6 wk)
Treatment Group
p Value
HIPJ Arthroplasty (n ¼ 13)
Standard Therapy (n ¼ 13)
55.38 13.02
54.77 12.28
5 8 12 13 4 1 6 8 1 13 13
2 11 12 13 7 4 11 4 1 13 13
(92.3) (100) (30.77) (7.69) (46.15) (61.54) (7.69) (100) (100)
(92.3) (100) (53.85) (30.77) (84.6) (30.77) (7.69) (100) (100)
.87* .38y
NA NA .45y .13y .81z .25y > .99y > .99z .001z
Abbreviations: HIPJ, hallux interphalangeal joint; NA, not applicable. Data presented as mean standard deviation, n, or n (%). * Student’s paired t test. y Exact McNemar’s test. z Wilcoxon signed ranks test. and conservative treatment groups were tested using either Student’s paired t test or paired Wilcoxon signed ranks test. Differences in nominal categorical variables were tested using the exact McNemar test. Differences in ordinal categorical variables were tested using paired Wilcoxon signed ranks tests. The time-to-heal was analyzed using Cox proportional hazards models, with clusters defined by the matched group number, with the comparison between groups performed using the Wald test. The median healing times were estimated using the Kaplan-Meier method. All analyses were performed using R software, version 3.0.1 (R Project for Statistical Computing, Vienna, Austria). A 5% level of significance was used for all testing.
Results A total of 158 patients with chronic plantar hallux ulcerations were treated from January 2008 to May 2013 at the Podiatric Centers of the Cleveland Clinic (Cleveland, OH). After applying the exclusion criteria, 55 patients were included in our retrospective cohort study. Of the 55 patients, 13 (24%) underwent HIPJ arthroplasty and 42 (76%) underwent standard therapy. Thus, 13 age-matched patients were selected from the standard therapy group for comparison with the 13 patients in the surgery group. Therefore, 13 patients each were present in the surgical and standard therapy groups, and 26 patients were reviewed. A statistical comparison of the baseline risk factors for each treatment group is presented in Table 1. The overall mean patient age was 55 13.0 years, and 73% of the patients were male. The mean followup duration was 19.5 (range 1.2 to 47.9) months. The median time to healing was 9 [2, 17.29] weeks in the standard therapy group and 3.5 [2.5, 4.25] weeks in the surgical group (p ¼ .033). The incidence of recurrent hallux ulceration was 7.7% (1 of 13) in the surgical group and 53.9% (7 of 13) in the standard therapy group (p ¼ .041). The incidence of amputation was 0% (0 of 13) in the surgical group and 38.5% (5 of 13) in the standard therapy group (p ¼ .063). Comparisons of the healing time, ulcer recurrence, and amputations between the treatment groups are presented in Table 2. A total of 4 postoperative complications (30.8%) developed in the surgical group. These included incisional wound dehiscence in 2 (15.4%), infection in 1 (7.7%), and new (transfer) ulcer formation in 2 (15.4%). A new ulcer formed at the plantar aspect of the second toe in 1 patient, and another formed on the plantar aspect of the second metatarsophalangeal joint. Both of these transfer ulcerations healed with local supportive care. Case 1 A 50-year-old male presented with a left foot, plantar-medial hallux interphalangeal ulceration. He stated that the ulcer had been recurrent
Interval to heal (wk) Patients with recurrence Patients with amputation
Treatment Group
p Value
HIPJ Arthroplasty (n ¼ 13)
Standard Therapy (n ¼ 13)
3.5 (2, 17.29) 1 (7.69) 0 (0)
9 (4.25, 25) 7 (53.85) 5 (38.46)
.033* .031y .063y
Abbreviation: HIPJ, hallux interphalangeal joint. Data presented as median (minimum, maximum) or n (%). * Wald test from Cox proportional hazards regression. y Exact McNemar test.
for longer than 15 years. Previous treatment had included debridement, iodine cleansing, and the use of silver impregnated dressings. The patient’s medical history included type 2 diabetes mellitus with neuropathy, hypertension, and hyperlipidemia. The patient’s most recent hemoglobin A1c was 7.0 mg/dL, his body mass index was 38 kg/ m2, and he was a current smoker with a 30-pack-year history. On physical examination, the patient’s neurovascular status was intact, including the presence of protective touch-pressure sensation. A full-thickness ulcer was noted, localized to the plantar aspect of his left HIPJ and measuring 2.0 cm in length 2.0 cm in width 0.4 cm in depth. The ulceration did not probe to the bone, and no evidence of purulence, cellulitis, or signs of infection were noted. Radiographs revealed intact hallux phalangeal cortices without signs of lysis or proliferation. The patient spent a lot of time on his feet daily as a heating, ventilation, and air conditioning technician and was unable to take time off from work. Conservative treatment was attempted, consisting of weekly debridement of nonviable tissue and daily home dressing changes. The patient ambulated in new custom orthotics in regular shoes but refused to wear pressure relief shoes. Four weeks later, the patient’s course was complicated by forefoot cellulitis that resolved with oral antibiotics. The blood culture results were negative, no ascending lymphangitis was present, and no evidence of osseous lysis or proliferation was found on radiographic inspection. During the next several months, the ulceration waxed and waned in size, ranging from 0.5 cm 0.4 cm 0.5 cm to 2.0 cm 0.8 cm 0.3 cm. After approximately 9 months of conservative treatment, the wound measured 0.5 cm 0.5 cm 0.2 cm (Fig. 10). At that point, the patient was frustrated with the chronicity of the ulceration and opted to pursue surgical options. Subsequently, the patient underwent left HIPJ arthroplasty. The patient continued frequent monitoring, offloading, and local wound care postoperatively. The ulceration progressively decreased in size and had fully healed at 4 weeks postoperatively (Fig. 11), and the patient resumed his work duties and regular shoe gear. The patient had neither complications nor recurrence at 12 months of follow-up. Case 2 A 59-year-old male with a medical history of uncontrolled diabetes and peripheral neuropathy presented with a chronic and recurrent right hallux ulceration that had been present for more than 5 years. Other significant elements of the patient’s medical history included hypertension, chronic kidney disease, and hyperlipidemia. The patient’s hemoglobin A1c was 8.8 mg/dL, and his body mass index was 37.3 kg/m2. He was also a half-a-pack per day smoker. On physical examination, the patient’s neurovascular status was intact, including protective touch-pressure sensation. A full-thickness skin ulceration was located at the plantar aspect of the left HIPJ. It measured 1.5 cm 0.7 cm 0.5 cm. The ulceration probed to the bone through a sinus tract. Radiographs revealed cortical disruption and lytic changes at the medial condyle of the base of the distal phalanx
546
E. Lew et al. / The Journal of Foot & Ankle Surgery 54 (2015) 541–548
Fig. 10. A 50-year-old diabetic male with a chronic neuropathic plantar hallux ulceration underwent hallux interphalangeal joint arthroplasty.
Fig. 12. An anteroposterior radiograph demonstrating a periosteal reaction and cortical lysis along the base of the distal phalanx and medial aspect of the hallux interphalangeal joint.
(Fig. 12). A white blood cell–labeled bone scan was indicative of chronic osteomyelitis of the distal phalanx. Thereafter, the patient underwent operative debridement and resection of the affected HIPJ. The intraoperative bone culture results were positive for Streptococcus agalactiae. Bone pathologic examination also revealed findings consistent with chronic osteomyelitis. He received appropriate intravenous antibiotic therapy on consultation
with an infectious disease specialist. The patient was followed up on a weekly basis with local wound care and offloading. His postoperative course was complicated by dorsal wound dehiscence and cellulitis at 7 days postoperatively. The wound eventually closed by secondary intention after operative debridement, antibiotic therapy, and local wound care. At 6 weeks postoperatively, the plantar wound had completely healed. He was maintained with custom pressurerelieving insoles and prescriptive depth-inlay shoe gear. He had no ulcer recurrence through 5 years of follow-up (Fig. 13). Discussion
Fig. 11. The patient’s wound had completely healed at 4 weeks postoperatively.
The concept of internal offloading of chronic ulcerations has been described previously and applied to other areas of the foot (1–5). Several surgical treatments have been proposed for healing digital ulcerations, including metatarsal head resection, arthrodesis, and amputation (3). A study by Armstrong et al (4), in 2003, concluded that first proximal phalangeal base resection, or Keller osteotomy, resulted in a faster healing time and fewer repeat ulcerations than standard conservative therapy. A study by Stephens (1), in 2000, reported using flexor hallucis longus contracture release to reduce plantar pressure and heal chronic plantar hallux ulcerations. Kim et al (5), in 2008, revealed that removal of a lesser digital phalanx can result in ulceration healing. Quebedeaux et al (7), in 1996, concluded that hallux amputation contributes to lesser digital deformities and new ulcer formation. Lavery et al (8) postulated that the reason for this was because, after a hallux amputation, the pedal pressure distribution is altered and increases the reamputation risk. In lieu of hallux amputation, toe-preserving surgical intervention such as HIPJ arthroplasty could serve as an alternative to address chronic HIPJ ulcerations and decrease the risk of altered pedal pressure distribution, thus decreasing the risk of lesser digital amputation.
E. Lew et al. / The Journal of Foot & Ankle Surgery 54 (2015) 541–548
547
Fig. 13. (A and B) A 59-year-old male with a chronic plantar medial hallux ulceration and osteomyelitis underwent debridement and resection interphalangeal joint arthroplasty. The plantar wound was healed at 6 weeks postoperatively. No ulcer recurrences were reported at the final follow-up visit at 5 years.
To our knowledge, only 1 study has evaluated the efficacy of HIPJ arthroplasty for the treatment of recalcitrant hallux ulceration (6). In 1994, Rosenblum et al (6) reviewed 46 HIPJ arthroplasties in 40 patients treated for chronic neuropathic ulcer of the hallux. They found that 91% of the patients had healed with no evidence of recurrence in the followup period, which ranged from 4 to 44 months. They also concluded that HIPJ arthroplasty is a valuable procedure for chronic ulcerations. The results of the present study revealed that HIPJ arthroplasty is an effective treatment option to aid in the healing of chronic plantar hallux nonischemic ulcerations in diabetic patients with neuropathy. These results correlate with those from the study by Rosenblum et al (6) performed more than 20 years earlier. An osseous prominence of the HIPJ predisposes the plantar aspect of the hallux to ulcer formation, along with plantar contractures and restricted motion, often observed in neuropathic patients. Resection of this joint prominence internally relieves pressure on the soft tissue and increases the range of motion, thereby eliminating key factors in ulcer development and recurrence. In the surgical group, the ulcerations healed within a shorter period, with no amputations required postoperatively. All patients continued the use of custom inserts and footgear after ulcer resolution, which we believe played a key role in preventing ulcer recurrence. One ulcer recurrence was observed in the HIPJ arthroplasty group at 2 years postoperatively. Additional evaluation of this patient revealed comorbidities characteristic of advanced disease. In addition to a diagnosis of diabetes mellitus, chronic kidney disease, and PAD, the patient had a hemoglobin A1c greater than 10%, an ankle brachial index greater than 1.3, and an ulcer volume greater than 5 cm3. This patient also required the longest to heal postoperatively (12 weeks). Given these findings, we expect that patients with vasculopathy, longstanding disease, and uncontrolled blood glucose levels will have poorer outcomes. The complications in the surgical group included 2 cases (15.4%) of minor incisional wound dehiscence, including 1 patient with associated infection and cellulitis that healed uneventfully. No cases of neuritis or deep vein thrombosis occurred. Another observed outcome was ulcer development localized to the adjacent digits, which occurred in 2 of the patients (15.4%) in the HIPJ arthroplasty group. A new ulcer formed at the plantar aspect of the second toe in 1 patient. In yet another patient, an ulcer developed on the plantar aspect of the second metatarsophalangeal joint. Both of these transfer ulcers healed uneventfully without surgical intervention and did not result in chronic wounds or amputation after addressing the lesions with accommodative offloading inserts. It is believed that excessive resection of the proximal phalangeal head can render the hallux unstable and shortened. This could lead to the potential development of transfer lesions to the second ray. An unstable hallux can also result in a cosmetically
displeasing joint and the possible lack of hallux purchase. Neither of these complications was encountered in the present cohort of patients. In contrast, reduction of the hallux purchase with this procedure proved therapeutic and aided in pressure relief and the prevention of ulcer recurrence. The patients were generally satisfied with appearance and function of their great toe. Ultimately, the foot and ankle surgeon will need to determine how much bone resection will be necessary, taking into account the lesser digital length, to avoid transfer lesions, ulcer recurrence, hallux instability, and excessive shortening. Removing the least amount of bone as possible will usually result in a more relatively stable joint that does not necessitate pin fixation. Another potential complication is hallux extensus, which can be avoided by not overshortening the extensor halluces longus tendon after tenotomy. The relative contraindications to this procedure include moderate to severe PAD. Patients with advanced PAD have a greater risk of surgical wound ischemia, infection, and secondary amputation. There is also an associated risk with surgery in diabetics with multiple comorbidities and advanced multisystem disease. The decision to perform surgery on these patients must be carefully considered, balancing the risks and benefits of this elective procedure. Foot and ankle surgeons are strongly encouraged to obtain appropriate medical, infectious disease, and vascular surgical consultations to optimize their patients’ success for healing. Lavery et al (2) in 2012 stated that despite the greater 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 ulceration. Osteomyelitic infection is not an absolute contraindication. One patient in the surgical group had osteomyelitis of the HIPJ. After surgical resection and prolonged antibiotic therapy, this patient’s ulceration had healed uneventfully at 6 weeks postoperatively. This patient did experience dorsal incisional wound dehiscence and infection that resolved with operative debridement and wound care. The patient had had no recurrence at 5 years of follow-up. Additional fixation with a 0.062-in. Kirschner wire should be avoided in the presence of deeper ulceration or suspicion of infection to avoid potential seeding of any pathogen proximally. The present study had several limitations, including the nature of a retrospective analysis, which can make it difficult to procure meaningful, reliable data in some cases. Also, the number of patients in the surgical group was small, limiting stronger statistical analysis measures pertaining to time-to-event outcomes. There was also the possibility of selection bias related to patient compliance in the surgical group. More advanced, higher powered, prospective or randomized controlled trials are necessary to further evaluate this procedure as a curative measure for chronic plantar hallux wounds.
548
E. Lew et al. / The Journal of Foot & Ankle Surgery 54 (2015) 541–548
In conclusion, HIPJ arthroplasty for the treatment of recalcitrant hallux ulcerations appears to yield favorable outcomes in the immediate and long-term periods, including a shorter healing time compared with a nonoperative control group. This procedure is a viable option for foot and ankle surgeons in preventing hallux ulcer recurrence and amputations in patients with diabetes-related neuropathy. The results of the present investigation could be used in the development of future prospective cohort studies and randomized controlled trials focusing on the treatment of diabetic foot ulceration localized to the HIPJ. Acknowledgments We express our appreciation to Colin O’Rourke, MS (Quantitative Health Sciences Department), for conducting the statistical analysis, and Tammy Owings, DEng (Department of Biomedical Engineering), for her kind assistance with patient data extraction.
References 1. Stephens HM. The diabetic plantar hallux ulcer: a curative soft tissue procedure. Foot Ankle Int 21:954–955, 2000. 2. Lavery LA. Effectiveness and safety of elective surgical procedures to improve wound healing and reduce re-ulceration in diabetic patients with foot ulcers. Diabetes Metab Res Rev 28(suppl 1):60–63, 2012. 3. Resch S. Corrective surgery in diabetic foot deformity. Diabetes Metab Res Rev 20(suppl 1):S34–S36, 2004. 4. Armstrong DG, Lavery LA, Vasquez JR, Short B, Kimbriel HR, Nixon BP, Boulton AJ. Clinical efficacy of the first metatarsophalangeal joint arthroplasty as a curative procedure for hallux interphalangeal joint wounds in patients with diabetes. Diabetes Care 26:3284–3287, 2003. 5. Kim JY, Kim TW, Park YE, Lee YJ. Modified resection arthroplasty for infected non-healing ulcers with toe deformity in diabetic patients. Foot Ankle Int 29: 493–497, 2008. 6. Rosenblum BI, Giurini JM, Chrzan JS, Habershaw GM. Preventing loss of the great toe with the hallux interphalangeal joint arthroplasty. J Foot and Ankle Surg 33:557–560, 1994. 7. Quebedeaux TL, Lavery LA, Lavery DC. The development of foot deformities and ulcers after great toe amputation in diabetes. Diabetes Care 19:165–167, 1996. 8. Lavery LA, Lavery DC, Quebedeax-Farnham TL. Increased foot pressures after great toe amputation in diabetes. Diabetes Care 18:1460–1462, 1995.
