Foot & Ankle International

There is an increasing need for orthopaedic practitioners to measure and collect patient-reported outcomes data. In an effort to better understand outcomes from operative treatment, the American Orthopaedic Foot & Ankle Society (AOFAS) established the Orthopaedic Foot and Ankle Outcomes Research (OFAR) Network, a national consortium of foot and ankle orthopaedic surgeons. We hypothesized that the OFAR Network could successfully collect, aggregate, and report patient-reported outcome (PRO) data using the National Institutes of Health (NIH) Patient Reported Outcomes Measurement Information System (PROMIS).

Ten sites enrolled consecutive patients undergoing elective surgery for 1 of 6 foot/ankle disorders. Outcome instruments were collected preoperatively and at 6 months postoperatively using the PROMIS online system: Foot and Ankle Ability Measure (FAAM), Foot Function Index (FFI), and PROMIS physical function (PF) and pain computerized adaptive tests (CAT). During the 3-month period, 328 patients were enrolled; 249 (76%) had completed preoperative patient-reported outcomes data and procedure-specific data. Of these, 140 (56%) also completed 6-month postoperative patient- reported outcomes data.

Ankle arthritis and flatfoot demonstrated consistently worse preoperative scores. Five of 6 disorders showed significant improvement at 6 months on PF CAT and FAAM, 4 of 6 showed improvement on pain interference CAT, and no disorders showed improvement on FFI. Ankle arthritis and flatfoot demonstrated the greatest magnitude of change on most patient-reported outcomes scales.

We were able to enroll large numbers of patients in a short enrollment period for this preliminary study. Data were easily aggregated and analyzed. Substantial loss of follow-up data indicates a critical area requiring further effort. The AOFAS OFAR Network is undergoing expansion with goals to ultimately facilitate large, prospective multicenter studies and optimize the quality and interpretation of available outcome instruments for the foot and ankle population.

Level II, prospective comparative study.

Little has been published on the management of infected total ankle replacements. This study reviewed our experience with this difficult clinical problem.

From 2002 to 2011 we retrospectively evaluated patients requiring prosthesis removal for infection following primary or revision total ankle replacement. Infections were classified as acute postoperative, late chronic, or remote hematogenous. The demographics, treatment, and outcome of the infected replacements were evaluated.

Of 613 replacements performed during this time period, 15 infections occurred at our institution and 4 from other institutions were treated. These 19 cases included 15 late chronic infections, 3 early postoperative infections, and 1 acute hematogenous infection. Fourteen of the infections occurred in a subset of 433 prostheses (3.2%) placed between 2002 and 2009 when we used an implant with a unique design that necessitated a tibiofibular syndesmosis fusion for the tibial component support (Agility, Depuy). One infection occurred out of 139 other prostheses (0.7%) between 2009 and 2011 when we switched to an anatomic design fixed bearing implant system. Prosthesis salvage was attempted in the 3 early postoperative and 1 acute hematogenous cases, however they all required subsequent removal of the prosthesis. Mean time to removal of the prosthesis after the index arthroplasty was 18 months. Mean follow-up was 19 months from the time of prosthesis removal. Three of the 19 patients underwent successful revision with replacement, 6 with arthrodesis, 7 with a permanent antibiotic spacer, and 3 patients required a transtibial amputation.

Only a limited number of patients who develop a deep infection following primary or revision total ankle arthroplasty can expect to undergo successful joint-preserving revision ankle arthroplasty.

Level III, retrospective case series.

Concomitant procedures are being performed with total ankle replacement (TAR) to improve alignment, function, and mobility. The purpose of this study was to examine the differences in outcomes between patients who had a concomitant triceps surae lengthening (gastrocnemius recession [GSR] or triple hemisection [TAL]) versus a group that underwent TAR alone preoperatively and 1 year after TAR.

For this prospective, nonrandomized study, 229 patients (37 GSR, 22 TAL, and 170 TAR alone) were examined. Patient-reported outcomes, physical performance, and lower extremity gait mechanics were completed preoperatively and 1 year postoperatively. A series of repeated measures ANOVAs were used to determine significant differences (P < .05), and Tukey’s post hoc testing was used to follow any significant ANOVA results.

No difference existed in BMI, age, gender, or dorsiflexion (DF) angle at heel strike between the triceps surae lengthening groups or between preoperative and 1 year following TAR. Walking speed, the physical performance measures, the AOFAS Hindfoot Score, SF-36, peak plantar flexion angle, and the peak plantar flexion moment were significantly improved (P < .001) postoperatively with no differences between the triceps surae lengthening groups. The peak DF angle (P = .006) and the ankle range of motion (P = .014) demonstrated a greater improvement from preoperative to 1 year postoperatively in the triceps surae lengthening groups in comparison to the TAR alone group.

Significant improvements existed between preoperative and 1 year postoperatively for most of the variables of interest independent of the triceps surae lengthening group. This study demonstrated that the use of a concomitant triceps surae lengthening procedure (GSR or TAL) resulted in equivalent outcomes when compared with a group undergoing TAR alone.

Level II, prospective comparative study.

Civilian literature has reported excellent outcomes after elective fasciotomy for chronic exertional compartment syndrome (CECS). Our study’s purpose was to objectively investigate the functional outcome of fasciotomies performed for CECS in a high demand military population.

A retrospective review of all fasciotomies performed for CECS at a single tertiary military medical center was performed. The primary outcome measure was the ability to return to full active duty. Diagnosis, operative technique, and number of compartments addressed were collected and analyzed. Patients were contacted and the visual analog scale (VAS) pain score, functional single assessment numeric evaluation (SANE) score, as well as overall satisfaction were reported. Return to duty status was collected on 70 of 70 (100%) consecutive operative extremities in 46 patients with an average follow-up of 26 months.

Only 19 patients (41.3%) were able to return to full active duty. Ten patients (21.7%) underwent a medical separation from the military and 17 patients (37%) remained in the military but were on restricted duty secondary to persistent leg pain. Thirty-five of 46 (76%) of the patients were contacted and provided subjective feedback. The average SANE score was 72.3, and there was a mean improvement of 4.4 points in VAS score postoperatively. Overall, 71% of patients were satisfied and would undergo the procedure again. Outcomes were correlated to operative technique, patient rank, and branch of military service.

Our study showed a return to full military duty in 41% of patients who underwent elective fasciotomy for CECS. Overall 78% of patients remained in the military, which is consistent with previous military literature. Subjective satisfaction rate was 71%. Both the return to activity and subjective outcomes in our study population were substantially lower than reported results in civilian populations.

Level IV, case series.

Instability of the lesser metatarsophalangeal (MTP) joints is a common cause of deformity and pain. The purpose of this study was to prospectively evaluate the surgical outcomes for the different grades of plantar plate tears and propose a treatment protocol.

Sixty-eight patients (100 MTP joints) were prospectively enrolled and graded according to the anatomical grade system for plantar plate tears. Based on this classification, the appropriate surgical procedure was chosen as follows: grades 0 and I, thermal shrinkage with radiofrequency; grades II and III, direct reinsertion of the plantar plate; and grade IV, flexor-to-extensor tendon transfer. All surgical procedures were associated with a Weil metatarsal osteotomy. Evaluations were performed before and after surgical treatment with a mean follow-up of 2 years (12-36 months), using clinical and radiological parameters: American Orthopaedic Foot and Ankle Society (AOFAS) Lesser MTP-IF Scale, visual analog scale (VAS), ground touch, joint stability, and toe purchase.

The analysis of the clinical parameters demonstrated a significant improvement of all groups (P < .0001) after surgical treatment, but grade IV had less VAS improvement and a fair AOFAS average score (72 points). All groups improved regarding physical examination parameters, but grade I, III, and IV tears had proportionally less stable MTP joints following surgery, as well as a lower proportion of normal postoperative toe purchase and ground touch. All groups showed a significant improvement regarding radiographic parameters.

All groups of operatively treated patients had significant improvement with regard to subjective and objective parameters. Grades I, III, and IV presented inferior results compared with grades 0 and II.

Level IV, case series

Substantial attention has recently been placed on fractures of the posterior malleolus. Fracture extension to the posteromedial rim (“posterior pilon variant”) may result in articular incongruity and talar subluxation. Current classification systems fail to account for these fractures. The relative frequency of this fracture, its associated patient characteristics, and the reliability of its diagnosis have never been reported in such a large series.

We retrospectively identified 270 patients who met our inclusion criteria. Basic demographic data were collected. The fractures were classified according to Lauge-Hansen and AO/OTA. Additional radiographic data included whether the fracture involved the posterior malleolus and whether the fracture represented a posterior pilon variant. Univariate statistical methods, chi-square analysis, and interobserver reliability were assessed.

The relative frequency of posterior malleolus fracture was 50%. The relative frequency of the posterior pilon variant was 20%. No significant difference was noted with respect to the frequency of posterior malleolar or posterior pilon variant between the subgroups of the AO/OTA and Lauge-Hansen classification systems when compared to the overall fracture distribution. Patients with posterior malleolar fractures and posterior pilon variants were significantly older. Females were significantly more likely than men to sustain posterior malleolar fractures and posterior pilon variants. Patients with diabetes trended toward a greater risk of both types of fractures. Interobserver reliability data revealed substantial agreement for posterior malleolar fractures and posterior pilon variants.

These data represent the highest reported rate of posterior malleolar involvement in operatively treated ankle fractures and is the first to describe the percentage of the posterior pilon variant in such a large series. The interobserver reliability data demonstrate substantial agreement in identification of posterior malleolar fractures and the posterior pilon variant based on plain radiographs. Certain patient characteristics such as age, sex, and diabetes may be associated with these fractures.

Level III, retrospective cohort study.

Triple arthrodesis is a powerful hindfoot corrective procedure involving fusion of the talonavicular (TN), calcaneocuboid (CC), and subtalar (ST) joints. A 2-incision approach, a single-incision medial triple arthrodesis, and a single-incision medial double arthrodesis are well described. We present a single-incision lateral approach to triple arthrodesis.

We retrospectively reviewed 70 patients who underwent triple arthrodesis at our institution from 2007 to 2011. Patients had either double-incision (n = 33) or single-incision lateral (n = 37) triple arthrodesis. A single surgeon performed all procedures. The most common diagnosis was stage III planovalgus deformity. Deformity correction, union rate, time of surgery, complications, wound healing, reoperations, and pre- and postoperative visual analog scale (VAS) pain scores were analyzed for both groups.

There were no statistical differences in deformity correction, wound healing, complications, reoperations, or improvement in VAS pain scores. Operation time was significantly shorter in the single-incision lateral group (86 minutes vs 95 minutes, P = .0395). There was no difference in union rates with regard to the TN, ST, or CC joints. Five patients had radiographic nonunions of the CC joint between both groups.

This is the first study that presents outcomes of a single lateral approach for triple arthrodesis. The single-incision approach was faster. The low rate of symptomatic nonunions suggests that fusion of the CC joint may not be important in symptomatic relief or deformity correction.

Level III, retrospective comparative study.

Latent syndesmotic instability is a common cause of chronic ankle pain. The diagnosis is not readily apparent on static imaging as the fibula remains reduced. The hypothesis of this study was that a previously undescribed novel finding on coronal MRI (lambda sign) is an independent indicator of latent syndesmosis instability. We also report on the utility of classic radiographic and physical exam findings.

A total of 23 patients with latent syndesmotic instability diagnosed via arthroscopy (group I) were compared to a cohort of 40 patients who were found to have a stable syndesmosis during arthroscopy for unrelated conditions (group II). A retrospective chart review was performed evaluating their clinical history, preoperative physical examination, and radiologic findings. The lambda sign is a high intensity signal seen on coronal MR imaging that resembles the Greek letter lambda.

All of the physical exam findings tested were statistically significant. Pain at the syndesmosis had the highest sensitivity (83%), while pain reproduced with the proximal squeeze test resulted in the highest specificity (89%). The external rotation stress test had the highest positive predictive value (75%). Of the radiographic examinations performed, only the lambda sign was found to have statistical significance with a sensitivity of 75% and a specificity of 63%. The presence of a lambda sign on the MRI of patients with physical exam findings suggestive of syndesmotic pain was highly sensitive (75%) and specific (85%).

The lambda sign noted on the coronal MRI was both sensitive and specific for injuries involving greater than 2 mm of diastasis on arthroscopic stress examination of the syndesmosis. While neither the lambda sign nor any other finding on physical or radiographic examination represented an independent predictor of syndesmotic instability, the presence of a lambda sign in concert with positive physical exam findings might help health care providers determine which patients might benefit from operative intervention or referral.

Level III, case control study.

Traditional treatment of talonavicular osteochondral lesions (OCLs) requires an open procedure. Arthroscopic microfracture of talonavicular OCLs may provide a viable, minimally invasive approach. The purpose of this study was to describe an arthroscopic approach for treatment of talonavicular OCLs, describe the proximity of arthroscopic portals to important structures in cadaver specimens, and report magnetic resonance imaging (MRI) findings and clinical outcomes of this technique.

Five cadaver specimens were dissected so proximity of portals to adjacent tendons and neurovascular structures could be assessed. Subsequently, 3 athletic patients with OCLs of the talonavicular joint were treated with arthroscopic debridement and microfracture. Patient records and imaging studies were retrospectively reviewed.

In the cadaver specimens, the mean distance between the neurovascular bundle and the medial border of the extensor hallucis longus (EHL) was 9.0 (range, 8 to 10) mm. The saphenous nerve was located a mean of 6.8 (range, 6 to 7) mm from the medial border of the tibialis anterior tendon. Therefore, portals were placed just medial to the EHL and tibialis anterior tendon to avoid the neurovascular bundle and saphenous nerve, respectively. In all patients, access, identification of the OCL, debridement, and microfracture were successfully performed. All patients demonstrated improvements in Foot and Ankle Outcome Scores and Short Form–12 scores and began gradual return to activity within 12 weeks following the operation. No significant complications occurred. MRI indicated signal consistent with reparative fibrocartilage in all patients.

Talonavicular arthroscopy allowed visualization, curettage, synovectomy, loose body removal, and microfracture of OCLs that would have otherwise required an open approach. At early follow-up, all patients had returned to their previous activity levels. Arthroscopy of the talonavicular joint was a viable approach for microfracture of OCLs.

Level IV, case series.

There is heterogeneity in the literature regarding the anatomy and number of ligamentous bands that form the deltoid ligament (DL). Anatomic knowledge of the DL and its variations are important for surgeons who repair ankle fractures.

The DL was dissected in 33 ankles from 17 formalin-fixed cadavers (mean age at death, 76.6 years) to examine its morphology. The length, width, and thickness of its constituent bands were recorded with a digital caliper. Descriptive and correlational statistics were used to investigate the relationships between band size, age at death, and sex. A literature review was conducted to compare our data to those of previous studies.

The DL has superficial and deep layers with up to 8 different bands.

The DL stabilizes the medial ankle and should be evaluated in flatfoot deformities and severe ankle fractures.

Anatomic knowledge of DL variations should aid the surgeon in repairing torn DLs.

Many techniques have been described for lateral ligament reconstruction. One frequently overlooked autograft option is the plantaris tendon, potentially due to the paucity of data on its mechanical characteristics. This study examined the structural properties of double and quadruple plantaris tendon constructs.

Plantaris tendons were harvested from 35 fresh-frozen human cadaver specimens (mean age, 66 years [range, 43-89 years]; 17 female, 13 male). The tendon ends were sutured in a running locking technique and then woven onto a template board to create double or quadruple graft constructs with a 20-mm functional length. If additional tendon length remained, a single 40-mm specimen was isolated to provide tissue material properties. Structural properties were calculated from the results of cyclic and failure uniaxial tensile tests.

Quadruple-strand constructs had a tensile strength of 205.8 ± 68.2 N and a stiffness of 133.1 ± 46.3 N/mm. Single strands had a tensile strength of 66.9 ± 26.3 N and a stiffness of 43.8 ± 14.7 N/mm. Material properties were similar to a prior study.

The average maximum tensile strength for the quadrupled plantaris grafts exceeded the strength of the intact anterior talofibular ligament of 139 to 161 N; therefore, the quadruple plantaris construct may be a viable autograft for foot and ankle ligament reconstruction.

The tensile strength of the plantaris tendon is comparable to, or stronger than, other grafts already in use and offers a donor site that may result in negligible loss of strength.

Increased navicular drop (NDro) and navicular drift (NDri) are associated with musculoskeletal pathology in adults. The aim of this study was to investigate navicular motion in children, with respect to foot posture, and identify altered patterns of motion that demonstrate midfoot dysfunction. Navicular motion in different activities was evaluated as well as the role of flexibility and body mass index (BMI).

Twenty-five children with flatfeet and 26 with neutral feet (age range, 8-15) underwent gait analysis using a 12-camera Vicon MX system (Vicon, UK). Navicular motion indices were calculated from marker coordinates. Student t tests and Pearson’s correlation coefficient (R) were used to investigate navicular motion differences between groups. The relationship between NDRo, NDRi, and their dynamic counterparts was also assessed.

Normalized NDri (NNDri) and normalized NDro (NNDro) correlated strongly in neutral feet (R = 0.56, P = .003) but not in flatfeet (R = 0.18, P > .05). Flatfeet demonstrated reduced NNDri compared to neutral footed children (0.7 vs 1.6, P = .007). No difference was observed in NNDro between groups. Standard and dynamic measures of NDri and NDRo were highly correlated. Navicular motion correlated poorly with BMI and flexibility.

Motion of the navicular in the transverse and the sagittal plane is important when investigating foot function. Uncoupling of this motion in flatfeet may indicate impaired midfoot function. Reduced navicular medial translation in flatfeet may indicate altered alignment of the talonavicular joint.

The measurement of dynamic navicular motion indices did not add information about dynamic foot function compared to measurement of static indices.