Abstract
Clinical and pedobarograph evaluation was performed on 16 patients following flexor hallucis longus (FHL) tendon transfers to determine the resulting morbidity due to the loss of FHL function. All patients underwent FHL tendon transfer for either chronic tendon Achilles rupture or chronic Achilles tendinosis. Clinical evaluation of hallux function was performed using the American Orthopaedic Foot and Ankle Society (AOFAS) hallux metatarsophalangeal-interphalangeal scale, the SF-36 score, and a clinical questionnaire to assess alteration in the clinical function of the hallux during activities of daily living. Pedobarography was carried out using the Musgrave pedobarograph system to detect changes in forefoot loading in comparison to the contralateral normal foot. Fourteen of the 16 patients scored maximally on the hallux metatarsophalangeal-interphalangeal scale and none of the patients noticed functional weakness of the hallux during activities of daily living at a mean follow-up of 43.6 months (range, 5–120 months). Pedobarograph readings showed a trend toward reduction in peak pressure loading on the distal phalanx, but this was not significant for the numbers of patients studied. There was no significant increase in loading of the first or second metatarsophalangeal joints to suggest that transfer metatarsalgia may complicate FHL tendon transfer. According to the results of the study morbidity from FHL transfer should be clinically insignificant.
Introduction
Flexor hallucis longus (FHL) tendon transfer is an accepted treatment for chronic rupture of the tendon Achilles (TA). 3,4,9,14,15 The advantages of using FHL in this situation include phasic contraction with the triceps surae, 14 the relative strength of the FHL musculotendinous unit, 13 and the distally situated muscle belly of FHL. The latter allows the recruitment of an increased blood supply to the repaired Achilles tendon constuct. 4,14 Alternatives to FHL, such as flexor digitorum longus (FDL), peroneus brevis, and plantaris, are less advantageous in these regards. Also, tendinous distal interconnections between FHL and FDL 12 theoretically permit some residual interphalangeal joint (IPJ) flexion of the hallux by the tenodesis effect of FDL following sectioning of the FHL tendon.
Ideally, donor morbidity following tendon transfer should be minimized. Hansen stated that concerns about loss of strength after transferring FHL are unfounded. 4 The purpose of this study was to evaluate objectively the functional deficit to the forefoot following FHL tendon transfer.
Materials and Methods
Seventeen patients were identified who had undergone an FHL tendon transfer for a chronic rupture or tendinosis of the TA. Inclusion criteria for this study were symptom duration of greater than 4 weeks in the involved foot prior to surgery and a normal contralateral foot, which was used as a control. One patient with a chronic TA rupture was excluded due to the presence of a bilateral diabetic sensory lower limb neuropathy.
Harvesting of FHL was performed through either a single- or a two-incision technique. The two-incision technique permits the harvesting of a longer tendon graft and tenodesis of the distal stump of FHL to FDL following FHL tendon harvest. We do not believe that the benefits of the double incision are sufficient to warrant potential increased surgical morbidity, so we changed to a single-incision technique after the first four patients in this study.
Clinical assessment comprised the American Orthopaedic Foot and Ankle Society (AOFAS) hallux metatarsophalangeal-interphalangeal scale 8 (pain 40 points, function 45 points, alignment 15 points, total 100 points), the SF-36 score, and a questionnaire regarding difficulties experienced with walking, running, stair climbing, and rising from a squatting position.
Pedobarography was performed using the Musgrave system (Musgrave Medical Ltd., Llangollen, UK). Each footplate had an area of 650 × 297 mm containing 2048 force-sensing resistors. The size of each sensor was 0.25 cm2. The scanning time was 18 ms and the operating frequency was 56 Hz. The parameters of this system conform to those previously recommended. 11,16 The footplate sensors were appropriately placed adjusting for the stride length and stance width of each patient. Following a series of “dummy” runs to ensure the patient was striking the footplates without breaking their normal stride, two sets of bilateral footprints were recorded. The peak pressures occurring on heel raise in three regions were recorded for analysis: the distal phalanx of the hallux, the first metatarsophalangeal joint (MTPJ) and the second MTPJ. We used the sum of the peak pressures over the distal phalanx of the hallux to represent FHL function at toe-off (Fig. 1). All peak pressure recordings from the distal phalanx were summated and the average total from the two footprints used as the index value.
Each footprint was composed of 64 images taken from heel strike to the end of toe-off. By forwarding through these images for each footprint we were able to confirm that the maximal peak pressures occurred following heel raise and not during the heel strike or stance phases of the gait cycle.
The normal contralateral footprint was used to identify normal peak pressures in the three areas and these were compared to the peak pressure readings from the corresponding areas on the operated side. Analysis of peak pressures over the first and second metatarsals was more problematic than that over the distal phalanx because pressure readings in these areas were not as discrete due to potential overlap of readings from closely adjacent areas of the forefoot. The normal footprint was used to identify the largest peak pressure sensor reading for the first and second MTPJs. Based around this single sensor reading, a 1.5 × 1.5 cm square of nine peak pressure recordings was summated to represent the first metatarsophalangeal joint peak pressure. Similarly, a 1.0 × 1.0 cm square of four peak pressure recordings was used to represent the second MTPJ peak pressure (Fig. 1). The same procedure was then applied to the operated footprint.
Areas of peak pressure recordings used to represent distal phalanx, first metatarsal, and second metatarsal peak loading.
The mean and standard deviation of the two readings for each foot were calculated. The paired Student t test was used to compare the means for the normal side with the means for the operated side for each of the three regions.
Results
The results for each of the 16 patients are listed in Table 1. The average hallux metatarsophalangeal-interphalangeal score was 97 (range, 85–100). Only two patients scored less than the maximum 100 points on this scale. One of these patients (score 85) had pre-existing hallux rigidus and one patient (score 88) had pre-existing hallux valgus. The average SF-36 score was 141.1 (SD 3.98). On the clinical questionnaire, none of the patients admitted to any impairment of walking, running (where appropriate), stair climbing, or rising from a crouched position due to weakness of the hallux. On clinical assessment, all patients demonstrated absence of active plantarflexion of the hallux at the IPJ with normal active plantarflexion of the first MTPJ.
Results of study
Pedobarography assessment showed a trend toward reduced total peak pressure loading of the distal phalanx in 12 of the 16 patients, one of whom (patient 1) had no pressure recording from the distal phalanx on the operated side. However this trend was not statistically significant on the paired Student t test (p = .082) for the number of patients studied. There was no significant difference in peak loading of the first (p = .1) or second (p = .39) MTPJs.
Five patients underwent a two-incision procedure with tenodesis of the FHL stump to FDL. These subgroups were too small to permit meaningful comparative statistical analysis, but there was no obvious difference between the two groups.
Discussion
The degree of morbidity following FHL tendon transfer has not been previously evaluated. Frenette and Jackson described three professional athletes who had returned to their respective sports at the same level following penetrating lacerations of FHL which had not been amenable to surgical repair. 2
All patients in this study clearly demonstrated clinical weakness of FHL function. Pedobarographic changes were also consistent with loss of FHL function. These findings did not translate into noticeable morbidity. The lack of evidence of transfer of pressure to the proximal forefoot in the FHL-deficient foot was reassuring.
The use of pedobarography has become increasingly established as a clinical research tool. 1,10 In this small number of subjects, we were not able to identify significant differences in forefoot loading patterns between the normal and operated foot, but a false-negative result cannot be excluded. The clinical use of pedobrography is still a relatively new science. Although certain guidance principles have been established, 5 -7,11 we had to make our own assumptions for the purposes of this study. First, we used the total of the peak pressure recordings over the distal phalanx of the hallux to represent FHL function. The contact area for the distal phalanx is discrete from the rest of the forefoot and was therefore relatively easy to delineate. For the first and second metatarsals, the contact area is not so discrete. To avoid contamination from adjacent sensor readings we chose to use only the central part of the contact area to represent loading in these two areas. By using the contralateral foot as a control, error was minimized. We found no evidence of transfer pressures to the first or second MTPJs following surgery. We accept that the use of the contralateral foot as a control may be criticized on the grounds that the contralateral foot may have accommodated to prevent limp. None of our patients were noted to have an antalgic gait at follow-up.
Passive extension of the IPJ of the hallux is restricted by the shape of the articulating surfaces. Therefore, hyperextension of the IPJ does not occur in the absence of FHL function. As the flexor hallucis brevis (FHB) contracts during push-off, active flexion occurs at the first MTPJ. As the IPJ is forced into extension, the joint locks, permitting some push-off pressure to be transmitted to the distal phalanx.
The patients in this study were not athletes and caution is advised in extrapolating our results to an athletic population. We conclude that the clinical morbidity following FHL tendon transfer is negligible.