First Metatarsophalangeal Arthrodesis: A Clinical, Pedobarographic and Gait Analysis Study

Transcription

1 FOOT & ANKLE INTERNATIONAL 2002 by the American Orthopaedic Foot & Ankle Society, Inc First Metatarsophalangeal Arthrodesis: A Clinical, Pedobarographic and Gait Analysis Study Paul Francis DeFrino, M.D.; James White Brodsky, M.D.; Fabian E. PolIo, Ph.D.; Stephanie J. Crenshaw, M.S.;Andrew D. Beischer M.D., FRACS, Dip Anat Palos Heights, IL, Dallas, TX, and Melbourne, NSW; Australia ABSTRACT INTRODUCTION This study investigated the results of first metatarsophalangeal (MTP) arthrodesis in terms of clinical outcome measures, plantar pressure distribution, and gait patterns. Ten feet in nine patients with severe hallux rigidus (HA) who underwent first MTP arthrodesis were studied. The preoperative evaluation included a subjective questionnaire, physical exam, AOFAS hallux score, radiographs and dynamic pedobarography (EMED). At followup (average 34 months) these were repeated, and gait analysis studies were obtained. Patients showed significant clinical improvement based on the subjective criteria. The mean AOFAS score improved from 38 preoperatively to 90 postoperatively. Postoperative EMED analysis showed restoration of the weightbearing function of the first ray, with greater maximum force carried by the distal hallux at toe-off. Kinematic and kinetic gait analysis from each patient's operative limb were compared to the unaffected contralateral limb and to age- and sex-matched healthy subjects. The kinematic data indicated a significantly shorter step length with some loss in ankle plantar flexion at toe-off on the fused side. The kinetic data indicated a reduction in both ankle torque and ankle power at push-off. Clinical results indicated effective pain relief and a high level of patient satisfaction, consistent with previous reports in patients with symptomatic Hallux The success of first metatarsophalangeal (MTP) arthrodesis for the treatment of severe hallux rigidus (HA) is well documented in the literature.4,7-10,14,17,21,22 The intent of this salvage procedure is to provide long-term relief of pain as well as preserve the length and stability of the first ray. Internal fixation methods have provided a reliable and reproducible means of fusion. Studies have reported fusion rates of greater than 95% with screw fixation,1,14,19 threaded pins18,21 and a small fragment plate.7,8 While excellent clinical results have been reported following first MTP arthrodesis, considerably less is known regarding the effects of fusion on plantar pressure distribution and gait. Mann and Oates18 observed their fusion patients, without the benefit of three-dimensional gait analysis, and found the overall gait pattern in the majority of their patients "to be most satisfactory." Betts2,3 and Duckworth11 used dynamic pedobarography to evaluate pathologic forefoot conditions and the effect that treatment had on those pressures. They reported that patients with HA who underwent a first MTP arthrodesis showed more normal peak pressure levels beneath their great toe postoperatively. This study was undertaken ~o quantify the effects of first metatarsophalangeal arthrodesis on gait and plantar pressures. Aigidus. MATERIALS AND METHODS Key Words: Gait Analysis; Arthrodesis; Pedobarography Corresponding Author: James White Brodsky, M.D. Clinical Professor, Orthopaedic Surgery, The University of Texas Southwestern Medical School and Director, Foot and Ankle Fellowship at Baylor University Medical Center 411 N. Washington Street Suite 7000 Dallas, TX Phone: (214) Fax: (214) jbrodsky@dallasortho.com Metatarsophalangeal A prospective clinical and pedobarographic study was performed on 10 feet in nine patients undergoing first MTP fusion for severe HA. Patients who had undergone previous lower-extremity surgery or had abnormal function of the ispilateral or contralateral ankle, knee and/or hip were excluded from the study. Of the 10 patients identified who fulfilled the inclusion 496 criteria and were initially studied, nine were available for the study. The study group consisted of five women and four men with an average age of 56 years (range, 38 to 72). The mean time to follow-up was 34 months (range, 26 to 44). A group of age- (average of 57.6 years) and sex-matched normal subjects were also studied with

2 Foot & Ankle InternationalNol. 23, No. 6/June 2002 FIRST MTP ARTHRODESIS 497 gait analysis to obtain data for comparison. One patient had bilateral MTP fusions and was then only compared to the control population. Operative Procedure Ten arthrodeses were performed on nine patients. The preoperative diagnosis in all patients was severe HR (Fig. 1 ). All patients had failed nonoperative measures including nonsteroidal anti-inflammatories and shoewear modifications. Surgical technique was identical in all patients and performed by the senior author (JWB). Straight dorsal approach was used followed by debridement of the joint surfaces and osteophytes and removal of subchondral bone. Rigid internal fixation was achieved by parallel 3.5-mm cortical screws directed from dorsal-proximal to distal-plantar. The screw heads were countersunk with a power bur to reduce their subcutaneous prominence and to prevent breaking out of the metatarsal head dorsally. After surgery, all patients were placed in a soft dressing and a rigid-bottom postoperative shoe and allowed to bear limited weight on the heel for two weeks, followed by weightbearing 'as tolerated until two months postoperatively. Clinical and radiographic fusion was achieved in all patients (Fig. 2). Postoperative radiographs showed that patients were fused in dorsiflexion with an average of 15.7 :t 6.9 angle between the first metatarsal and the proximal phalanx. One patient had a complication consisting of a deep-vein thrombosis (DVT) that required anti-coagulation therapy. No patients required hardware removal. Two of the nine patients underwent concurrent procedures. One of these patients had bilateral second hammertoe corrections. The second of these patients underwent ipsilateral second hammertoe correction and great toenail ablation. One of the nine patients had bilateral arthrodesis performed as separate procedures, five months apart. All nine patients underwent preoperative clinical, radiographic and dynamic pedobarographic analysis. As part of the follow up study, these patients returned for repeat clinical, radiographic, as well as three-dimensional dynamic pedobarography gait analysis study. Fig. 1: AP radiograph depicting hallux rigidus preoperatively. Clinical Assessment Each patient was interviewed both preoperatively and postoperatively regarding the specifics of pain location and duration, shoewear type and use of in-shoe orthoses. Range of motion, pain, and function were assessed using the AOFAS Hallux Metatarsophalangeal-Interphalangeal Scale.16 This scale was modified by exclusion of the 10 points attributed to first MTP motion, as this was eliminated by the arthrodesis. The maximum total score of 90 points was then normalized to a score of 100. Weightbearing anteroposterior and lat- Fig. 2: AP radiograph depicting fused MTP joint two years postoperatively. eral radiographs and non-weightbearing oblique views of the foot were taken preoperatively and postoperatively with a standardized technique. Preoperative and postoperative radiographs were measured for intermetatarsal angle (IMA), hallux valgus angle (HVA) and Interphalangeal/Fitzgerald'2 score.

3 498 DEFRINO ET AL, Foot & Ankle InternationalNol. 23, No. 6/June 2002 Pedobarographic Analysis Plantar pressures were recorded using the EMED-SF system (Novel Gmbh, Munich, Germany). This system has 2048 capacitive resistance sensors arranged in a matrix over a recording area of 36 cm x 18 cm, with a sensor density of 2 sensors/cm2. The sampling frequency of the EMED-SF was 71 samples/second. For each patient, age, weight and height information was obtained. Patients were then asked to walk barefoot over the pressure platform using the two-step protocol outlined by Cavanaughand Ulbrecht.5 Based on this protocol, patients were instructed to take only a single step before stepping on the pressure plate with the study foot. A successful trial occurred when the patient stepped in the center of the pressure plate. Three trials were obtained both preoperatively and postoperatively for each study patient, and the results were averaged. The pressure distribution of each foot was divided into the heel, midfoot, and forefoot (consisting of 1 st, 2nd, 3rd, 4th and 5th metatarsal heads), hallux, second toe, and the lateral toe areas or "masks." Maximum force and pressure parameters were recorded for each area and normalized to that patient's body mass in kilograms. Gait Analysis The patients were asked to wear shorts and a tee shirt allowing for maximum exposure of their legs and pelvic area during gait analysis testing. Both the operated and unoperated sides were tested for each patient. A group of nine healthy, age- and sex-matched controls were also tested for comparison. Kinematic parameters were collected using a 5-camera Motion Analysis System (Motion Analysis Corporation, Inc., Santa Rosa, CA) recording at 60 Hz. Ground reaction forces were collected at 1,000 Hz with an AMTI OR6-5 force platform (Advanced Medical Technology, Inc., Newton, MA) embedded in the center of a 10-meter walkway. Inverse dynamics were used to derive kinetic information from the kinematic and the ground reaction force data. Reflective markers were secured to anatomical locations with adhesive tape using the Helen Hayes Marker configuration.15 Markers were placed bilaterally over the anterior superior iliac spine, lateral knee epicondyle, lateral malleolus, on the forefoot over the head of the third metatarsal and the posterior aspect of the heel at the same level as the marker over the third metatarsal head. Short 11-cm wands, with markers attached to the ends, were also secured laterally to the mid-thigh and mid-shank regions. Markers were also placed over the medial malleoli and medial femoral epicondyles for collection of a static trial just prior to testing to identify the centers of the ankle and knee joints, respectively. The patients were then asked to walk barefoot down the 1O-meter walkway at a self-selected speed. Data were collected until five "good" trials were recorded for each limb. A trial was considered "good" if the subject struck the force platform with only one foot, and this foot was completely within the boundaries of the force platform. The patients were allowed rest breaks whenever they deemed them necessary. The collected data was normalized to the gait cycle and the results of six cycles were averaged for statistical analysis. Statistics The pre-op and post-op AOFAS score and radiographic values were compared using a Student's paired t-test. The pedobarographic data for the operated and nonoperated limbs were also analyzed with a Student's paired t-test. All gait analysis data (temporalspatial, kinematic and kinetic) were analyzed using a one-way analysis of variance (ANOVA) for the three conditions consisting of 1.the operated limb, 2. the nonoperated limb of the study patients, and 3. the group of normal controls. RESULTS Clinical Review The average preoperative AOFAS Hallux MTP-IP scale was 38 (range, 20 to 62). The score improved to an average of 90 (range, 74 to 100) at the most recent follow-up, which was significant at p= Of the 10 arthrodesis procedures performed, four patients reported no pain, and minimal/occasional pain was reported in the other six. In reviewing the questionnaires it appears that most patients describe occasional mild symptoms in the operated area, defined as a diffuse but mild ache with extended periods of walking. This discomfort was described as being located over a wide area from the fusion site to the midfoot, and it was experienced intermittently, but only when patients were standing or walking for long periods of time. Shoewear problems were present in three women who reported difficulty wearing high fashion shoes. No limitations in activity were reported in six patients, while limitations in recreational activity were present in three. All patients who were participating in running recreation preoperatively were able to return to those activities. All patients stated that they had "great" improvement as compared to their preoperative condition. None required postoperative in-shoe orthoses. When asked if they would undergo this surgery again, eight of the nine responded "yes" and one responded "yes with reservations." This one patient suffered a postoperative DVT and her reservations were in regards to the postoperative difficulties

4 Foot & Ankle InternationalNol. 23, No.61 Ine 20C related to that condition. All patients in the study group were examined both standing and sitting, and asked to walk to assess for evidence of a visible limp. On physical examination, all nine patients had clinical evidence of solid fusion at the arthrodesed joint. Satisfactory alignment of the hallux was noted in all patients without evidence of malunion in varus/valgus or in dorsiflexion/plantarflexion. No individuals in the study group reported pain beneath the first MTP, hallux or lesser metatarsal heads. Additionally, no pain was reported at the hallux interphalangeal joint with either passive range of motion or during ambulation. Gait was observed visually with the patient walking barefoot and no patient in the study group showed clinical evidence of a limp. Radiographic Review The preoperative mean IMA, HVA and IP/Fitzgerald scores (Table 1) were 100 (range, 9 to 11), 11 (range, 7 to 16), and 1 (range, 1 to 2) respectively., these mean values were measured to be 10 (range, 9 to 12), 6 (range, 0 to 10), and 2 (range, 1 to Table 1 : Radiographic values measured pre- and postoperatively IMA HVA Fitzgerald/lP Radiographic Value 10 ::t :f: :t: ::t :f: :t: * * significant at p<o.os Table 2: Pedobarographic Contact Area (cm2) Maximum Force (N/kg) Peak Pressure (N/cm2okg) data of the forefoot. 1 st 2nd Metatarsal Head Metatarsal Head FIRST MTP ARTHRODESIS 499 3). The preoperative and postoperative values for the IMA and IP/Fitzgerald were not significantly different from each other. However, the hallux valgus angle was significantly reduced postoperatively (P=O.OO5). None of the patients in the study gave a history at follow-up of any new trauma or new symptoms relative to the proximal joints of the ankle and foot. In addition, there were no physical examination or radiographic changes in the remaining joints of the forefoot, midfoot or hindfoot in any of the patient's follow-up clinic visits and radiographs. Pedobarographic Data None of the pedobarographic measurements in the metatarsal head regions were different between preoperative and postoperative evaluations (Table 2). The maximum force and peak pressure under the hallux significantly increased between the preoperative and postoperative evaluations (p=0.006 and p=0.004, respectively, see Table 3). The contact area under the entire foot and under the hallux significantly increased between the preoperative and postoperative evaluations (p=0.011 and 0.024, respectively, see Table 3). Subjects also seemed to walk approximately 6% faster over the pedobarographic plate at the postoperative visit, which was not statistically significant but noteworthy. Gait Analysis Gait analysis data for the fused limb, the contralateral "normal" limb, and the limb of the age- and sexmatched controls were compared using a one-way ANOVA. In comparison to the patient's normal limb and the healthy control, the patient's operated limb was found to have a significantly decreased 3rd 4th Sth step length when nor- Metatarsal Metatarsal Metatarsal malized for height, Head Head Head when stepping off that side. The average decrease in step length was 2.4 cm (range, 5.5 cm to cm). The kinematic data also revealed an average of 4 less of ankle plantarflexion at the end of stance phase during toe-off on the fused side as com pared to the normal ankle; however, it was not statistically significant (p=0.077).

5 500 DEFRINO ET AL. Foot & Ankle InternationalNol. 23, No. 61June 2002 This was demonstrated with no significant change in the hip or knee kinematics. In evaluating the ankle kinetics in the sagittal plane, the study found a reduction in both plantarflexor moment and power during toe-off as compared to the nonoperative and the healthy control limbs. Figure 3 illustrates the comparison of ankle joint power between the operative and nonoperative limb in a sample patient. Kinetics of the hip and knee were not significantly different among the three conditions. DISCUSSION TABLE 3: Pedobarographic data for total foot and the toes Contact Area (cm2) Maximum Peak Pressure Force (N/kg) (N/cm2.kg) Total Hallux 2nd Toe *, * * * * * rd, 4th and Sth Toes This study objectively evaluated I. significant at p<o.os the function of first MTP arthrode- ~ sis within the foot in terms of Ankle Power weightbearing function and pressure distribution, as well as the effect on gait. The patients in this 4 study demonstrated subjective and objective clinical findings that 3 first MTP arthrodesis improves Fused Limb Healthy Limb function, reduces pain and allows 2 most patients to resume normal biomechanical function and return 1 to recreational activities. The AOFAS scores reflected this high O - degree of subjective satisfaction and objective improvement. The AOFAS Clinical Scoring system, ~ while widely utilized, reflects the % Gait Cycle current absence of a comprehen- IFig. 3: Graph of ankle power from a sample patient comparing the fused limb to the healthy sive dated and scoring universally system accepted for the vali- foot l limb and ankle. Despite this limitation of the study, it does provide some guidelines in evaluating patient improvement. Many authors have surmised that first MTP fusion is successful because it re-establishes the weightbearing role of the first ray.6,13,19 Betts2,3 found that those patients with HA had abnormally high peak pressures beneath the hallux, but due to early toe-off, the pressure was present for a only a short duration. These preoperative pedobarographic findings reflect the common clinical findings seen in HA. Stiffness and pain at the first MTP joint restrict normal motion at third rocker resulting in initial high peak pressures followed by early toe-off. Following first MTP arthrodesis, these same patients were thought to be more likely to have normal peak pressures beneath the hallux. However, even with a solid arthrodesis, loading of the hallux occurred over a shorter duration than in normal subjects. The authors explained this finding by the fused position of the hallux commonly being in a slight dorsiflexed position. As the forefoot rolled over into third rocker, a hallux fused in dorsiflexion would contact the floor for a shorter period of time. In our study, preoperative and postoperative dynamic pedobarography studies followed an established protocol.5 The findings demonstrate that arthrodesis of the first MTP joint allows greater weight to be carried across the hallux during third rocker. This finding confirms the work of previous studies indicating that first MTP

6 Foot & Ankle InternationalNol. 23, No. 6/June 2002 FIRST MTP ARTHRODESIS 501 arthrodesis in patients with severe HA reestablishes the weight-bearing role of the f(rst ray.1,20 We expected to find reduced loads under the lesser toes at the postoperative pedobarographic study, since the hallux was bearing more weight. However, this was not the case, and in most of our patients, the force and pressure in the lesser toes increased postoperatively. This increased pressure even occurred despite increased contact area under those regions postoperatively. We surmise that the increased loads were attributable to an increase in the velocity postoperatively and subsequent loading throughout the entire foot. To explain why patients, despite a solid arthrodesis, load the hallux for a shorter period of time, we utilized three-dimensional gait analysis. We observed that patients had a shorter step length when pushing off from the side with the MTP fusion. We additionally noted a reduction in the plantarflexion angle of the ankle at toeoff on the fused side, which was accomplished with no change in the hip or knee angle. This appears to be a compensatory action in order to keep the vertical motion of the center of gravity of the body to a minimum. We also noted a significant decrease in ankle plantarflexion moment and ankle power at toe-off. This is most likely a combination of the loss of the motion of the first ray, which contributes to generating power at pushoff. Secondly, the reduction in ankle plantarflexion would also reduce the amount of power that can be generated in that portion of stance phase. Hip and knee kinematics and kinetics seemed to be undisturbed by the MTP fusion when compared to the nonoperated side and the group of normal controls. The sensitivity of three-dimensional gait analysis uncovered alterations in gait that had been difficult to perceive by both patient and practitioner. While these findings are statistically significant, the question remains regarding their clinical significance. The majority of subjective and objective clinical data regarding first MTP arthrodesis point toward a high degree of patient satisfaction with good to excellent functional outcomes. Our findings, although based on small numbers, reflect the same level of patient satisfaction seen in larger studies. Our conclusion is that the changes observed through gait analysis represent compensatory changes made by the study patients to accommodate a foot that is made both more rigid and effectively longer (during push-off) by the arthrodesis. The impact of rigidity and length on gait is most significant during the third rocker phase of gait. In normal gait, the MTP joint dorsiflexes allowing the ankle to rollover the foot in a larger arc, in turn generating greater torque. A functioning MTP joint also allows the hallux to contact the floor for a longer period without generating exces- Sively high peak pressures. Following arthrodesis, the patient must adapt to a rigid hallux that in effect increases the length of the lever arm of the foot. As patients push off with the longer lever arm, a longer step length is made. Without a compensatory decrease in ankle plantarflexion, a vertical displacement in the body's center of gravity would occur. This would manifest as a lurch with each step taken off the fused foot. The degree to which patients were able to compensate is shown by the fact that none of the patients in the study group walked with a visible limp. In conclusion, these data show that first MTP fusion provides excellent pain relief and establishes a more normal plantar pressure pattern in the foot. Fusion does not, however, re-establish a normal gait pattern. Data from gait kinetics and kinematics show that patients appear to be compensating well for the increased rigidity of their fused hallux. These gait changes are quite subtle and do not appear to compromise the overall satisfaction level reported by patients in the study group. A prospective study has been initiated as a result of this preliminary data, which will examine the same parameters both preoperatively and postoperatively in a larger patient population. REFERENCES 1. Beauchamp, CG; Kirby, T; Audge, SA; et al: Fusion of the first metatarsophalangeal joint in forefoot arthroplasty. Clin Orthop , Betts, AP; Franks, CI; Duckworth, T: Foot Pressure Studies: Normal and Pathologic Gait Analyses. In Jahss MH (ed): Disorders of the Foot and Ankle, ed Second. Philadelphia: W. B. Saunders, 1991, Voll, pp Betts, AP; Franks, CI; Duckworth, T; et al: Static and dynamic foot-pressure measurements in clinical orthopaedics. Med Bioi Eng Comput 18:674-84, Bonney, G; MacNab, I: Hallux Valgus and Hallux Rigidus: A Critical Survey of Operative Results. The Journal of Bone and Joint Surgery 348: , Caganagh, PA; Ulbrecht, JS: Clinical plantar pressure measurement in diabetes: rationale and methodology. The Foot 4: , Chana GS; Andrew TA; Cotterill CP: A simple method of arthrodesis of the first metatarsophalangeal joint. J Bone Joint Surg Br 66:703-5, Coughlin, MJ: Arthrodesis of the first metatarsophalangeal joint. Orthop Rev 19:177-86, Coughlin, MJ: Arthrodesis of the first metatarsophalangeal joint with mini-fragment plate fixation. Orthopedics 13: , Coughlin, MJ: Conditions of the Forefoot. In DeLee JC, Drez D (eds): Orthopaedic Sports Medicine: Principles and Practice. Philadelphia: W. B. Saunders, 1994, Vo12, pp Coughlin, MJ; Abdo, AV: Arthrodesis of the first metatarsophalangeal joint with Vitallium plate fixation. Foot Ankle Int 15:18-28, Duckworth, T; 8etts, AP; Franks, CI; et al: The measurement of pressures under the foot. Foot Ankle 3:130-41, Fitzgerald, JA: A review of long-term results of arthrodesis of the first metatarso-phalangeal joint. J Bone Joint Surg [8r] 51 :488-93, Henry, AP; Waugh, W; Wood, H: The use of footprints in assess-

7 502 DEFRINO ET AL. Foot & Ankle InternationalNol. 23, No. 6/June 2002 ing the results of operations for hallux valgus. A comparison of Keller's operation and arthrodesis. J Bone Joint Surg [Br] 57:478-81, Johansson, JE; Barrington, TW: Cone arthrodesis of the first metatarsophalangeal joint. Foot Ankle 4:244-8, Kadaba, MP; Aamakrishnan, HK; Wootten, ME: Measurement of lower extremity kinematics during level walking. J Orthop Res 8:383-92, Kitaoka, HB; Alexander, IJ; Adelaar, AS; et al: Clinical rating systems for the ankle-hindfoot, midfoot, hallux, and lesser toes. Foot Ankle Int 15:349-53, Mankey, M; Mann, A: Arthrodesis of the first metatarsophalangeal joint using a dorsal plate. In 7th Annual Summer Meeting of the American Orthopaedic Foot and Ankle Society, Boston, MA 18. Mann, AA; Oates, JC: Arthrodesis of the first metatarsophalangeal joint. Foot Ankle 1: , Moynihan, FJ: Arthrodesis of the metatarso-phalangeal joint of the great toe. J Bone Joint Surg [Br] 49:544-51, Samnegard, E; Turan, I; Lanshammar, H: Postoperative evaluation of Keller's arthroplasty and arthrodesis of the first metatarsophalangeal joint using the EMED gait analysis system. J Foot Surg 30:373-4, Smith, N: Hallux Valgus and rigidus treated by arthrodesis of the first metatarsophalangeal joint. J Br Med 2: , Southgate, JJ; Urry, SA: Hallux rigidus: the long-term results of dorsal wedge osteotomy and arthrodesis in adults. J Foot Ankle Surg 36:136-40; discussion 161, 1997.