Kinematic Changes of the Foot and Ankle in Patients with Systemic Rheumatoid Arthritis and Forefoot Deformity

Transcription

1 Kinematic Changes of the Foot and Ankle in Patients with Systemic Rheumatoid Arthritis and Forefoot Deformity Michael Khazzam, 1 Jason T. Long, 1,2 Richard M. Marks, 2 Gerald F. Harris 1,2 1 Orthopaedic and Rehabilitation Engineering Center, Marquette University/Medical College of Wisconsin, 9200 West Wisconsin Avenue, Milwaukee, Wisconsin Department of Orthopaedic Surgery, Medical College of Wisconsin, Milwaukee, Wisconsin Received 15 February 2005; accepted 10 August 2006 Published online 1 December 2006 in Wiley InterScience ( DOI /jor ABSTRACT: Minimal published data exist characterizing the effect of rheumatoid arthritis of the forefoot (RA) on multi-segmental gait kinematics. The purpose of this study was to examine specific changes in segmental foot motion in patients with RA as compared to persons without foot/ankle pathology. This was a cross-sectional, descriptive study consisting of 22 preoperative adult patients (29 feet) diagnosed with RA and 25 adult patients with no known foot pathology (Control). All RA patients were evaluated by the same orthopaedic surgeon. This group consisted of 20 women and 2 men with a mean age of 54 years (range, years). The Control cohort consisted of 13 men and 12 women with a mean age of 41 years (range, years). Foot and ankle motion data for the RA population were obtained using a 15-camera Vicon Motion Analysis System (Vicon Motion Systems, Inc., Lake Forest, CA). Anterior posterior, lateral, and modified coronal radiographic views were obtained to relate marker position to underlying bony anatomy. Temporal and three-dimensional kinematic parameters were obtained via the 4-segment Milwaukee Foot Model. Quantitative comparisons of range of motion values during the seven phases of gait were made between RA and Control ankles using unpaired nonparametric methods. The RA group showed significant differences ( p < 0.001) as compared to Controls with prolonged stance time, shortened stride length, increased cadence, and a walking speed that was 80% of Control. Overall, kinematic data in the RA cohort showed significant differences ( p < 0.001) in motion for tibial, hindfoot, and forefoot motion as compared to Controls. The effect of RA on segmental foot motion is poorly understood. This study characterized the effect that RA has on motion about the foot and ankle during gait, providing insight into this pathology to improve quantitative assessment, treatment planning, and rehabilitative care. ß 2006 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. J Orthop Res 25: , 2007 Keywords: rheumatoid arthritis; gait; kinematics; foot and ankle; Milwaukee foot model INTRODUCTION Correspondence to: Jason Long (Telephone: ; Fax: ; jlong@mcw.edu) ß 2006 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. Rheumatoid arthritis (RA) is a progressive systemic autoimmune disorder that affects the musculoskeletal system. Typically, forefoot symptoms present early, and as the disease progresses, so does their severity. 1 The most common forefoot deformities are hallux valgus, subluxation or dislocation of the lesser metatarsophalangeal (MTP) joints, and fixed hammer toe or claw toe deformities of the lesser toes. 1 4 These changes result from chronic MTP joint inflammation leading to joint capsular distension followed by loss of ligament integrity. 3,4 Soft tissue instability, articular cartilage damage, and subchondral bone resorption lead to these common RA forefoot deformities. 2,4 Frequently, with dorsal dislocation of the lesser toes, a pistoning effect occurs, which coupled with painful callosities and stretching out of the foot pad leads to a decreased area for weight-bearing. 1 4 These painful forefoot deformities result in alteration of the normal motion of the foot and ankle during ambulation. To examine how the pathology caused by RA affects foot and ankle motion, the normal motion of the foot and ankle during gait must be understood. Ankle motion is commonly referred to in terms of the three rockers of foot and ankle motion The first rocker occurs during the loading response; at heel strike (0% stride), the ankle undergoes rapid plantar flexion that peaks at 7% of the gait cycle, concluding with the foot flat on the floor (first rocker or initial double limb support). Next, the ankle progressively dorsiflexes until roughly 30% of the JOURNAL OF ORTHOPAEDIC RESEARCH MARCH

2 320 KHAZZAM ET AL. cycle (second rocker or single limb support), then as the heel lifts, active plantarflexion continues to a maximum at toe-off (258) or 60% of the cycle (third rocker or double limb support). The second rocker occurs during mid-stance; the third rocker occurs during terminal stance. During terminal stance, the foot acts as a rigid lever for toe-off. Acting through the windlass mechanism, dorsiflexion of the toes helps to stabilize the plantar arch of the foot by shortening the plantar aponeurosis, resulting in elevation of the arch, passive inversion of the heel, and increased stability. 7,10 This action also converts the normally flexible foot into a rigid lever for toe-off. 10 Finally, during early swing phase, the ankle rapidly dorsiflexes, allowing the foot to clear the floor and positioning it for the next heel strike. 7,8,10,12 Changes in foot/ankle biomechanics secondary to RA have been studied by several investigators, but the majority modeled the foot and ankle as a single rigid segment. Siegel et al. evaluated four patients with RA using a single-segment foot model, finding a prolonged period of foot-flat with diminished plantarflexion and heel rise at toe-off. 13 This data set included RA patients with various foot deformities and did not focus on patients with isolated forefoot involvement; of the patients tested, two were described as having somewhat isolated forefoot arthritis. Platto et al. evaluated temporal-spatial parameters during gait in 31 patients with RA, and reported decreased walking speed, cadence, and stride length. 14 O Connell et al. compared the gait of 10 RA patients with 7 normals, obtaining temporal-spatial data with a foot switch system and kinematic data via a single-segment foot model. 15 They reported that RA patients had decreased kinematic and temporal values as compared to normal, but the study s kinematic analysis of ankle motion was limited to the sagittal plane. Investigations of plantar pressure have also provided insight into the effects of RA on foot and ankle biomechanics. In a plantar pressure assessment of 124 patients with foot deformities secondary to RA, Minns and Craxford found pressures under the metatarsal head 2 3 times higher than that of normals. 16 A lateral shift in pressure center to the 4th and 5th metatarsals just before toe-off was compared to the normal pattern of loading at the medial aspect of the foot (head of the 1st metatarsal). More recently, Otter et al. investigated 25 patients with chronic RA during ambulation over a pressure measurement platform. 17 No significant findings related to magnitudes of peak pressures were found, but the RA group exhibited significant differences in the temporal aspects of loading, with prolonged loading periods and increased force-time integrals. The authors suggested that these temporal characteristics of pressure distribution may be a better means of describing the RA foot than pressure magnitude. The use of a multi-segmental foot model to examine foot and ankle motion during gait has gained in popularity, and several authors have used them to analyze foot and ankle pathologies However, reports of multi-segmental motion in the RA population are limited. Woodburn et al. evaluated 11 RA patients with clinical foot impairments and deformity. 20 Kinematic and temporal parameters were obtained using a 4-segment foot model 22 that showed 3D motion of the tibia, hindfoot, and forefoot, and only the sagittal plane of motion of the hallux. The authors found evidence of forefoot weight-bearing avoidance via delayed onset and decreased hindfoot plantar flexion during pre-swing, indicating loss of the third rocker. Johnson et al. described normal motion using the Milwaukee Foot Model (MFM), a 4-segment rigidbody model validated for both adults 8,23,24 and children. 25 The 4 segments in the model are tibia/ fibula, hindfoot, forefoot, and hallux. Three-dimensional kinematics of these segments were assessed in five normal adults; surface marker data were referenced to bony landmarks via anteroposterior, lateral, and modified coronal hindfoot alignment 9 radiographs. The ways in which forefoot RA affects gait are not well understood. We aimed to describe quantitatively the multi-segmental kinematic changes about the foot and ankle during gait as a result of RA forefoot deformity. This information provides a baseline to understand better the dynamics of this deformity and measure disease progression, as well as to provide a reference of comparison for outcome after surgical intervention. We hypothesized that foot ankle range of motion (ROM) measured in an RA population would differ from that measured in a population without pathology and that these differences would be observed in multiple segments and planes. Specifically, we expected reduced sagittal plane motion of the forefoot during gait. METHODS This is a cross-sectional, descriptive study consisting of 22 preoperative adult patients (29 feet) diagnosed with forefoot RA and 25 healthy normal patients with no known ankle pathology (Control, Table 1). All subjects gave informed consent under authorization by the IRB. All RA patients were evaluated by the same surgeon (RMM) and tested preoperatively between the years JOURNAL OF ORTHOPAEDIC RESEARCH MARCH 2007 DOI /jor

3 KINEMATIC CHANGES OF THE FOOT AND ANKLE 321 Table 1. Subject Demographics Characteristics Control RA No. of subjects (29 feet) Age (average) 41 years (range, 27 73) 54 years (range, 17 76) Sex 13 men/12 women 2 men/20 women Affected side (R/L) 11 R/18 L Inclusion criteria for the RA cohort were: independent ambulators with pain and deformity of the forefoot and a documented RA diagnosis (as established by the American College of Rheumatology 26 ). A patient is classified as having RA if at least four of the following criteria are met: morning stiffness, arthritis of 3 joint areas, arthritis of the hand joints, symmetric arthritis, rheumatoid nodules, serum rheumatoid factor, and radiographic changes. Patients were excluded based on: previous surgery for their forefoot deformity; pain or documented arthritis in any other joint in the affected extremity being tested; and medical contra-indication preventing surgery. The Control population was drawn from a previously collected database. Following marker placement, the subject stood comfortably in the center of the laboratory walkway, using the previously acquired tracing to establish foot position. A static trial (3-s duration) was collected to record the marker positions. The subject was then instructed to ambulate along the walkway (6-m length) at a selfselected speed. During each trial, video data were captured at 120 frames per second (fps). A minimum of five trials were collected and approved for further analysis, subject to clinical screening. The subject was instructed to rest whenever necessary, for as long as needed. No subjects indicated undue discomfort or fatigue. Clinical Assessment Milwaukee Foot Model Temporal-spatial and 3D kinematic foot and ankle motion data were obtained using a 15-camera Vicon 524 Motion Analysis System (Vicon Motion Systems, Inc., Lake Forest, CA). The system uses infrared strobes to illuminate reflective markers placed on key anatomic landmarks and measures the marker positions within the global laboratory coordinate system. Marker position information was analyzed using the 4-segment MFM (Fig. 1), previously described. 8,23 25 The MFM calculates kinematics of 4 segments (tibia, hindfoot, forefoot, and hallux) in the sagittal, coronal, and transverse planes. Segment motion is measured relative to the next most proximal segment, with tibial segment motion measured relative to the global coordinate system. Marker motion is referenced to the underlying bony anatomy with rotational offsets based on weightbearing plane AP, lateral, and modified hindfoot coronal alignment (Milwaukee view) radiographs. 9 The model also calculates temporal-spatial parameters, including stride length, cadence, stance duration, and walking speed. Each subject stood on a sheet of cardboard in a comfortable position while the positions of his/her feet were traced. This tracing was assumed to represent a standing foot position and was subsequently used during both motion analysis testing and radiograph collection to standardize foot position. Nine reflective spherical markers (16-mm diameter) were placed on bony landmarks on the tibia, ankle, and foot to define each MFM segment. Due to its limited size, the hallux was marked with an orthogonal triad with adequate marker separation. The patients were interviewed regarding pain, function, level of activity, and problems with footwear. A physical examination of the feet was performed. The American Orthopaedic Foot and Ankle Society (AOFAS) hallux and lesser toes scores were used for clinical assessment. 27 These assessment tools provide information Figure 1. Milwaukee Foot Model (MFM). Superior, posterior, and lateral views of foot marker placement. DOI /jor JOURNAL OF ORTHOPAEDIC RESEARCH MARCH 2007

4 322 KHAZZAM ET AL. about pain, activity level, deformity, motion, alignment, stability, and footwear issues. A score of points is considered excellent; points, good; points, average; points, fair; and <60 points, poor. Radiographs A unique radiographic view was used to demonstrate true alignment of the calcaneus relative to the tibia (Fig. 2). The subject stood in a natural position and tracings of both feet were drawn on the cardboard. A line was drawn that bisected the silhouette of the 2nd toe and the heel. Another line was drawn across the tips of the toes perpendicular to the foot line. The foot tracings were then cut along this line. The tracing allowed reproducibility of foot position and assured that the foot was perpendicular to the x-ray plate. The subject stood on an x-ray stand with both feet positioned on the tracing. The beam was then angled to the floor and a posterior x-ray was taken with the cassette at 908 to the floor. To measure calcaneal/tibial alignment, an ellipse was superimposed over the posterior calcaneal tuberosity. The angle between the tibial axis and the semi-major axis of the ellipse was defined as the coronal hindfoot alignment. Clinical Radiographs The radiographic assessment performed preoperatively included evaluation of the hallux valgus angle, the first intermetatarsal angle, the angle of dorsiflexion at the first metatarsophalangeal joint, and the number of subluxed and/or dislocated lesser toes. Statistics ROM values were calculated for each subject during each of the seven gait phases, designated as: load response (0 16% stance), mid-stance (16 48% stance), terminal stance (48 81% stance), pre-swing (81 100% stance), initial swing (0 32% swing), mid-swing (32 66% swing), and terminal swing (66 100% swing). 28 Temporal-spatial parameters (cadence, stride length, stance duration, and walking speed) were also calculated. In each plane, we tested the hypothesis that ROM measures were the same at each phase using unpaired nonparametric methods (Mann Whitney U-test). To adjust for multiple tests over the set of seven test points, a Bonferroni correction was used to achieve a familywise 5% overall error rate. We made all comparisons at p < (slightly less than the level needed to achieve the desired overall error rate). RESULTS Comparison of temporal parameters between RA and Control subjects showed significant differences (Table 2). Overall walking speed was reduced from 1.12 m/s to 0.89 m/s. RA walking Figure 2. Modified coronal radiograph of hindfoot (Milwaukee view). speed was found to be 80% of the Control walking speed. Stance time in the RA patients was prolonged from 62% to 66%, stride length was shortened from 1.28 m to 0.96 m, and cadence was increased from steps/min to steps/min. Differences in walking speed, stance duration, and stride length were statistically significant. JOURNAL OF ORTHOPAEDIC RESEARCH MARCH 2007 DOI /jor

5 KINEMATIC CHANGES OF THE FOOT AND ANKLE 323 Table 2. Temporal Spatial Parameters (RA vs. Control) Parameter Control RA p Stride length (m) < Cadence (steps/min) Stance duration (%) < Walking speed (m/s) < The RA population exhibited a generally decreased ROM throughout all foot segments (tibia, hindfoot, forefoot, and hallux) as compared to Controls. Tibial motion of the RA population was decreased from mid-swing through terminal swing phase in the sagittal, coronal, and transverse planes (Fig. 3; Table 3). The hindfoot of the RA population demonstrated decreased motion in the sagittal plane at load response as well as from pre-swing through initial swing phase compared to Controls (Fig. 4; Table 4). Eversion ROM decreased significantly in the coronal plane from load response through terminal stance and during initial swing and terminal swing phases. The hindfoot was more externally rotated in the RA group, but decreased rotation ROM occurred compared to Controls throughout the gait cycle. In the RA population, the forefoot showed significant loss of motion in the sagittal, coronal, and transverse planes as compared to Controls (Fig. 5; Table 5). Sagittal motion (dorsiflexion) decreased significantly from load response through mid-stance and from pre-swing through terminal swing. Forefoot motion in the coronal plane demonstrated loss of forefoot varus from midstance through terminal swing. The forefoot also demonstrated a significant decrease in transverse plane motion (abduction) throughout the gait cycle. The hallux segment displayed a large degree of valgus in the transverse plane throughout the cycle as a result of the fixed hallux valgus deformity, a characteristic of RA of the forefoot (Fig. 6; Table 6). Sagittal motion of the hallux was decreased from terminal stance through pre-swing. The average AOFAS Hallux score was 26, and the average Lesser Toes score was 28, indicating that the RA group had severe pain and a low level of function. All 22 patients (29 feet) had longstanding RA with hallux valgus deformity, Figure 3. Tibia segment kinematics and range of motion (mean 1 SD) in sagittal, coronal, and transverse planes for RA vs. Control. (* indicates statistically significant difference in ROM at p < 0.001). DOI /jor JOURNAL OF ORTHOPAEDIC RESEARCH MARCH 2007

6 324 KHAZZAM ET AL. Table 3. ROM of Tibia Segment in Three Planes for RA vs. Control a Plane Phase Control (8) RA (8) p* Sagittal Load response Mid-stance Terminal stance Pre-swing Initial swing Mid-swing < X Terminal swing < X Coronal Load response Mid-stance Terminal stance Pre-swing X Initial swing Mid-swing X Terminal swing < X Transverse Load response Mid-stance < X Terminal stance Pre-swing Initial swing Mid-swing < X Terminal swing < X Heel strike Toe-off a Mean 1 SD. *X denotes statistical significance at p < Figure 4. Hindfoot segment kinematics and range of motion (mean 1 SD) in sagittal, coronal, and transverse planes for RA vs. Control. (* indicates statistically significant difference in ROM at p < 0.001). JOURNAL OF ORTHOPAEDIC RESEARCH MARCH 2007 DOI /jor

7 KINEMATIC CHANGES OF THE FOOT AND ANKLE 325 Table 4. ROM of Hindfoot Segment in Three Planes for RA vs. Control a Plane Phase Control (8) RA (8) p* Sagittal Load response X Mid-stance Terminal stance Pre-swing < X Initial swing < X Mid swing Terminal swing Coronal Load response X Mid-stance < X Terminal stance < X Pre-swing Initial swing < X Mid-swing < X Terminal swing < X Transverse Load response < X Mid-stance < X Terminal stance < X Pre-swing X Initial swing < X Mid-swing < X Terminal swing < X a Mean 1 SD. *X denotes statistical significance at p < plantar callosities, and axial malalignment of the lesser toes (subluxation or dislocation). All subjects exhibited at least one of hammertoe, claw-toe, or cock-up deformities. Three of 29 feet had bunionette deformity of the 5th metatarsal. Radiographic evaluation revealed bony angles characteristic of an RA population (Table 7). Figure 5. Forefoot segment kinematics and range of motion (mean 1 SD) in sagittal, coronal, and transverse planes for RA vs. Control. (* indicates statistically significant difference in ROM at p < 0.001). DOI /jor JOURNAL OF ORTHOPAEDIC RESEARCH MARCH 2007

8 326 KHAZZAM ET AL. Table 5. ROM of Forefoot Segment in Three Planes for RA vs. Control a Plane Phase Control (8) RA (8) p* Sagittal Load response < X Mid-stance < X Terminal stance Pre-swing X Initial swing < X Mid-swing < X Terminal swing < X Coronal Load response Mid-stance < X Terminal stance < X Pre-swing X Initial swing < X Mid swing < X Terminal swing < X Transverse Load response < X Mid-stance < X Terminal stance < X Pre-swing < X Initial swing < X Mid-swing < X Terminal swing < X a Mean 1 SD. *X denotes statistical significance at p < DISCUSSION This study examined 22 RA patients (29 feet) and found a generalized decreased ROM in multiple foot and ankle segments throughout the majority of the gait cycle. Significant alterations in temporal-spatial parameters associated with decreased walking speed were also observed. Figure 6. Hallux segment kinematics and range of motion (mean 1 SD) in sagittal, coronal, and transverse planes for RA vs. Control. (* indicates statistically significant difference in ROM at p < 0.001). JOURNAL OF ORTHOPAEDIC RESEARCH MARCH 2007 DOI /jor

9 KINEMATIC CHANGES OF THE FOOT AND ANKLE 327 Table 6. ROM of Hallux Segment in Three Planes for RA vs. Control a Plane Phase Control (8) RA (8) P* Sagittal Load response Mid-stance Terminal stance < X Pre-swing < X Initial swing Mid-swing Terminal swing Coronal Load response Mid-stance Terminal stance Pre-swing Initial swing X Mid-swing Terminal swing Transverse Load response < X Mid-stance < X Terminal stance < X Pre-swing Initial swing < X Mid-swing < X Terminal swing < X a Mean 1 SD. *X denotes statistical significance at p < Table 7. Radiographic Measures from RA Population HV angle IM1 angle MTP1 dorsiflexion No. of lesser toes subluxed 3 1 These results confirm that during gait, significant alterations in foot segment kinematics occur in patients with forefoot changes associated with RA. All RA subjects were tested prior to planned surgical intervention for chronic RA and were symptomatic despite attempts at appropriate means of conservative treatment, including shoe modifications, orthotics/bracing, and medical management. The results of this study may not be as applicable to patients who are in the early stages of RA or are undergoing conservative care (orthotics, shoe modifications, etc.). While a large number of significant differences were observed, the clinical relevance of kinematic changes less than 1 28 is uncertain. Finally, of the 15 RA subjects tested unilaterally, 13 reported current or previous discomfort in the contralateral limb. Of these, seven had a previous successful forefoot reconstruction on the nontested side. The remaining six were being treated successfully with conservative measures on the contralateral side. Subjects were monitored for comfort during testing; their gait patterns were assumed to be natural and repeatable, as none reported undue or excessive pain while walking. Further investigation may be necessary to discern the effects of contralateral pathology on the tested limb; such an investigation may require stratification of a larger number of subjects at different stages of disease progression. Normal motion of the foot and ankle during gait has been well described in terms of 4 segments with respect to the phase of the gait cycle. 5 8,10,11,29 While the gait of these RA patients may be generally characterized as having less overall motion, we note several areas of interest. Significantly reduced dorsiflexion of the hallux was noted during terminal stance and pre-swing (Fig. 6). The Control group s rapid transition of the hallux to a position of peak dorsiflexion is characteristic of normal gait, 8 and the initial component in activating the windlass mechanism. In contrast, the RA group s peak dorsiflexion was 208 less than Control. The RA group s gait deficits seem to stem at least in part from this limitation, as reduced peak dorsiflexion at toe-off leads to reduced effectiveness of the windlass mechanism. 30 Similar reductions in hallux motion were observed by Woodburn et al., 20 who noted a strong association between these deficiencies and deformity and stiffness of the 1st MTP joint. DOI /jor JOURNAL OF ORTHOPAEDIC RESEARCH MARCH 2007

10 328 KHAZZAM ET AL. Further evidence is seen in assessing forefoot motion during the same period; the Control group exhibited increasing varus and adduction consistent with increasing supination, but the RA group exhibited only mildly increased varus and maintained an abducted position for most of the gait cycle (Fig. 5). The effects of this reduction in supination are also apparent in the temporalspatial parameters. All measures were significantly different, but clinical significance can be ascribed to the RA group s prolonged stance phase (6% longer than Control), reduced stride length (75% of Control), and reduced walking speed (79% of Control). A similar relationship between speed and hip and knee ROM was previously demonstrated. 31 The focus of this study was characterization of gait patterns in RA patients. The moderate to severe pain noted by most patients during clinical exam is a likely reason for the patients own voluntary alteration of motion, in an effort to avoid loading in areas that are painful. Previous work noted lateral shifting of the center of pressure (COP) in the forefoot during late stance, in an avoidance of toe-off at MT1-2, 16 and altered gait patterns in an attempt to avoid loading the forefoot. 20 While no direct comparison exists between our work and those, the changes in foot position we observed are intuitively consistent with the plantar pressure shifting reported by Minns and Craxford. 16 Pain stemming from hallux valgus and lesser toe deformities might also factor into the decrease in motion, as these deformities generally restrict MTP joint motion. Muscle weakness (specifically associated posterior tibial tendon dysfunction) may also contribute to these changes. Our findings agree with previous works. Though examination of 10 RA patients with 7 normals by O Connell et al. 15 was limited to sagittal plane motion of the ankle, both studies found decreased dorsiflexion/plantarflexion during stance with the largest difference between groups seen during late stance. The authors also noted decreased walking speed and stride length in RA compared to normals. Siegel and colleagues examination of four patients with RA 13 found similar kinematic and temporalspatial characteristics, with prolonged foot-flat during stance; they also noted prolonged ankle dorsiflexion and decreased heel rise and plantarflexion at toe-off. The recent work by Woodburn et al. 20 is most similar to our study. Both studies found reduced triplanar ROM in the forefoot throughout the gait cycle and excessive hindfoot external rotation. Unlike the Woodburn study, we measured hindfoot inversion, though the position was not significantly different from Control. The difference between the studies is likely due to our radiographic indexing through the use of the MFM, which accounts for alterations in underlying bony anatomy. Correction angles, such as those for valgus misalignment in the hindfoot, are applied with this technique. 9 Our results supplement the existing works by providing information from a larger population of RA patients, using motion analysis methods based on radiographic indexing to analyze a full gait cycle. The inclusion of radiographic and preoperative functional measures helped to characterize these patients; additional testing using such tools as the Foot Health Status Questionnaire, the Foot Function Index, or the Musculoskeletal Outcomes Data Evaluation and Management System (MOD- EMS) may be warranted in the future. This report is a subset of a larger study involving pre- and postoperative evaluation of the specified patient population. A future study of surgical follow-up and outcomes assessment is planned. The established study protocol involved two visits (about 1 month before surgery and 12 months after surgery), with a single test session for each visit. Consequently, within- and between-day variability could not be established for the collected data. Woodburn et al. 20 previously reported high withinday coefficients of multiple correlation (CMC) for a population with generalized chronic RA ( ); as our recruitment criteria represented a subset of their criteria, the variability should be similar between the populations. Our protocol also did not involve collection of ground reaction force (GRF) or plantar pressure data. Synchronized collection of these data with kinematics data would represent a novel opportunity for further characterization of this population s gait. In conclusion, the bony deformities that are characteristic of RA can cause severe disability, profoundly affecting ambulation. Relatively little information has been published that quantitatively characterizes dynamic kinematic changes that occur during gait as a result of RA of the forefoot. Examination of the dynamic differences between motion of patients with forefoot RA and normals helps to characterize the pathomechanics of this condition. Our results demonstrate significant decreases in temporal-spatial parameters (walking speed and stride length), accompanied by decreased motion of the hindfoot, forefoot, and hallux in all planes. This assessment reveals biomechanical alterations in a population ambulating with structural deformity and severe pain. These results support further characterization of JOURNAL OF ORTHOPAEDIC RESEARCH MARCH 2007 DOI /jor

11 KINEMATIC CHANGES OF THE FOOT AND ANKLE 329 pathologic gait kinematics. Findings from our study may prove useful in the preoperative evaluation; postoperative analysis using these same methods will allow a quantitative assessment of the efficacy of surgical intervention. Further evaluation via kinetic analysis will also be useful, pending the availability of technology capable of measuring the distribution of vertical and shear forces on the plantar surface of the foot. These technological needs are beyond the capability of current commercially available plantar pressure systems, but represent an opportunity for future work. REFERENCES 1. Ouzounian T Rheumatoid arthritis of the foot and ankle. In: Myerson MS, editor. Foot and ankle disorders. Philadelphia: Saunders; p Abdo R, Iorio L Rheumatoid arthritis of the foot and ankle. J Am Acad Orthop Surg 2: Coughlin MJ Athritides. In: Mann RA, Coughlin MJ, editors. Surgery of the foot, 7th ed. St. Louis: Mosby; p Coughlin MJ Rheumatoid forefoot reconstruction. A long-term follow-up study. J Bone Joint Surg [Am] 82: Beyaert C, Sirveaux F, Paysant J, et al The effect of tibio-talar arthrodesis on foot kinematics and ground reaction force progression during walking. Gait Posture 20: Castro MD Ankle biomechanics. Foot Ankle Clin 7: Harris GF Analysis of ankle and subtalar motion during human locomotion. In: Stiehl JB, editor. Inman s joints of the ankle, 2nd ed. Baltimore: Williams & Wilkins; p Johnson JE, Kidder SM, Abuzzahab FS Jr Threedimensional motion analysis of the adult foot and ankle. In: Harris GF, Smith PA, editors. Human motion analysis. New York: IEEE Press; p Johnson JE, Lamdan R, Granberry WF, et al Hindfoot coronal alignment: a modified radiographic method. Foot Ankle Int 20: Mann RA Biomechanics of the foot and ankle. In: Mann RA, editor. Surgery of the foot, 7th ed. St. Louis: Mosby; p Stauffer RN, Chao EY, Brewster RC Force and motion analysis of the normal, diseased, and prosthetic ankle joint. Clin Orthop Relat Res 127: Schwartz M Kinematics of normal gait. In: Gage JR, editor. The treatment of gait problems in cerebral palsy. London: Mac Keith Press; p Siegel KL, Kepple TM, O Connell PG, et al A technique to evaluate foot function during the stance phase of gait. Foot Ankle Int 16: Platto MJ, O Connell PG, Hicks JE, et al The relationship of pain and deformity of the rheumatoid foot to gait and an index of functional ambulation. J Rheumatol 18: O Connell PG, Lohmann Siegel K, Kepple TM, et al Forefoot deformity, pain, and mobility in rheumatoid and nonarthritic subjects. J Rheumatol 25: Minns RJ, Craxford AD Pressure under the forefoot in rheumatoid arthritis. A comparison of static and dynamic methods of assessment. Clin Orthop Relat Res 187: Otter SJ, Bowen CJ, Young AK Forefoot plantar pressures in rheumatoid arthritis. J Am Podiatr Med Assoc 94: MacWilliams BA, Cowley M, Nicholson DE Foot kinematics and kinetics during adolescent gait. Gait Posture 17: Ringleb SI, Hansen DK, Kotajarvi BR Changes in gait associated with posterior tibial tendon dysfunction. In: 10th Annual Meeting of the GCMAS, Portland, OR. Gait and Posture (in press). 20. Woodburn J, Nelson KM, Siegel KL, et al Multisegment foot motion during gait: proof of concept in rheumatoid arthritis. J Rheumatol 31: Wu WL, Su FC, Cheng YM, et al Gait analysis after ankle arthrodesis. Gait Posture 11: Carson MC, Harrington ME, Thompson N, et al Kinematic analysis of a multi-segment foot model for research and clinical applications: a repeatability analysis. J Biomech 34: Abuzzahab FS Jr, Harris GF, Kidder SM Foot and ankle motion analysis system: instrumentation, calibration, and validation. In: Harris GF, Smith PA, editors. Human motion analysis. New York: IEEE Press; p Kidder SM, Abuzzahab FS Jr, Harris GF, et al A system for the analysis of foot and ankle kinematics during gait. IEEE Trans Rehabil Eng 4: Myers KA, Wang M, Marks RM, et al Validation of a multisegment foot and ankle kinematic model for pediatric gait. IEEE Trans Neural Syst Rehabil Eng 12: Arnett FC, Edworthy SM, Bloch DA, et al The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum 31: Kitaoka HB, Alexander IJ, Adelaar RS, et al Clinical rating systems for the ankle-hindfoot, midfoot, hallux, and lesser toes. Foot Ankle Int 15: Perry J Gait analysis: normal and pathologic function. Thorofare, NJ: SLACK, Inc. 29. Johnson JE, Harris GF Pathomechanics of posterior tibial tendon insufficiency. Foot Ankle Clin 2: Fuller EA The windlass mechanism of the foot. A mechanical model to explain pathology. J Am Podiatr Med Assoc 90: Escalante A, Lichtenstein MJ, Hazuda HP Walking velocity in aged persons: its association with lower extremity joint range of motion. Arthritis Rheum 45: DOI /jor JOURNAL OF ORTHOPAEDIC RESEARCH MARCH 2007