Clinical Biomechanics 20 (2005) 526–531 www.elsevier.com/locate/clinbiomech
The relationship between hallux dorsiflexion and ankle joint complex frontal plane kinematics: A preliminary study Jill Halstead *, Deborah. E. Turner, Anthony. C. Redmond Academic Unit of Musculoskeletal Disease, School of Medicine, University of Leeds, 36 Clarendon Road, Leeds LS2 9NZ, UK Received 19 September 2004; accepted 12 January 2005
Abstract Background. It has been suggested that the function of the first metatarsophalangeal joint may be related to the motion of the ankle joint complex. Objective. This study explored the relationship between ankle joint complex and first metatarsophalangeal joint motion during gait in a group of 14 who demonstrated clinically limited passive hallux dorsiflexion in quiet standing (cases), and 15 matched controls. Method. An electromagnetic tracking system was used to measure the ankle joint complex frontal plane motion and first metatarsophalangeal joint sagittal plane motion during gait, in both cases and controls. The case group was then evaluated further to investigate the effect of an orthosis on first metatarsophalangeal joint motion. Findings. The correlation between maximum ankle joint complex eversion and maximum first metatarsophalangeal joint dorsiflexion during gait was r = 0.471. Within the case group, maximum rearfoot eversion was reduced following the application of the orthoses, but there was no change in sagittal first metatarsophalangeal joint rotations. Interpretation. The relationship between maximum ankle joint complex eversion and first metatarsophalangeal joint dorsiflexion kinematics found in this study was moderate, and decreasing maximum ankle joint complex eversion with an orthosis did not result in any increase in first metatarsophalangeal joint dorsiflexion during gait in patients with functional first metatarsophalangeal joint limitation. These results do not support the assumption that ankle joint complex eversion influences first metatarsophalangeal joint motion substantially. Ó 2005 Elsevier Ltd. All rights reserved. Keywords: Foot; First metatarsophalangeal joint; Ankle; Kinematics; Gait analysis
1. Introduction The first metatarsophalangeal joint (MTPJ1) is the second most common site of osteoarthritis in the lower limb after the knee (Muehleman et al., 1997). The development of osteoarthritis has been associated with mechanical factors that disrupt normal synthesis of articular cartilage chrondrocytes, extracellular matrix, and subchondral bone (Creamer and Hochberg, 1997). Excessive or high velocity motion at the ankle joint com*
Corresponding author. E-mail addresses: jillhalstead@gmail.com (J. Halstead), a.redmond@leeds.ac.uk (Anthony. C. Redmond). 0268-0033/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.clinbiomech.2005.01.004
plex (AJC) has been implicated as a mechanical factor affecting motion of the MTPJ1 and leading to abnormal forces within the joint (Short, 2000). Patients with MTPJ1 osteoarthritis have been shown to have higher medial forefoot pressures and more pronated foot postures in comparison to normal participants (Kernozek et al., 2003; Bryant et al., 1999; Greenburg, 1979). Ankle joint complex motions and resulting foot pressures can be altered systematically by foot orthoses (Van Gheluwe and Dananberg, 2004; Nester et al., 2001; Masse Genova and Gross, 2000; Novick and Kelley, 1990; Bates et al., 1979), and orthotic treatment has improved MTPJ1 osteoarthritis pain and quality of life scores in symptomatic patients (Torkki et al., 2001, 2003;
J. Halstead et al. / Clinical Biomechanics 20 (2005) 526–531
Grady et al., 2002). The mechanism by which the orthotics may have reduced pain is not well understood however. Functional impairment of the MTPJ1 is said to exist where there is a functionally reduced range of motion during walking in a joint that has no structural limitations. This presentation is sometimes referred to as functional hallux limitus (Dananberg, 1993), and has been implicated as a factor in the development of MTPJ1 osteoarthritis (Drago et al., 1984). A moderate relationship (r = 0.61) has been described between maximum passive MTPJ1 dorsiflexion during standing and maximum MTPJ1 dorsiflexion during gait in normals (Nawoczenski et al., 1999). The relationship between standing AJC position and passively induced MTPJ1 dorsiflexion has been investigated previously in three studies of static standing. One in vitro study (n = 5) showed that incremental increases in rearfoot eversion were associated with increased metatarsal dorsiflexion and eversion (Oldenbrook and Smith, 1979). A larger in vivo study (n = 30) showed that rearfoot eversion (using lateral rearfoot wedging) significantly reduced peak hallux dorsiflexion in simulated standing (Harradine and Bevan, 2000), while in contrast, medially posted orthoses produced no significant changes in passive hallux dorsiflexion in a sample of 30 patients with MTPJ1 osteoarthritis (Hogan and Kidd, 2001). Two studies have described the MTPJ1 motions during gait and the effect of an orthotic device. The first study (n = 5) found a small reduction in MTPJ1 motions following orthotic therapy (Kilmartin et al., 1991), while the second study (n = 18) found that two types of orthosis produced similarly small increases in MTPJ1 motion, although results were highly variable (Nawoczenski and Ludewig, 2004). Neither the relationship between standing and walking MTPJ1 motions, nor the relationship between MTPJ1 and AJC has been explored in patients with MTPJ1 dysfunction. The aims of this study were: (1) To explore the relationship between frontal plane AJC, and MTPJ1 sagittal dorsiflexion during gait in a group of patients with clinical functional hallux limitus, and in age and gender matched controls. (2) To investigate in the case group, the influence on MTPJ1 dorsiflexion, of a 10° rearfoot medial wedge intended to alter frontal plane motion at the AJC.
527
limb musculoskeletal assessment who demonstrated clinical functional hallux limitus. Functional hallux limitus was defined as the presence of passive hallux dorsiflexion in quiet stance <40° despite a non-weight-bearing range of motion P50°, the cutoffs were based upon the normative data of Nawoczenski et al. (1999). The control group were 15 age and gender matched volunteers who demonstrated normal passive hallux dorsiflexion during quiet standing (P40°). Participants were screened to ensure that none had a history of lower limb trauma in the last six months, ankle or foot surgery in the last 18 months, or musculoskeletal disease in the lower limb. Ethical approval for the study was granted by the Leeds Teaching Hospitals Trust Local Research Ethics Committee. 2.1. Instrumentation Joint motion data were captured at 30 Hz using four channels of a FastrakTM (Polhemus Inc., Colchester, VT, USA) electromagnetic motion tracking system configured with a long range transmitter. Stance phase events were defined using force sensing resistors (Interlink Electronics Inc, Santa Barbara, CA, USA), incorporated into a split platform within the electromagnetic motion tracking field. Sensor data were filtered and time synchronised using the 6D ResearchTM motion analysis software package (Skill Technologies Inc., Phoenix, AZ, USA) as described previously (Woodburn et al., 1999). Sensor placement protocols and joint coordinate systems have been described previously for the AJC and MTPJ1, and were employed in this study (Woodburn et al., 1999; Umberger et al., 1999). Excursion of extensor hallucis longus tendon has been described previously as a significant source of error when mounting a sensor over the dorsum of the hallux (Umberger et al., 1999). This source of error was minimised by securing the sensor with a Velcro strap that encircled the toe according to a recently described protocol modification (Longworth et al., 2005). Following data capture, segment and joint models were constructed in the software, angular rotations derived, and walking data normalised to 101 centiles of the stance phase. In this study, the angular rotations of interest were AJC inversion/eversion (designated AJCb) and MTPJ1 dorsiflexion/plantar flexion (designated MTPJ1a). 2.2. Procedure
2. Method The case group comprised 14 patients attending a hospital out-patients foot health department for lower
All participants underwent a clinical examination of passive hallux dorsiflexion to compare the range of available dorsiflexion weight bearing and non-weight bearing. Weight-bearing examinations were taken with the patients standing in relaxed double limb support,
528
J. Halstead et al. / Clinical Biomechanics 20 (2005) 526–531
knees fully extended, and with the foot in its relaxed position. Non-weight-bearing examinations were taken with the patient sitting, knee and ankle flexed to 90° and the heel aligned to the long axis of the shank. A single researcher (JH) passively dorsiflexed the hallux by pushing up on the proximal phalangeal till a maximum dorsiflexion was reached. In the 14 cases the limb with the most severe limitation of MTPJ1 dorsiflexion was used for the electromagnetic motion tracking studies. In the control group the dominant limb of the participant was determined as the limb used to initiate walking, and this limb was instrumented for the kinematic studies. Sensors were applied and cables secured to the study limb. The reference position was established with participants standing with the calcaneus aligned to a near vertical position with the talus head equally palpable at anterior of the ankle joint (Elveru et al., 1988). The standing passive hallux dorsiflexion tests were repeated three times per participant and the mean calculated. Participants then completed five walking trials across a 9 m walkway at a self-selected walking speed. The control group undertook barefoot walking trials only, while the case group were assigned barefoot and intervention walking trials, according to a predetermined randomisation protocol. The intervention was a single plane 10° medial wedge made from high density (450 kg/M3) ethyl vinyl acetate, applied directly to the plantar surface of both heels. The five repeat walking trials were normalised and averaged for each participant in each of the barefoot and intervention conditions.
3. Results 3.1. Consistency The CMCÕs indicate a high degree of within-subject consistency at both joint complexes during gait (see Table 1). 3.2. Relationship between AJC and MTPJ1 The correlation between maximum MTPJ1a (at 98% of stance) and AJCb (at 68% of stance) was moderate (r = 0.471, P = 0.01). The logistic regression investigating the relationship between maximum AJC eversion during gait, and the standing test of functional hallux limitus indicated poor discernment between cases and controls on the basis of rearfoot eversion during gait (Cox and Snell R2 = 0.035, B = 0.083, SE = 0.84, Wald = 0.97, P = 0.325). 3.3. Case and control groups There were no significant differences in the age or gender profile, between cases and controls (see Table 2). There were significant differences in maximum MTPJ1astanding between case and control groups in quiet standing, with the mean for the case group 20° (50%) lower than the mean of the controls, confirming the clinical test (see Table 2). In contrast however, mean MTPJ1adynamic during gait, differed by only 1.3° (3%) between cases and controls (see Table 2), a difference that was not statistically significant. The motion time curves are presented in Figs. 1 and 2.
2.3. Analyses 3.4. Intervention in the case group All analyses were carried out using Microsoft ExcelTM and SPSS v.11.5. Coefficients of Multiple Correlation (CMC) were derived according to the method of Kadaba (Kadaba et al., 1989). Differences between maximum MTPJ1a motion during quiet standing (MTPJ1astatic) in cases and controls were tested with a Students un-paired t-test. Maximum AJC eversion was defined within the joint coordinate system as minimum AJCbdynamic, as this parameter is thought to potentially predict the presence of hallux limitation in static standing. A logistic regression model was constructed to investigate the degree to which maximum AJC eversion (minimum AJCbdynamic) predicted the presence of a functional hallux limitus (MTPJ1astatic). The relationship between peak AJC eversion and MTPJ1a during gait was tested using PearsonÕs product moment correlation coefficient. In the case group, the significance of differences in maximum AJC eversion and MTPJ1a between baseline and the 10° rearfoot wedge intervention were investigated using a Students paired t-test.
The results of the case group at baseline (barefoot) and following the addition of the intervention of the 10° rearfoot medial wedge are presented in Table 3. The intervention reduced mean peak AJCbdynamic by 2.76° (25%). Despite having a significant influence on AJCb kinematics however, the intervention of the 10° wedging had a minimal effect on peak MTPJ1a during gait.
4. Discussion Within-subject consistency was high at both joint complexes, with CMCs > 0.93 for the five repeat Table 1 Coefficients of multiple correlation for within-subject variability at the two joint complexes in cases and controls
ACJb MTPJ1a
Within-subject CMC Within-subject CMC
Case
Control
0.964 0.966
0.936 0.956
J. Halstead et al. / Clinical Biomechanics 20 (2005) 526–531
529
Table 2 Participant characteristics and mean peak angular rotation, and instant of peak rotation in cases and controls Case
Control
Mean Number Gender (F/M) Age Age range Mean MTPJ1astanding Mean MTPJ1adynamic Time MTPJ1adynamic Mean AJCbdynamic Time mean AJCbdynamic
14 8/6 35.1
(SD)
Mean
(9.2)
15 10/5 35.5
18–52
Difference between groups (SD)
Mean
(95% CI)
(6.1)
– – 0.4
( 7 to 6.2)
(6.06) (7.98) (5.5) (5.68) (9.51)
20° 1.32° 1.57% 1.73° 2.25%
( ( ( ( (
25–49
19.32° 38.21° 99.57% 10.82° 68.78%
(9.15) (5.77) (2.24) (3.41) (15.67)
39.35° 36.89° 97% 9.09° 66.53%
26.04 to 14.01)** 3.97 to 6.61) 1.65 to 4.8) 5.31 to 1.84) 7.81 to 12.31)
Time = Mean instant of maximum joint rotation (AJC eversion, MTPJ1 dorsiflexion). ** P < 0.001.
5.0 3.0
Angular rotation
1.0
Inversion
-1.0 -3.0 -5.0 -7.0
Eversion -9.0 -11.0 -13.0 0
Key: ____
10
20
30
40
50 60 % Stance phase
70
80
90
100
Case Control
Fig. 1. Motion time curves for AJCb (mean ± 95% CI) for the cases and controls.
45.0
Angular rotation
35.0
25.0
Dorsiflexion
15.0
Plantarflexion 5.0
-5.0 0
10
20
30
40
50 60 % Stance phase
70
80
90
100
Key: ____
Case
____
Control
Fig. 2. Motion time curves for MTPJ1a (mean ± 95% CI) for the cases and controls.
walking trials. The correlation between peak AJC eversion and MTPJ1 dorsiflexion during the stance phase of gait indicated a moderate relationship between motions at these two joint complexes. It was not possible to
quantify fully the extent of the predictive relationship between these two variables, as the MTPJ1adynamic data violated the assumptions of homogeneity of variance and normal distribution, and was not suitable for full linear regression modelling. The motion time data indicates that peak AJCb angle occurs 30 centiles before the peak MTPJ1a angle, and so peak AJCb motion may be less influential on the MTPJ1a during late stance phase. Coupling of transverse plane motions of the tibia to the frontal plane motions of the navicular and metatarsal have been reported (Cornwall and McPoil, 2002), but the relationship between the AJC and the forefoot joints requires further investigation. Mean differences between the maximum MTPJ1a of the cases and controls measured in quiet standing, confirmed the capacity of the clinical test to differentiate the case and control groups according to the test protocol. The differences in peak standing MTPJ1a observed between cases and controls did not however translate into differences between these groups during the stance phase of gait, calling into question the construct validity of the test. Similar values were found for both MTPJ1astanding and MTPJ1adynamic in the control group, findings in agreement with the 38° and 42° reported in two studies of normals (Nawoczenski and Ludewig, 2004; Nawoczenski et al., 1999). The sensitivity and specificity of the passive test for limited MTPJ1 dorsiflexion has not been reported previously in the literature. Maximum AJC eversion during gait was a poor predictor of the outcome of the standing test of passive dorsiflexion of MTPJ1. The lack of consistency between standing and walking MTPJ1 motions in the case group and predictive association between peak AJCbdynamic and the MTPJ1adynamic during walking raises further questions over the appropriateness of using this clinical test of passive hallux dorsiflexion in quiet standing. The intervention with a 10° medial wedge reduced the mean peak AJCbdynamic by 2.76°, a similar change to the 2.2° to 3° reported previously (Masse Genova and Gross, 2000; Redmond et al., 2003). Despite a systematic
530
J. Halstead et al. / Clinical Biomechanics 20 (2005) 526–531
Table 3 Mean peak angular rotation for the case group at baseline and after intervention with a 10° medial heel wedge Baseline
Mean MTPJ1adynamic Mean AJCbdynamic **
10° Intervention
Difference between trials
Mean
(SD)
Mean
(SD)
Mean
(95% CI)
39.23° 10.84°
(6.06°) (3.45°)
39.33° 8.08°
(7.17°) (2.83°)
0.1° 2.76°
( 2.84 to 2.63) ( 3.74 to 1.77)**
P < 0.001.
change in AJCb kinematics there was however, no corresponding change in peak MTPJ1adynamic. In two previous studies mean change in MTPJ1 sagittal plane motion in response to orthoses has not exceeded 3° (Kilmartin et al., 1991; Nawoczenski and Ludewig, 2004). These results indicate altering AJC kinematics with an orthoses (10° medial heel wedge) does not result in a systematic change in MTPJ1 sagittal motion during gait.
5. Conclusion The clinical test of limited passive hallux dorsiflexion when repeated using an electromagnetic motion tracking system showed a clear distinction between the case and control groups. Functional limitation of the MTPJ1 demonstrated by the standing clinical test did not however, appear to predict altered MTPJ1 dorsiflexion during gait. The results of this study also indicate that the relationship between maximum AJCb and first MTPJa kinematics during walking is incomplete. The moderate correlation, and the absence of effect upon MTPJ1a motion when AJCb was altered, calls into question the strength of any clinical association and merits further investigation.
Acknowledgements The authors would like to acknowledge Dr. J. Woodburn and the Foot and Ankle studies in Rheumatology programme for assistance with data capture, and the Foot Health Department at Leeds Teaching Hospital Trust, UK, for their assistance with recruiting the patient cohorts. We are also grateful to Professor M. Cornwall, Northern Arizona University, USA, for the software routines used to normalise the gait data and calculate CMCs. ACR and DET are funded by the Arthritis Research Campaign.
References Bates, B.T., Osternig, L.R., Mason, B., James, L.S., 1979. Foot orthotic devices to modify selected aspects of lower extremity mechanics. American Journal of Sports Medicine 7, 338–342.
Bryant, A., Tinley, P., Singer, K., 1999. Plantar pressure distribution in normal, hallux valgus, hallux limitus feet. The Foot 10, 18–22. Cornwall, M.W., McPoil, T.G., 2002. Motion of the calcaneus, navicular, and first metatarsal during the stance phase of walking. Journal of the American Podiatric Medical Association 92, 67–76. Creamer, P., Hochberg, M.C., 1997. Osteoarthritis. Lancet 350 (9076), 503–508. Dananberg, H.J., 1993. Gait style as an etiology to chronic postural pain. Part 1: Functional hallux limitus. Journal of the American Podiatric Medial Association 83, 433–441. Drago, J.J., Oloff, L., Jacobs, A.M., 1984. A comprehensive review of hallux limitus. Journal of Foot Surgery 23, 213–220. Elveru, R.A., Rothstein, J.M., Lamb, R.L., Riddle, D.L., 1988. Methods for taking subtalar joint measurements. A clinical report. Physical Therapy 68, 678–682. Grady, J.F., Axe, T.M., Zager, E.J., Sheldon, L.A., 2002. A retrospective analysis of 772 patients with hallux limitus. Journal of the American Podiatric Medical Association 92, 102–108. Greenburg, G.S., 1979. Relationship of hallux abductus angle and first metatarsal angle to severity of pronation. Journal of the American Podiatry Association 69, 29–34. Harradine, P.D., Bevan, L.S., 2000. The effect of rearfoot eversion on maximal hallux dorsiflexion. Journal of the American Podiatric Medical Association 90, 390–393. Hogan, D., Kidd, R., 2001. Do functional foot orthoses change the range of motion of the first metatarsophalangeal joint of hallux limitus/rigidus? Australasian Journal of Podiatric Medicine 35, 39– 41. Kadaba, M., Ramakrishnan, H., Wootten, M.E., Gainey, J., Gorton, G., Cochran, G.V., 1989. Repeatablility of kinematic, kinetic, and electromyographic data in normal adult gait. Journal of Orthopaedic Research 7, 849–860. Kernozek, T.W., Elfessi, A., Sterriker, S., 2003. Clinical and biomechanical risk factors of patients diagnosed with hallux valgus. Journal of the American Podiatric Medical Association 93, 97–103. Kilmartin, T.E.A., Wallace, W., Hill, T.W., 1991. Orthotic effect on metatarsophalangeal joint extension: a preliminary study. Journal of the American Podiatric Medical Association. 81, 414–417. Longworth, R., Chockalingam, N., Woodburn, J., Redmond, A., 2005. An enhanced protocol to reduce error in electromagnetic tracking of first metatarsophalangeal joint motions. International Journal of podiatric Biomechanics 2, 43–44. Masse Genova, J., Gross, M.T., 2000. Effect of foot orthoses on calcaneal eversion during standing and treadmill walking for subjects with abnormal pronation. Journal of Orthopaedic Sports Physical Therapy 30, 664–675. Muehleman, C.D., Bareither, K., Huch Cole, A., Kuettner, K.E., 1997. Prevalence of degenerative morphological changes in the joints of the lower extremity. Osteoarthritis and Cartilage 5, 23–37. Nawoczenski, D.A.J., Ludewig, P.M., 2004. The effect of forefoot and arch posting orthotic designs on first metatarsophalangeal joint kinematics during gait. Journal of Orthopaedic and Sports Physical Therapy 34, 317–327. Nawoczenski, D.A.J., Baumhauer, J.F., Umberger, B.R., 1999. Relationship between clinical measurement and motion of the first metatarsophalangeal joint during gait. Journal of Bone and Joint Surgery—American 81, 370–376.
J. Halstead et al. / Clinical Biomechanics 20 (2005) 526–531 Nester, C.J., Hutchins, S., Bowker, P., 2001. Effect of foot orthoses on rearfoot complex kinematics during walking gait. Foot & Ankle International 22, 133–139. Novick, A., Kelley, D.L., 1990. Position and movement changes of the foot with orthotic intervention during the loading response of gait. Journal of Orthopaedic & Sports Physical Therapy 11, 301–312. Oldenbrook, L.L., Smith, C.M., 1979. Metatarsal head motion secondary to rearfoot pronation and supination. Journal of the American Podiatry Association 69, 24–28. Redmond, A.C., Woodburn, J., Helliwell, P.S., Emery, P., 2003. Single plane medial wedge may offer an alternative to expensive, cast functional foot orthoses. Annals of Rheumatic Diseases 62 (Suppl. 1), 394–395. Short, A., 2000. Treatment of osteoarthritis. Australasian Journal of Podiatric Medicine 34, 93–99. Torkki, M., Malmivaara, A., Seitsalo, S., Hoikka, V., Laippala, P., Paavolainen, P., 2001. Surgery vs orthosis vs watchful waiting for
531
hallux valgus: a randomized controlled trial. Journal of the American Medical Association 285, 2474–2480. Torkki, M., Malmivaara, A., Seitsalo, S., Hoikka, V., Laippala, P., Paavolainen, P., 2003. Hallux valgus: immediate operation versus 1 year of waiting with or without orthoses; a controlled randomised trial of 209 patients. Acta Orthopaedica Scandinavica 74, 209–215. Umberger, B.R.D., Nawoczenski, A., Baumhauer, J.F., 1999. Reliability and validity of first metatarsophalangeal joint orientation measures with an electromagnetic tracking device. Clinical Biomechanics 14, 74–76. Van Gheluwe, B., Dananberg, H.J., 2004. Changes in plantar foot pressure with in-shoe varus or valgus wedging. Journal of the American Podiatric Medical Association 94, 1–11. Woodburn, J., Turner, D.E., Helliwell, P.S., Barker, S., 1999. A preliminary study determining the feasibility of electromagnetic tracking for kinematics at the ankle joint complex. Rheumatology 38, 1260–1268.