Secondary Logo

Journal Logo

Altered Neuromuscular Control in Individuals with Exercise-Related Leg Pain

FRANETTOVICH, MELINDA1,2; CHAPMAN, ANDREW R.1,2,3; BLANCH, PETER2; VICENZINO, BILL1

Medicine & Science in Sports & Exercise: March 2010 - Volume 42 - Issue 3 - p 546-555
doi: 10.1249/MSS.0b013e3181b64c62
APPLIED SCIENCES: Symposium
Free

Purpose: To compare neuromuscular control of the lower limb during gait between individuals with and without a history of exercise-related leg pain (ERLP).

Methods: Fourteen females with a history of ERLP and 14 age-, height-, and weight-matched asymptomatic female controls participated in the study. Electromyographic activity, normalized to maximum voluntary contraction (MVC), from 12 lower limb muscles during walking gait was the primary outcome. Secondary outcomes were three-dimensional kinematics of the lower limb during gait, measurements of static foot posture (arch height and midfoot width in weight bearing and non-weight bearing), and foot mobility (difference in arch height and midfoot width from non-weight bearing to weight bearing and foot mobility magnitude).

Results: Individuals with a history of ERLP demonstrated lower peak activation (13.7% MVC, 95% confidence interval (CI) = 3.2%-24.3% MVC) and lower average activation of gluteus medius (2.3% MVC, 95% CI = 0.3%-4.3% MVC) when compared with controls (P < 0.05). This reduction in gluteus medius activation was moderately determined (57.1%, P = 0.01) by the duration (β = 0.555) and severity of pain (β = −0.516). Peak and average activation of lateral gastrocnemius were also lower than controls (20.5% MVC, 95% CI = 0.6%-40.5% MVC and 1.7% MVC, 95% CI = 0.2%-3.1% MVC, respectively) but were not explained by pain duration or severity. No differences in kinematics at the ankle, knee, hip and pelvis, or differences in static foot posture and mobility were observed between groups (P > 0.05).

Conclusions: This study provides evidence of altered neuromuscular control of gait in females with a history of ERLP. Further work is required to discern the clinical relevance of this finding.

1The University of Queensland, Brisbane, AUSTRALIA; 2Australian Institute of Sport, Canberra, AUSTRALIA; and 3McGill University, Montreal, CANADA

Address for correspondence: Bill Vicenzino, Ph.D., Division of Physiotherapy, School of Health and Rehabilitation Sciences, The University of Queensland, Brisbane QLD 4072, Australia; E-mail: b.vicenzino@uq.edu.au.

Submitted for publication March 2009.

Accepted for publication July 2009.

Exercise-related leg pain (ERLP) describes pain that is located between the knee and the ankle that is experienced during weight-bearing activities and is diminished/ceases when activities cease. ERLP encompasses the clinical and pathologic features of several commonly used labels such as shin splints, medial tibial stress syndrome, periostitis, and stress fractures (37). This condition affects between 11% and 38% of college athletes and military recruits (2,26,28-30,36,37,39) and is second only to knee pain as the most common lower extremity running injury (34). The condition often leads to interruption of training or participation in physical activity and limitation of performance; for example, ERLP caused interruption to training for 50%-60% of running athletes (28-30) and 67% of military recruits (39), and up to 60% of college running athletes reported that the condition negatively affected race performance (28,30). The following injury-related factors have been investigated in the literature: previous history of injury (26,28,29,39), gender (2,5,26,30,37,39), navicular drop (1,9,26,28-30,32), static foot posture (1,2,5,29,35,39), talocrural and subtalar joint range of motion (2,5,23,30,35,37,39), muscle endurance (20), and lower limb biomechanics and foot plantar pressures during running (36,37).

Several prospective studies have reported a previous history of injury as a factor in the development of ERLP in populations such as college athletes, runners, and naval recruits (26,28,29,39). Female gender has also been associated with the development of ERLP by six prospective studies across these populations (2,5,26,30,37,39). Conversely, numerous studies have reported that range of motion at the talocrural and subtalar joints is not associated with ERLP (2,5,23,30,37,39). Recent prospective studies by Willems et al. (36,37) included three-dimensional motion analyses of the ankle and knee during barefoot and shod running and showed that individuals who developed ERLP demonstrated increased ankle eversion excursion, velocity of ankle eversion, and ankle abduction excursion compared with those who remained uninjured. In the same studies, Willems et al. (36,37) also reported differences in plantar pressures between groups, with individuals who developed ERLP exhibiting a more laterally directed center of pressure at forefoot push-off and last foot contact. There is less conclusive and often contradictory evidence for other factors. Whereas some authors have reported an increase in navicular drop in individuals with ERLP (1,2,9,28), other authors have not (26,29,30). Similarly, some studies support a relationship between static foot posture (rear foot position, longitudinal arch angle, foot posture index) and ERLP (32,35,39), whereas others have been unable to demonstrate the same relationship (1,2,5,29). Madeley et al. (20) used the standing heel rise test to evaluate isotonic endurance of plantarflexor muscles of the ankle and reported that individuals with medial tibial stress syndrome performed less heel rise repetitions than the control group (23 ± 5.6 compared with 33 ± 8.6).

Consistent findings are evident for injury history, gender, talocrural and subtalar joint motion, and ankle and knee biomechanics and foot plantar pressures during running. Less conclusive evidence has been presented for other factors; for example, investigations on static foot posture and ERLP have produced conflicting results, and investigations of muscle endurance and ERLP are limited to only one study.

Whereas Madeley et al. (20) compared muscle endurance in individuals with and without ERLP, other aspects of neuromuscular control including muscle recruitment patterns during functional tasks have not been investigated. Altered muscle recruitment has been demonstrated in other musculoskeletal conditions (19); for example, altered quadriceps recruitment in patellofemoral pain (8) and altered trunk muscle recruitment in low back pain (15). Given altered muscle recruitment during weight bearing exercise can result in altered bone and soft tissue loading (7), altered muscle recruitment could feasibly be related to ERLP, and it is timely that muscle recruitment patterns be investigated in individuals with ERLP.

The primary aim of this investigation was to compare activity of lower limb muscles during gait between individuals with and without ERLP. Secondary aims were to further compare lower limb motion between these two groups and to confirm a lack of differences in foot posture and foot mobility. We hypothesized that individuals with ERLP would demonstrate different lower limb muscle recruitment and motion patterns when compared with uninjured individuals.

Back to Top | Article Outline

METHODS

Participants

Fourteen females with a history of ERLP and 14 age-, height-, and weight-matched asymptomatic female controls were recruited to participate in the study. All individuals provided informed written consent, and the study was approved by the University of Queensland's human research ethics committees. Individuals were excluded from either group if they had a history of surgery to the lower limb, had blood clotting or bleeding abnormalities, or had a neurologic or a cardiac condition.

ERLP was defined as pain located between the knee and ankle that is brought on or aggravated by loading during intensive weight bearing activities and diminishes or ceases when activity ceases (30,37). Individuals with a history of ERLP within the 12 months before the study were included in the ERLP group. Included individuals did not have point bone tenderness on palpation of the posterior-medial border of the tibia. For the purpose of this study, individuals were excluded if there was a medical diagnosis of compartment syndrome or tibial stress fracture within the last 12 months. Participants were also excluded if there were signs and symptoms of radiculopathy or other neurologic involvement. Information was also obtained regarding the duration of symptoms (months), the time since last symptoms (weeks), and pain severity (worst pain in the week before testing rated on a 10-cm visual analog scale). Individuals in the control group had no current lower limb injury or a history of lower limb injury in the past 12 months that required treatment or interfered with work or leisure.

Back to Top | Article Outline

Procedure

Individuals walked barefoot on a treadmill at a self-selected speed for 10 min. Barefoot walking was selected to avoid potential measurement error introduced by a shod condition (i.e., movement of foot within shoe where markers are mounted on the shoe) (31). Running was not assessed because running was a provocative activity for some individuals with ERLP, and we did not want to confound results with the effect of pain on muscle activity and motion. Measurements of static foot posture and foot mobility were performed before walking, whereas electromyographic (EMG) and kinematic data were recorded throughout the 10 min of walking.

Back to Top | Article Outline

Electromyography

We measured EMG activity (Noraxon, Scottsdale, AZ) from 12 lower limb muscles: tibialis posterior (TP), tibialis anterior (TA), peroneus longus (PL), soleus (SOL), gastrocnemius medialis and lateralis (MG and LG), rectus femoris (RF), vastus medialis obliquus (VMO), vastus lateralis (VL), semitendinosus (ST), biceps femoris (BF), and gluteus medius (GM). Because of its deep location in the leg, an intramuscular recording was performed for TP to reduce contamination from attenuation of the signal or cross talk from overlying muscles (11,25). Bipolar intramuscular electrodes were fabricated from a Teflon-coated stainless steel wire (California Fine Wire Company, Grover Beach, CA) from which 2 mm of insulation was removed before inserting into a hypodermic needle (0.41 × 32 mm): the exposed tips were bent back by 2 and 4 mm to prevent contact. TP insertions were made from the posteromedial aspect of the leg using real-time ultrasound guidance (Nemio 20; Toshiba, Japan) to ensure accurate positioning of the intramuscular electrode (16). For all other muscles, recordings were performed using bipolar silver/silver chloride surface electrodes of 10-mm-diameter contact area and a fixed interelectrode distance of 20 mm (Nicolet Biomedical). Surface electrodes were applied in accordance with SENIAM recommendations (14), and electrode positions were chosen with reference to recommendations from Perotto (24) and innervations zone locations reported by Rainoldi et al. (27) (Table 1). A ground electrode was placed over the proximal shaft of the tibia (3M Health Care). EMG data were sampled at 3000 Hz and band-pass filtered between 10 and 1000 Hz.

TABLE 1

TABLE 1

Back to Top | Article Outline

Kinematics

An eight-camera VICON motion analysis system (Oxford Metrics, Oxford, United Kingdom) was used to measure three-dimensional motion of the lower limb at a sampling rate of 250 Hz. Placement of retroflective markers and calculation of kinematic data were based on the Plug-in Gait model (Oxford Metrics) that has been previously described (13). Joint rotations were referenced to standing position. From the model, we derived the following motions (negative-positive direction): pelvic posterior-anterior tilt in the sagittal plane, mediolateral tilt in the frontal plane, and external-internal rotation in the transverse plane; hip extension-flexion in the sagittal plane, abduction-adduction in the frontal plane, and external-internal rotation in the transverse plane; knee extension-flexion in the sagittal plane, valgus-varus in the frontal plane, and external-internal rotation in the transverse plane; and ankle plantar-dorsiflexion in the sagittal plane and abduction-adduction in the transverse plane. Because only two markers defined the foot segment, frontal plane rotation was not calculated for the ankle (18).

Back to Top | Article Outline

Foot Posture and Mobility

A specially constructed platform was used to perform measurements of foot posture as described in detail by McPoil et al. (21). Weight-bearing and non-weight-bearing measurements of arch height and midfoot width were taken and then used to calculate three indices of foot mobility (21). The difference between weight-bearing and non-weight-bearing measurements of arch height and midfoot width (termed difference in arch height and difference in midfoot width) provides an indication of vertical and medial-lateral midfoot mobility, respectively. These indices are used to calculate the foot mobility magnitude, a composite measure of the vertical and medial-lateral midfoot mobility, where, based on Pythagorean theorem, foot mobility magnitude is equal to the square root of the sum of difference in arch height squared and difference in midfoot width squared (21).

Back to Top | Article Outline

Data Management

Signal processing.

EMG recordings were adjusted for direct current offset, full-wave rectified, and filtered to remove low-frequency movement artifact using a fourth-order Butterworth filter with a high-pass cutoff of 10 Hz for surface recordings and 50 Hz for intramuscular recordings (6). For kinematic data, a generalized cross-validatory spline was used to remove low-frequency movement artifact from marker trajectories (38).

Back to Top | Article Outline

Event detection.

Ten consecutive strides (defined as foot contact to subsequent foot contact) were selected from each minute of data. Gait events (foot contact and foot off) were detected from the heel marker trajectory using a method that has been previously validated against force plate detection (40).

Back to Top | Article Outline

Signal normalization.

For each stride, EMG and kinematic data were time normalized to 100 points. EMG data were amplitude normalized to the peak activation during maximum voluntary contraction (MVC).

Back to Top | Article Outline

Data Analysis

Comparison of participants' characteristics.

Age, weight, and height were compared between groups to confirm that an adequate control was achieved. Speed of walking was also compared between groups to ensure it was not a confounding factor because previous studies have demonstrated that speed of gait influences amplitude of muscle recruitment (12).

Back to Top | Article Outline

Repeatability of outcome measures.

Comparisons between control and ERLP groups were made only after repeatability of outcome measures had been confirmed. Repeatability of EMG and kinematic data were calculated using the root mean square error (RMSE) and coefficient of multiple correlations (CMC) between the average stride for each minute of gait. Repeatability of foot posture and foot mobility measures was calculated using intraclass correlation coefficients (ICC) and standard error of measurement (SEM).

Back to Top | Article Outline

Comparison of muscle recruitment between groups.

Amplitude (peak, stance, and swing phase average) and temporal (duration and time of onset and offset) indices of muscle activity were derived from EMG recordings (Fig. 1) (11).

FIGURE 1-E

FIGURE 1-E

Back to Top | Article Outline

Comparison of movement patterns between groups.

The minimum, maximum, and total excursion in each plane at the ankle, knee, hip, and pelvis were derived from kinematic data.

Back to Top | Article Outline

Comparison of foot posture and foot mobility between groups.

Measurements of foot posture and foot mobility were compared between groups.

A series of one-way multivariate ANOVA procedures with between-subject factor of group were performed to compare dependent variables between the control and ERLP groups for measurements of muscle recruitment, movement, and foot posture/mobility. The α level was set at 0.05. Statistically significant main effects were followed up with (i) tests of simple effects to obtain point estimates of effect (95% confidence intervals (CI)) for differences between groups and (ii) regression analyses that included pain severity, pain duration, and time since last symptoms as potential predictor variables. Both forward and backward stepwise regressions were performed with the P value of entering a potential predictor variable as 0.05 and for removing the variable as 0.10. Standardized mean differences (SMD = mean difference/pooled SD) were calculated to provide an estimate of the effect that can be used for comparing measures with different units and as a proxy for estimation of clinical meaningfulness given there are no widely accepted indices of clinical utility for the measures used in this study. SMD were classified as small (0.2-0.6), moderate (0.6-1.2), or large (>1.2) (17). We considered that a clinically meaningful effect would be one in which the mean difference was larger than the SD (i.e., SMD > 1.0). Power calculations indicated that 14 per group would be adequate to detect such effects at a power of 80% and P value of 0.05 (10).

Back to Top | Article Outline

RESULTS

Participant characteristics

There was no difference in age, weight, height, or speed of walking between the two groups (P > 0.05; Table 2). For the ERLP group, the mean (range) duration of symptoms was 32.5 months (2-132 months), time since last symptoms was 3.6 wk (0-12 wk), and pain severity was 1.43 cm (0-4.9 cm).

TABLE 2

TABLE 2

Back to Top | Article Outline

Repeatability of outcome measures

For details on the repeatability of EMG and kinematic data, see Tables 3 and 4, respectively. The CMC between the average stride for each minute of data were greater than 0.9 for all muscles, indicating high consistency of EMG recordings through the 10 min of walking. RMSE ranged from 2.8% to 16.2% MVC for the leg muscles and from 0.9% to 3.3% MVC for the hip and thigh muscles. With the exception of anterior-posterior pelvic tilt (CMC = 0.7), kinematic data in all planes for each joint were also highly consistent with CMC values greater than 0.9. RMSE ranged from 5.9° to 6.4° at the ankle, 3.3° to 16.0° at the knee, 3.6° to 13.7° at the hip, and 0.7° to 3.1° for the pelvis. All measurements of foot posture and mobility displayed excellent repeatability, with ICC above 0.9 and SEM less than 0.08 cm (Table 5).

TABLE 3

TABLE 3

TABLE 4

TABLE 4

TABLE 5

TABLE 5

Back to Top | Article Outline

Comparison of EMG between groups

The group mean and SD for EMG variables are displayed in Table 6. The multivariate ANOVA identified a difference for GM and LG (multivariate statistic, P = 0.018 and P = 0.048, respectively; univariate statistics, P < 0.05) but did not identify a difference in muscle activity for other muscles (P > 0.05). As illustrated in Figure 2, individuals with ERLP demonstrated lower activation of GM and LG. Follow-up tests revealed that, for GM, peak activity was lower by 13.7% MVC (95% CI = 3.2%-24.2% MVC, P = 0.01) during stance and by 4.8% MVC (95% CI = 1.6%-8.1% MVC, P = 0.01) during swing phase. Average stance phase activity was also lower by 2.3% MVC (95% CI = 0.3%-4.3% MVC, P = 0.03). These reductions in GM activity were moderate to large with SMD of 1.1, 1.4, and 0.9. Follow-up tests also revealed that, for LG, peak activity was lower by 20.5% MVC (95% CI = 0.6%-40.5% MVC, P = 0.04) and average swing phase activity was lower by 1.7% MVC (0.2%-3.1% MVC, P = 0.03). The reductions in LG activity were moderate with SMD of 0.9 and 1.0. Neither was there a difference in amplitude of activation for any other muscle nor was there any difference in temporal aspects of activation for any muscle (P > 0.05).

TABLE 6

TABLE 6

FIGURE 2

FIGURE 2

Back to Top | Article Outline

Relationship of pain characteristics to EMG differences between groups

Duration of symptoms and pain severity explained 57.1% of the reduction in GM activity (standardized β = 0.555 and −0.516, respectively, P = 0.01). That is, there was a greater deficit when there was a shorter duration of symptoms and a greater severity of pain. The summary table from the regression analyses is shown in Table 7. For other dependent variables showing significant main effects in EMG data (i.e., peak swing and stance average of GM, peak and swing average of LG), pain characteristics were not related (P > 0.05).

TABLE 7

TABLE 7

Back to Top | Article Outline

Comparison of kinematics between groups

Minimum, maximum, and total excursion in the sagittal, frontal, and transverse planes at the ankle, knee, hip, and pelvis were not different between groups for the combined or separate dependent variables (P > 0.05; Table 8).

TABLE 8

TABLE 8

Back to Top | Article Outline

Comparison of foot posture and foot mobility between groups

There was no difference in foot posture or mobility between the two groups for the combined or separate dependent variables (P > 0.05; Table 9).

TABLE 9

TABLE 9

Back to Top | Article Outline

DISCUSSION

This is the first study to investigate lower limb muscle activity during gait in individuals with ERLP. As a secondary aim, we also compared lower limb motion during gait and static foot posture and foot mobility between individuals with and without ERLP. The results support our hypothesis that individuals with ERLP demonstrate different lower limb muscle recruitment but not foot posture and foot mobility. In contrast, the data did not support our hypothesis that individuals with ERLP demonstrate different lower limb motion.

The reductions in GM and LG activation that were observed could be a predisposing factor in ERLP, a consequence of the pain and injury, or a combination of both. Although the underlying pathology of ERLP remains controversial (inflammation of periosteum, muscle and tendon, bone stress reaction or stress fracture, increased compartmental pressure) (3), pathomechanically, there are two main hypotheses, namely, soft tissue traction and tibial bending (3). The latter proposes that tibial bending is induced by long-term repetitive weight-bearing activity, which leads to a bone stress injury at the site of maximum bending (3). It is plausible that reduced activation of LG, which is in the lateral aspect of the tibia and diagonal from the typical site of injury (medial tibia), results in a varus force to the tibia and, consequently, increased load to the medial side. This is purely speculative because we did not measure load on bone tissue in our study. Alternatively, the reduction in LG activation may be a consequence of ERLP and related to the influence of pain on motor control (19). This alternative possibility was not supported by our analysis of the relationship of the LG deficit to pain duration and severity. However, we found that duration of symptoms and pain severity explained 57.1% of the reduction in peak GM activation in individuals with ERLP, with shorter duration and greater severity of pain being associated with a greater deficit. To our knowledge, the link between a reduction in GM activation and ERLP has not been suggested before. Altered muscle recruitment has been shown to persist up to 66 months after injury (4,33) and suggests that without targeted interventions, altered muscle recruitment may continue well after the individual has returned to activity, which may be related to injury recurrence. This is consistent with findings of previous prospective studies that highlight a strong relationship between ERLP history and recurrence (26,28,29,39). Altered neuromuscular control of hip muscles may be implicated in this relationship. Our data of LG and GM changes in ERLP provide a basis on which to further study the role of the neuromuscular system on the genesis of ERLP, which should lead to more effective interventions.

Alterations in muscle activation observed in the current study were not accompanied by alterations in lower limb motion. Possible explanations for this are that there is redundancy in the neuromuscular system and that changes in motion may have been small and within the sensitivity of the measurements performed. This is a novel finding for the pelvis and hip but, for the ankle and knee, is in agreement with Willems et al. (36,37), who also reported no difference in ankle motion in the sagittal plane or knee joint motion in all three planes between individuals who developed ERLP compared with those who did not. In contrast to the current study, Willems et al. (37) reported an increase in maximum foot abduction excursion (12.92° compared with 11.43°) in individuals who developed ERLP, whereas we did not. Although the finding of Willems et al. was statistically significant, the difference was small in magnitude (SMD = 0.3), and the clinical significance of a change of this magnitude is unclear.

We did not observe a difference in foot posture and mobility between the two groups. This finding is consistent with six previous studies, which reported that foot posture and mobility are not associated with ERLP (1,2,5,26,29,30), but is in contrast to seven other studies that did report an association (1,2,9,28,32,35,39). Measurement error is one plausible explanation for this discrepancy. To date, there is no widely accepted method for quantifying foot posture and mobility, and many of the methods previously used have demonstrated poor interrater reliability. We chose measurements that have previously demonstrated excellent intrarater and interrater reliability (0.97-0.99 and 0.83-0.99, respectively) (21) and have been validated using radiographs as the reference standard (r = 0.93) (22).

A limitation of the current study is the retrospective design that allowed us to identify a relationship between altered lower limb muscle activation and ERLP but not to discern the direction of causality. A second limitation is the large number of comparisons (dependent variables) in proportion to the sample size (14 each group), which potentially inflates the type I error rate. We attempted to address this by considering critical values from statistical tests in conjunction with standardized mean differences to provide an indication of the magnitude of observed differences. The differences we report for GM and LG were evident not only on one variable but also on multiple variables of a similar construct (i.e., peak and average amplitude), which strengthens our interpretation that these effects are real and not a function of the analysis. Our investigation was performed on female participants, and as some studies have identified differences in anthropometric measurements in males with medial tibial pain but not females (5), caution is required when generalizing our results to male populations. We chose the experimental activity of barefoot walking to reduce measurement error and to avoid confounding the results with the effect of pain on muscle activity and motion for individuals whom running was a provocative task, but we acknowledge that this may limit the external validity of the study. Finally, we would like to highlight that although we identified that reduced GM and LG activation was associated with ERLP, there is also strong evidence that history of injury and gender are associated with ERLP, and therefore, the factors identified in our study should be considered as a part of a multifactorial cause and sequelae of ERLP.

Back to Top | Article Outline

CONCLUSIONS

This is the first study to demonstrate altered muscle activation of gait in individuals with a history of ERLP. Specifically, individuals with a history of ERLP demonstrated lower activation of GM and LG compared with uninjured controls. We observed no differences in foot posture, foot mobility, or motion at the pelvis, hip, knee, and ankle between individuals with and without a history of ERLP. Whereas further work is required to discern the causal relationship between ERLP and altered muscle recruitment, this study highlights that interventions that address muscle recruitment are indicated in the management of ERLP. Further work is required to determine clinical interventions most effective in addressing control of GM and LG.

Melinda Franettovich is supported by the National Health and Medical Research Council of Australia. Dr. Andrew R. Chapman is supported by the Australian Research Council. There is no conflict of interest relevant to this study for any author.

The results of the current study do not constitute endorsement by American College of Sports Medicine.

Back to Top | Article Outline

REFERENCES

1. Bandholm T, Boysen L, Haugaard S, Kreutzfeldt M, Bencke J. Foot medial longitudinal-arch deformation during quiet standing and gait in subjects with medial tibial stress syndrome. J Foot Ankle Surg. 2008;47(2):89-95.
2. Bennett JE, Reinking MF, Pluemer B, Pentel A, Seaton M, Killian C. Factors contributing to the development of medial tibial stress syndrome in high school runners. J Orthop Sports Phys Ther. 2001;31(9):504-10.
3. Bouche RT, Johnson CH. Medial tibial stress syndrome (tibial fasciitis): a proposed pathomechanical model involving fascial traction. J Am Podiatr Med Assoc. 2007;97(1):31-6.
4. Branch TP, Hunter R, Donath M. Dynamic EMG analysis of anterior cruciate deficient legs with and without bracing during cutting. Am J Sports Med. 1989;17(1):35-41.
5. Burne SG, Khan KM, Boudville PB, et al. Risk factors associated with exertional medial tibial pain: a 12 month prospective clinical study. Br J Sports Med. 2004;38:441-5.
6. Chapman AR, Vicenzino B, Blanch P, Hodges PW. Patterns of leg muscle recruitment vary between novice and highly trained cyclists. J Electromyogr Kinesiol. 2007;18(3):359-71.
7. Christina KA, White SC, Gilchrist LA. Effect of localised muscle fatigue on vertical ground reaction forces and ankle joint motion during running. Hum Mov Sci. 2001;20(3):257-76.
8. Cowan SM, Bennell KL, Hodges PW, Crossley KM, McConnell J. Delayed onset of electromyographic activity of vastus medialis obliquus relative to vastus lateralis in subjects with patellofemoral pain syndrome. Arch Phys Med Rehabil. 2001;82:183-9.
9. DeLacerda F. A study of anatomical factors involved in shin splints. J Orthop Sports Phys Ther. 1980;2(2):55-9.
10. Faul F, Erdfelder E, Lang A-G, Buchner A. G*Power 3: a flexible statistical power analysis program for the social, and biomedical sciences. Behav Res Methods. 2007;29(2):175-91.
11. Franettovich M, Chapman A, Vicenzino B. A pilot study of the initial effects of anti-pronation taping on electromyographic activity of lower leg muscles during walking. J Sci Med Sport. 2006;9(suppl 1):33-4.
12. Gazendam MGJ, Hof AL. Averaged EMG profiles in jogging and running at difference speeds. Gait Posture. 2006;25(4):604-14.
13. Growney E, Meglan D, Johnson M, Cahalan T, An K. Repeated measures of adult normal walking using a video tracking system. Gait Posture. 1997;6:147-62.
14. Hermens HJ, Freriks B, Disselhorst-Klug C, Rau G. Development of recommendations for SEMG sensors and sensor placement procedures. J Electromyogr Kinesiol. 2000;10:361-74.
15. Hodges PW, Bui B. A comparison of computer based methods for the determination of onset of muscle contraction using electromyography. Electroencephalogr Clin Neurophysiol. 1996;101:511-9.
16. Hodges PW, Kippers V, Richardson CA. Validation of a technique for accurate fine-wire electrode placement into posterior gluteus medius using real-time ultrasound guidance. Electromyogr Clin Neurophysiol. 1997;37(1):39-47.
17. Hopkins WG, Marshall SW, Batterham AM, Hanin J. Progressive statistics for studies in sports medicine and exercise science. Med Sci Sports Exerc. 2009;41(1):3-12.
18. Kadaba MP, Ramakrishnan HK, Wootten ME. Measurement of lower extremity kinematics during level walking. J Orthop Res. 1990;8:383-92.
19. Kolt GS, Snyder-Mackler L, editors. Physical Therapies in Sport and Exercise. 2nd ed. Philadelphia (PA): Churchill Livingstone, Elsevier Limited; 2007. p. 121-6.
20. Madeley LT, Munteanu SE, Bonanno DR. Endurance of the ankle joint plantar flexor muscles in athletes with medial tibial stress syndrome: a case-control study. J Sci Med Sport. 2007;10:356-62.
21. McPoil T, Vicenzino B, Cornwall M, Collins N, Warren M. Reliability and normative values for the foot mobility magnitude: a composite measure of vertical and medio-lateral mobility of the midfoot. J Foot Ankle Res. 2009;2:6.
22. McPoil TG, Cornwall MW, Medoff L, Vicenzino B, Forsberg K, Hilz D. Arch height change during sit-to-stand: an alternative for the navicular drop test. J Foot Ankle Res. 2009;2:17.
23. Messier SP, Pittala KA. Etiologic factors associated with selected running injuries. Med Sci Sports Exerc. 1988;20(5):501-5.
24. Perotto AO. Anatomical Guide for the Electromyographer: The Limbs and Trunk. 3rd ed. Springfield (IL): Charles C. Thomas; 1994. p. 142-69.
25. Perry J, Easterday CS, Antonelli DJ. Surface versus intramuscular electrodes for electromyography of superficial and deep muscles. Phys Ther. 1981;61(1):7-15.
26. Plisky MS, Rauh MJ, Heiderscheit B, Underwood FB, Tank RT. Medial tibial stress syndrome in high school cross-country runners: incidence and risk factors. J Orthop Sports Phys Ther. 2007;37(2):40-7.
27. Rainoldi A, Melchiorri G, Caruso I. A method for positioning electrodes during surface EMG recordings in lower limb muscles. J Neurosci Methods. 2004;134:37-43.
28. Reinking MF. Exercise-related leg pain in female collegiate athletes: the influence of intrinsic and extrinsic factors. Am J Sports Med. 2006;34(9):1500-7.
29. Reinking MF, Austin TM, Hayes AM. Exercise-related leg pain in collegiate cross-country athletes: extrinsic and intrinsic risk factors. J Orthop Sports Phys Ther. 2007;37(11):670-8.
30. Reinking MF, Hayes AM. Intrinsic factors associated with exercise-related leg pain in collegiate cross-country runners. Clin J Sport Med. 2006;16(1):10-4.
31. Reinschmidt C, van den Bogert AJ, Murphy N, Lundberg A, Nigg BM. Tibiocalcaneal motion during running, measured with external and bone markers. Clin Biomech. 1997;12(1):8-16.
32. Sommer HM, Vallentyne SW. Effect of foot posture on the incidence of medial tibial stress syndrome. Med Sci Sports Exerc. 1995;27(6):800-4.
33. Swanik CB, Lephart SM, Giraldo JL, DeMont RG, Fu FH. Reactive muscle firing of anterior cruciate ligament-injured females during functional activities. J Athletic Train. 1999;34(2):121-9.
34. van Gent RN, Siem D, van Middelkoop M, van Os AG, Bierma-Zeinstra SMA, Koes BW. Incidence and determinants of lower extremity running injuries in long distance runners: a systematic review. Br J Sports Med. 2007;41:469-80.
35. Viitasalo JT, Kvist M. Some biomechanical aspects of the foot and ankle in athletes with and without shin splints. Am J Sports Med. 1983;11(3):125-30.
36. Willems T, Witvrouw E, De Cock A, De Clercq D. Gait-related risk factors for exercise-related lower-leg pain during shod running. Med Sci Sports Exerc. 2007;39(2):330-9.
37. Willems TM, De Clercq D, Delbaere K, Vanderstraeten G, De Cock A, Witvrouw E. A prospective study of gait related risk factors for exercise-related lower leg pain. Gait Posture. 2006;23(1):91-8.
38. Woltring HJ. A FORTRAN package for generalized, cross-validatory spline smoothing and differentiation. Adv Eng Softw. 1986;8:104-13.
39. Yates B, White S. The incidence and risk factors in the development of medial tibial stress syndrome among naval recruits. Am J Sports Med. 2004;32(3):772-80.
40. Zeni JA, Richards JG, Higginson JS. Two simple methods for determining gait events during treadmill and overground walking using kinematic data. Gait Posture. 2007;27(4):710-4.
Keywords:

ELECTROMYOGRAPHY (EMG); MUSCLE ACTIVATION; MUSCULOSKELETAL PAIN AND INJURY; MOTOR CONTROL; SHIN SPLINTS

©2010The American College of Sports Medicine