The role of hip abductor and external rotator muscle strength in the development of exertional medial tibial pain: a prospective study
- Ruth Verrelst1,
- Tine Marieke Willems1,
- Dirk De Clercq2,
- Philip Roosen1,
- Lennert Goossens2,
- Erik Witvrouw1
- 1Department of Rehabilitation Sciences and Physiotherapy, Faculty of Medicine and Health Sciences, Ghent University, Ghent, Belgium
- 2Department of Movement and Sports Sciences, Faculty of Medicine and Health Sciences, Ghent University, Ghent, Belgium
- Correspondence to Ruth Verrelst, Department of Rehabilitation Sciences and Physiotherapy, Faculty of Medicine and Health Sciences, Ghent University, De Pintelaan 185 3B3, Ghent 9000, Belgium;
- Received 24 August 2012
- Revised 17 December 2012
- Accepted 16 January 2013
- Published Online First 8 February 2013
Objective To prospectively identify proximal risk factors contributing to the development of exertional medial tibial pain (EMTP).
Methods Data were prospectively collected on healthy female students in physical education, who were freshmen in 2010–2011 and 2011–2012. 95 female students, aged 18.15±0.84, were tested at the beginning of their first academic year. Testing included isokinetic hip strength measurements of the abductors, adductors, internal rotators and external rotators. The follow-up of the individulas was assessed using a weekly online questionnaire and a 3-monthly retrospective control questionnaire. EMTP was diagnosed by an experienced MD (Doctor of Medicine). Cox regression analysis was used to identify the potential risk factors for the development of EMTP.
Results 21 individuals were diagnosed with EMTP during follow-up. The results of this study identified that decreased hip abductor concentric strength is a predictive parameter for the development of EMTP in females. More specifically, total work (p=0.010) and average power (p=0.045) for concentric abduction strength were found to be significant predictors for this lower leg overuse injury.
Conclusions Hip abductor weakness is a significant predictor for EMTP in women. Preventive screening methods for EMTP should therefore include this proximal contributing factor.
Exertional medial tibial pain (EMTP) is a common and complex overuse problem among civilian, military and athletic populations1–3 and is usually induced by repetitive lower leg strain imposed by loading during weight-bearing activities.4 EMTP is characterised by exertional pain along the posteromedial border of the middle and distal thirds of the tibia5 and can include diagnostic entities of medial tibial stress syndrome, tibial stress fracture, chronic exertional compartment syndrome and muscular and tendon injuries.6 ,7 The treatment of such an injury is often difficult, expensive and time-consuming.8 Hence, being injured can have a significant effect on a physically active lifestyle, and it can sometimes end activity-related careers entirely.4 The understanding of injury causation should therefore be of primary concern. In a systemic review on the prevention of medial tibial pain, Thacker et al1 concluded that there is a lack of objective evidence to support the use of existing interventions to prevent medial tibial pain. A primary reason for this scarcity of evidence is that risk factors for EMTP are not well understood.9
In general, risk factors are commonly divided into two categories: intrinsic and extrinsic factors.9 Neely10 concluded in a review on the intrinsic risk factors of the lower extremity (LE) overuse injuries that ‘overwhelming evidence’ supported several intrinsic risk factors, including female gender. Additionally, Chuter et al11 described the important role of distal and proximal contributions to LE injury. Women have been shown to exhibit significantly greater external knee adduction moments, greater knee valgus, hip internal rotation and hip adduction during landing and other athletic tasks.12 ,13 Furthermore, the contribution of distal and proximal factors to the specific lower limb injuries has been investigated by several studies.4 ,11 ,14 Excessive foot pronation has already been extensively described as a link in the development of numerous lower limb pathologies, including EMTP.4 ,11 In addition to the role of this distal-orientated model, the role of a proximal-to-distal directed mechanism in LE injuries in general has also been described in the literature.11 Nevertheless, the specific role of these proximal parameters in the development of EMTP has not been described widely.6
To date, no prospective studies have been published on the role of hip strength parameters in the development of EMTP specifically. The purpose of this prospective cohort investigation was therefore to determine hip strength-related risk factors for EMTP in a young physically active population. We hypothesised that female athletes who developed EMTP would have weaker hip abductor and external rotator strength, compared with individuals who did not develop EMTP.
The individuals in this study were 95 female students, who were freshmen in 2010–2011(n=52) and 2011–2012(n=43) in Physical Education at Ghent University, Ghent University College and Ghent Artevelde University College, Belgium. The mean age of these students was 18.15±0.84 years. All signed informed consent and knew the goals of the study. Exclusion criteria for participating in the study were: (1) pain, ache or soreness in the LE within the previous year; (2) surgery of the LE and (3) neurological problems that would affect LE function.
Hip muscle strength of the students was evaluated at the beginning of the academic year. Freshmen in 2010–2011 were followed throughout two academic years, and freshmen in 2011–2012 were followed throughout one academic year. During their education, the participants followed a similar sports programme, under similar environmental conditions, for 29 weeks/academic-year. In addition to this basic sports programme, the amount of extramural (physical activities beyond sports lessons at school) and non-supervised practice activities were also registered. This individual amount of sport participation was then used as time at risk for every person.
After the follow-up period, students were divided into two groups: an injury group of 21 individuals who developed EMTP and a control group of 63 individuals who did not have any LE overuse injury during this study. Of the 95 individuals, 11 developed other LE injuries and were excluded from the comparison, so in total 84 individuals were taken into account for statistical analysis.
The injured leg of the individuals who developed EMTP was used in the statistical analysis. If an injured individual developed bilateral symptoms, the ‘most painful’ leg based on a visual analogue score was taken into account as the injured leg. Those 21 injured legs were matched with legs of the control group. The number of non-dominant/dominant legs in the injured group was matched with the number of non-dominant/dominant legs in the control group by creating a similar ratio among the 63 participants in the control group; hence, leg dominance would not play a role in the statistical outcome. Therefore, the uninjured leg of the injured individuals and one leg per participant of the control group were eliminated at random until the number of non-dominant/dominant legs of the injured group was also present in the control group (figure 1).
In this study, an isokinetic protocol was used to evaluate the hip muscle strength. Concentric torque of the hip abductors, adductors, internal rotators and external rotators, as well as eccentric torque of the abductors and external rotators, were assessed using an isokinetic dynamometer (Biodex System 4). Every test contained two series of five repetitions. All tests were performed at 60°/s. The order of strength testing series for all individuals was (1) concentric abduction–concentric adduction, (2) concentric abduction–eccentric abduction, (3) concentric external rotation–concentric internal rotation and (4) concentric external rotation–eccentric external rotation. This order was chosen for purposes of experimental convenience and was assumed to have negligible effects on test–retest reliability.15
Before the actual testing, participants completed a 5 min warm-up on a cycle ergometer and were provided with detailed instructions for the strength procedure. This warm-up and explanation phase was followed by a familiarisation procedure for every set of strength testing. Every test contained two series of five repetitions: practice and actual testing. The first of those two series consisted of submaximal practice repetitions; hence, the second and actual maximal testing series could be performed in a smooth manner. Between the practice and the actual testing series, there was a 1 min rest-interval. Between the different sets of strength testing, however, there was a 3 min rest-interval.
For the abduction–adduction test, the participant assumed a functional standing position on the non-tested leg and performed the testing as described by Claiborne et al.15 The abduction–adduction range of motion was set from 10° of hip adduction to 30° of hip abduction. The participants were instructed to keep their toes pointed forward and their knee extended in order to prevent alterations in muscle recruitment and compensation during testing (figure 2).
The hip external–internal rotation test was performed with the participants seated and the hip and knee flexed to 90°. The range of motion was set from 10° of internal rotation to 10° of external rotation (figure 3).15
Verbal encouragement was provided during all tests. Before testing, the limb was weighed following the instructions from the dynamometer's operations manual, so that the influence of gravity effect torque on the data could be corrected. Peak torque to body weight (PT/BW), total work (TW) and average power (AP) were used for statistical procedures in this study.
Claiborne et al15 found moderate to high reliability (ICC range=0.62–0.89; SEM range=7.80–24.11 Nm) for all the movements exhibited in this study.
Injury registration and diagnostic criteria
The injury registration method and diagnostic criteria are very important in the recording of injuries. Therefore, a multilevel registration method and accurate diagnostic criteria were used.
Before the beginning of injury registration, students were well informed about the features of EMTP by using the following lay-term description: EMTP is a medial overuse complaint distal to the knee and proximal to the foot, caused by exertion.6 ,16 ,17
At first, a multilevel method for registering the injuries was used. As a primary registration method, the participants received a weekly reminder by email to register their injuries in an online questionnaire. In this online survey, the students could register their injuries, of which the localisation and other features could be specified. Once they registered the ‘lower leg’ for localisation of their injury, the individuals were asked to visit the MD for a further diagnosis. One fixed day every week, the MD attended the university and the participants could ask for a free consult.
‘Onset of the injury’ was not used for detection of EMTP, because in some cases it may be less obvious whether the injury should be classified as overuse or acute. Symptoms can have a sudden onset, although the injury may be the result of a long-term process instead of a specific, identifiable event.18 As a secondary registration method, participant interviews were conducted one-on-one in a quiet and private surrounding to double-check for the occurrence of EMTP every 3 months.
Second, the diagnosis of EMTP was performed on the following criteria: (1) an atraumatic occurrence of at least 1 week of medial tibial pain, exacerbated by running, (2) the presence of focal or diffuse palpation tenderness at the distal two-thirds of the posteromedial tibial border and (3) pain, ache or soreness of the posteromedial tibial border with possible functional limitation during physical active participation.6 ,18
At first, a Shapiro-Wilk test for normality was conducted and it showed that all parameters were normally distributed. Subsequently, Cox regression analysis (enter method) could be performed to look for significant contributors to the development of EMTP. Variables with a p value<0.05 in the Cox regression analysis were seen as significant predictors for EMTP. This approach has been chosen because Cox regression analysis assumes that risk factors affect injury in a proportional manner across time. Moreover, this method can adjust the fact that the amount of sport participation can vary between individuals.4 The time of sport exposure was measured from the start of the follow-up period until the injury, or the end of the follow-up period for students who were not injured or who dropped out of the education programme. In case of dropouts, the date of dropping out was taken into account for the individual time of exposure. Statistical analysis for this study was performed using SPSS (V.20.0), except for the calculation of the thresholds of hip muscle weakness, which may precipitate EMTP. To define these threshold values for both TWABC and APABC, a receiver operating characteristic (ROC)-curve analysis was performed using MedCalc—software.
During this study, 21 (22%) of the 95 individuals developed EMTP. Nine of them developed bilateral complaints. A total of 63 individuals did not sustain any overuse injury of the LE (66%) and were used as a control group. During the period of follow-up, three students dropped out of the education programme, none of whom presented with EMTP. Anthropometric data on the individuals are listed in table 1.
Cox regression analysis revealed that both TWABC (p=0.010) and APABC (p=0.045) were found to be significant predictors for the development of EMTP. No other parameters were found to be significant contributors to EMTP. Table 2 summarises the results of the Cox regression analysis. Table 3 shows the strength of the predictive value of both TWABC and APABC on the development of EMTP. The mean and SD values of both predictive values are shown in figure 4.
The results of the Cox regression analysis revealed that the hazard of developing EMTP at any time decreases by (1–0.991=0.009) 1% if TWABC increases by 1 joule (J). The hazard of developing EMTP at any time decreases by (1–0.963=0.037) 4% if APABC increases by 1 Watt (W).
Finally, the ROC analysis revealed a cut-off value of ≤253.1J for TWABC. Similarly, this ROC curve analysis showed a cut-off value of ≤52.4 W for APABC (table 4).
This is the first study to prospectively investigate proximal risk factors for the development of EMTP. The results of this study identified that decreased hip abductor concentric strength is a predictive parameter for the development of EMTP. More specifically, TW and AP for concentric abduction strength were found to be significant predictors for this lower limb overuse injury. The incidence of EMTP reported in our female population (22%) is in accordance with previous results.4
Since this is the first study to examine the role of these proximal isokinetic strength factors in EMTP, comparison with other studies is rather difficult. Moreover, in other studies, hip strength was mostly assessed with a hand-held dynamometer.14 ,19 Nevertheless, similar results can be found in the recent literature. Niemuth et al reported significant weaker hip abductor and flexor isometric strength of the injured leg compared with the uninjured leg in recreational runners. Several overuse injuries were included in the cross-sectional design of that study, and thus no specific cause-and-effect relationship could be established.19 Similarly, in a prospective study, Leetun et al found significantly less hip abduction and external rotation strength in athletes who sustained a lower limb injury. Despite the prospective design, no specific mechanism could be described on the role of hip strength parameters in the development of EMTP, since general lower limb injuries were included in the study.14
In order to interpret the results of this study, it seems interesting to describe a possible link between hip muscle weakness and lower leg strain, since EMTP usually is induced by repetitive lower leg strain imposed by loading during weight-bearing activities.4
Hip abductors play a very important role in LE alignment, both in frontal and transverse plane movements. Consequently, weakness of these hip muscles can result in altered motions of the femur, which may have an important influence further down the kinetic chain.20 ,21 In the frontal plane, hip abductors assist in stabilising the pelvis and the hip, and consequently, they prevent the movement into hip adduction during single limb support.14 Frontal plane movements such as frontal plane pelvic drop and femoral adduction may have implications for more distal biomechanical function. Female athletes already have been shown to have greater hip adduction and internal angular rotation during those single limb movements compared with male athletes.22 Therefore, female athletes with hip abductor weakness are more vulnerable to large external forces experienced by the hip and trunk, which reduces the ability to stabilise these segments.14 ,22 In addition to their important role in the frontal plane, abductors seem to contribute to the control of transverse–plane movements. Baldon et al23 observed that greater eccentric hip abductor torque exhibited not only less femur adduction but also less femur medial rotation. It is known that the main hip abductor strength generator is the glutaeus medius (GM), and this muscle may assist in external rotation (especially the posterior fibres).20 The link between abductor strength and the transverse–plane movements can potentially be explained by the GM. Therefore, it is possible that a deficit of the GM may increase femur adduction and, secondarily, the femur internal rotation.23 This model may thus explain an adducted and internal rotated knee position during single leg weight bearing.24–29
As aforementioned, hip muscle weakness can lead to altered hip movements,20 ,21 which may have consequences down the kinetic chain. Previous research indicated that these altered hip movements are the most important predictors of rearfoot motion.30 This important link between hip adduction, hip internal rotation and foot pronation has already been described in the literature.31 Moreover, excessive pronation has already been identified as a risk factor for EMTP, since excessive eversion may be associated with increased internal inversion moments as the inverter musculature attempts to control the motion.4 These moments may lead to excessive eccentric traction to the plantar flexor and inverter musculature.4 Nevertheless, in a cadaver study, Beck and Osternig32 demonstrated that muscle origins do not frequently correspond to the site of symptoms, so tractions to the tibia may not always be the direct result of traction of the lower leg musculature. In support of this notion, Stickley et al33 described deep posterior crural fascia as a potential site of traction rather than the origins of lower leg muscles. The link between the lower leg muscles and fascia is described in a study by Bouché and Johnson,34 in which fascial tenting with muscle tractions were demonstrated. In conclusion, resulting traction on the lower leg musculature may result in traction on the deep posterior crural fascia, which may be the explanation for traction conferred to the tibia. When each touchdown generates this traction on midtibial musculoskeletal structures, the musculoskeletal system may become overloaded and overuse injury of the lower leg such as EMTP may occur.4
Additionally, another interesting finding in the research is that tension strain on the medial tibia increases in fatigued individuals. In general, most overuse injuries follow muscle fatigue, when the capacity of muscles to protect bone from excessive loads is compromised.35 Moreover, muscle fatigue was directly correlated to significant increases in bone strain by Yoshikawa et al36 in dogs. Measuring the muscle fatigue of the LE was not part of the protocol during this study. Nevertheless, whole body fatigue affects tibial strains,35 and a link between lumbopelvic–hip muscle weakness and muscle fatigue has already been described in low back pain patients.37 Since the results of a study with low back pain patients cannot be generalised, further research concerning the link between hip muscle weakness, muscle fatigue and tibial strains is certainly recommended.
In conclusion on the link between hip muscle weakness and the development of EMTP, hip muscle weakness can lead to altered movements of the lower limb.20 ,21 This hip muscle weakness and resulting altered movements can compromise the ability to protect the bone against excessive loads.38 This mechanism may lead to increased tibial strains and the development of EMTP. As described in the literature, muscle fatigue plays an important role in EMTP,35 but this fatigue however was not identified in this study.
In this study, hip abductor weakness was found to be a predictive parameter EMTP, whereas external rotation strength was not significantly decreased at the injured side. A significant decrease in hip external rotation strength was not found in this study, probably due to the testing position during isokinetic strength evaluation. The abductor strength test was performed in a functional standing position, whereas the rotation test was performed in a sitting position. Consequently, several external rotators may not perform optimal strength results due to this non-functional position. Various biomechanical studies have described the reversion of the external rotary action of the piriformis and the anterior fibres of the glutaeus maximus into internal hip rotation as the hip is significantly flexed. Thus, a sagittal plane position of the hip can reverse the horizontal plane movements of the whole or parts of the external rotator muscles.39 ,40 Consequently, the non-significant results for hip external rotator strength may suggest the following conclusions: hip external strength is not a predictive parameter for EMTP, or more functional testing protocols for hip rotator strength are needed.
Additionally, only AP and TW parameters for concentric abduction strength were found to be significantly different between groups. Since controlling the movements of the LE seems to be an important function of the hip muscles,14 ,21 ,41 ,42 these muscles might be more challenged in a functional range output measure, rather than a point output measure. Therefore, the evaluation of parameters such as TW and AVP may be more relevant than the evaluation of PT.
As a limitation of this study, we note that these results concern female physical education students. Therefore, these results cannot be generalised to a civilian setting. Nevertheless, a female active population suffers disproportionately higher rates of EMTP.43 Studies concerning risk factors for EMTP in female populations should therefore be of primary concern. Furthermore, the intrinsic alignment of the femur was not measured, so confounding factors such as coxa vara and excessive anteversion deformities, which are common features of the female gender, were not excluded. Nevertheless, these deformities may be important as they can have an important influence on LE function and hip strength parameters.44 ,45 These parameters can be of importance for future research. Additionally, further research concerning proximal contributors to EMTP should also focus on the link between isokinetic strength parameters and both kinematic and kinetic changes during functional movements. Finally, it should be noted that functional and accurate strength-testing protocols including measurement of muscle fatigue are needed in order to develop comprehensive screening methods and prevention programmes.
This study is the first to identify concentric hip abduction strength parameters as risk factors for the development of EMTP in women. The role of hip abductors in controlling the movement of the lower limb seems to be important. External rotation strength parameters, in contrast, were not found to be risk factors for this lower limb overuse injury, probably due to the non-functional positioning during strength testing.
Whereas the role of both distal and proximal factors in specific lower limb injuries has already been investigated,11 more prospective studies are needed to both combine the investigation of these distal and proximal risk factors and identify the cause–effect relationships. Preventive screening methods and programmes should examine these distal and proximal parameters and apply them in order to reduce the incidence of EMTP.
What are the new findings—impact on clinical practice
This prospective study is the first to evaluate hip muscle strength parameters as risk factors for the development of exertional medial tibial pain (EMTP) in women.
The results of this study identified that decreased hip abductor concentric strength is a predictive parameter for the development of EMTP in women.
Controlling the movements of the lower extremity seems to be an important function of the hip muscles. These muscles might therefore be more challenged in a functional range output measure, rather than a point output measure.
Preventive screening methods and programmes should examine these hip strength parameters and apply them in order to reduce the incidence of EMTP.
The authors gratefully acknowledge Kathleen Vanden Borre and Tanneke Palmans for their assistance in analysis of the data.
Contributors RV was responsible for the overall content as guarantor; the conception and design of the study, or acquisition of data, or analysis and interpretation of data; writing and drafting the article or revising it critically for important intellectual content and final approval of the version to be submitted. TMW was responsible for the conception and design of the study, or acquisition of data, or analysis and interpretation of data; revising it critically for important intellectual content and final approval of the version to be submitted. DDC was responsible for the conception and design of the study, or acquisition of data and revising it critically for important intellectual content. PR was responsible for the conception and design of the study and revising it critically for important intellectual content. LG was responsible for acquisition of data and revising it critically for important intellectual content. EW was responsible for the conception and design of the study, or acquisition of data, or analysis and interpretation of data; revising it critically for important intellectual content and final approval of the version to be submitted.
Funding This research was funded by BOF-UGent 05V00910.
Competing interests None.
Patient consent Obtained.
Ethics approval The Ethical Committee of Ghent University Hospital approved the study.
Provenance and peer review Not commissioned; externally peer reviewed.