Incidence and risk factors of medial tibial stress syndrome: a prospective study in Physical Education Teacher Education students

Subjects

All subjects were first-year bachelor’s degree PETE students at the Amsterdam University of Applied Sciences. Apart from theoretical courses, the PETE programme consists of six different sports courses (gymnastics, field sports, martial arts, dance, athletics and swimming) with a weekly curricular exposition of approximately 11.5 hours. In total 285 students were invited to participate in the study during the academic year of 2016–2017.

Study procedure

A prospective study design was used with baseline measures (April–June 2016) and a 10-month follow-up period during the academic year of 2016–2017. Before the baseline measures, subjects were informed about the study procedures and were asked to complete an intake questionnaire and an informed consent. Information was obtained about sex, age, presence of MTSS, MTSS history, insole use, shoe type and participation in sports with high running and jump loads.

Baseline measures consisted of a necessary selection procedure and medical screening and an additional physical screening regarding this study. The selection procedure included a 3000 m running test supervised by teachers of the PETE programme. The medical screening included measurements of length (cm), weight (kg), BMI and fat percentage (%), and was conducted by sports physicians. The additional physical screening on the potential factors associated with MTSS was performed by bachelor’s degree Physical Therapy (PT) students who were trained by an experienced physical therapist (YF). This screening consisted of the following tests:

Navicular drop test was conducted after marking the navicular prominence in a seated (non-weight-bearing) position with the feet in shoulder width. The distance between the floor and the marked navicular prominence was measured in millimetres. After standing up (weight-bearing), without moving the feet, the measurement of the distance between the navicular prominence and the floor was repeated. The amount of drop for both feet was calculated by subtracting the seated score from the elevated score.14

Hip internal and external ROM measurements were conducted in a supine position with the hip and knee flexed to 90°. Each hip was internally and externally rotated to a firm feel. Angles were measured using a goniometer.15 The procedure was executed twice, and the scores were averaged per hip.

Hip adduction and abduction strength were measured in a side-lying position using a hand-held dynamometer (HHD) (JTech PowerTrack Commander I) using a 5 s break-test procedure.16 Before the test, the placing for the HHD was determined by marking 8 cm above the lateral and medial malleolus. The tested leg was held at a 0° angle, while the other leg was held at a 90° angle. Subjects were instructed to hold on to the side of the examination bed with their hands for stabilisation and to perform the test with maximal effort. The procedure was executed twice, and the scores were averaged per hip. An adduction:abduction ratio was calculated by dividing the average adduction score by the average abduction score.

Single leg squat test was conducted to determine varus or valgus angles. White markers (6) were placed at the spina iliaca anterior superior (SIAS), the greater trochanter, the lateral and medial femoral condyle, and the lateral and medial malleoli. Subjects were instructed to execute the single leg squat with their arms crossed in front of their chest with their hands on the shoulder, remain their knees in parallel and their vision forward. To provoke a natural execution of the squat, no technical instructions were given. After a series of practice squats, the squat test was executed twice for both legs and recorded on video (iPAD Mini, Apple, California, USA) in the frontal and sagittal planes. Videos were synchronised (Dartfish V.7, Alpharetta, Georgia, USA), and the knee angle (varus-valgus) and the squat angle were determined (Kinovea V.0.8.15) at the lowest squat position or the lowest position that the SIAS marker was visible. The results from the two consecutive squats were averaged per knee. Two-dimensional evaluation of the single leg squat knee angles has good reliability.17

Shin palpation for pain and oedema was conducted with the subject supine and the knee flexed in 90° and the foot on the examination bed. Place of palpation was determined by marking two-thirds of the distal medial surface of the tibiae. Presence of pain was assessed by palpating the posteromedial border of the tibia, and presence of pitting oedema was assessed by a 5 s hold of the medial tibial surface.18

Follow-up

During the follow-up period, subjects were asked to complete an MTSS registration form every five curricular weeks, in total on seven occasions. MTSS was defined as an exercise-induced pain at the medial side of the tibia.5 The registration form included the Dutch version of the MTSS score questionnaire, designed and validated by Winters et al (2016). The questionnaire contained an entry question regarding the presence of MTSS on the left leg, the right leg or both legs. When MTSS was indicated, the nature of the complaints had to be specified in four questions with four answer options: (1) limitations for participating in sporting activities (no limitations - no participation); (2) pain while performing sporting activities (no pain - unable to exercise); (3) pain during walking (no pain - unable to walk); and (4) pain at rest (no pain - very painful). Based on the answers an MTSS severity score (scale: 0–10, with 0 indicating no complaints and 10 indicating maximal complaints) was calculated.19

Data analysis and statistics

Subjects were included in the study when (1) at least three follow-up questionnaires were completed, and (2) only one or two questionnaires were completed, but with the indication of MTSS in at least one questionnaire. Legs with an MTSS severity score equal or higher than 1 in one or more of the follow-up questionnaires were considered as MTSS legs. Baseline test results were presented in mean and SD for continuous variables and numbers and percentages in dichotomous variables. Continuous data from men and women were combined after dichotomising the data as higher and lower than average for men and women separately. The association between the potential risk factors and MTSS was analysed at the leg level using logistic general estimating equation (GEE) analysis. GEE was used to take into account the dependency of the observations of the two legs within the subject. Both univariable and multivariable logistic GEE analyses were performed. Only variables with a p value lower than 0.20 in the univariable analysis were included in the multivariable analysis. P values smaller than 0.05 were considered significant. All statistical tests were performed with IBM SPSS V.24.

Comparisons with the literature

With an MTSS incidence of 25%, specifically 21% in men and 39% in women, we found relatively high results compared with other studies. Verrelst et al 6 found an MTSS incidence of 20% in female PETE students (n=81) during a follow-up period of 29 weeks.6 Sharma et al 4 found an incidence of 8% in male recruits (n=468) during a 26-week training period.4 Rauh et al 3 found an incidence of 7% in female Marine Corps recruits during a 13-week training period.3 Yates and White20found an incidence of 35% in 124 naval recruits, 26% for men (n=84) and 45% for women (n=40), during a 10-week basis training programme. In runners incidence rates between 14% and 20% are reported during follow-up periods of 12 months.21–24 Bennett et al 22 included 125 high school cross-country runners and found an incidence of 12% during an 8-week training programme. All the above-mentioned studies used a clinical diagnosis of MTSS. Except for the study of Yates and White20, all studies found a lower incidence compared with our study.

There are three major differences between these studies and our study. First, our study used self-evaluation of MTSS complaints, which may also be sensitive to several other injuries in the lower extremities (eg, tibial stress fracture, chronic exertional compartment syndrome, and muscle and tendon injuries).6 Most studies in MTSS used clinical diagnosis of the MTSS, and this methodological difference helps to explain the lower incidence of MTSS in the literature compared with our study. We attempted to minimise self-evaluation errors by using a validated questionnaire19 and a clear definition.5 Nevertheless, the self-evaluation procedure of MTSS complaints may be the primary explanation for the higher MTSS incidence in our study. Second, our study consists of a 10-month follow-up period, which is substantially longer than most of the previous studies. A longer follow-up period may be associated with higher incidence of MTSS. Third, most studies involve a different population (runners or militaries) compared with our study. The training regimens of these populations may be very different regarding training frequency, volume and type compared with the training regimens of PETE students.

Only one study, in militaries, found a higher incidence compared with our study.20 The authors report a weekly physical activity of 16 hours and explain that this is relatively high for militaries. Also, that study is unique in the confidentiality of the diagnosis of MTSS, meaning that the subjects were probably more comfortable in coming forward with any complaints because there were no consequences of reporting this injury. The high weekly physical activity and the confidentiality may explain the higher incidence found in that study compared with other studies in militaries and our study.

Winters et al 19 is the single study in the literature that reports responses per item of the MTSS score questionnaire.19 Their study shows that in MTSS-diagnosed patients (N=133), in 78% sports participation was reduced due to MTSS-related pain. Furthermore, the study shows that 97% of the patients reported pain during sporting activities, 69% reported pain while walking and 64% reported pain at rest. Our results are, respectively, 35%, 78%, 51% and 56%. This indicates that Winters et al 19 found a substantially higher severity of MTSS compared with our study. This can be explained by the fact that we used the MTSS score questionnaire to monitor complaints in a non-patient group. This logically leads to the inclusion of cases with mild MTSS. This principle is well documented in the literature.25

No studies regarding the risk factors of MTSS specifically in PETE students were found. A history of MTSS, with an OR of 5.03 (95% CI 1.90 to 13.30), is the most relevant risk factor in our study. Injury history is, in general, a robust risk factor for injuries in the literature.26 27 This is also the case for MTSS.10 11 Based on five prospective studies, Newman et al 11 report an overall OR of 3.74 (95% CI 1.17 to 11.91) for subjects with a history of MTSS to repeat occurrence of MTSS. Compared with this review study, our study found a relatively strong association between MTSS history and the reoccurrence of MTSS. However, ORs up to 18.3,28 20.029 and 30.030 can be found in the literature. Therefore, our results are still in agreement with the literature.

Review studies report that women (athletes, runners and militaries) are more likely to develop MTSS compared with men by 2.35 (95% CI 1.58 to 3.50)10 and 1.71 (95% CI 1.15 to 2.54)11 times. Our study found a relative risk of 3.14 (95% CI 1.39 to 7.11) for the female sex, which is slightly higher but in agreement with the literature. It is unknown why women are more predisposed to develop MTSS. Newman et al 11 suggest that differences in running kinematics between men and women may be attributed to the increased risk for women.

Most review studies on the risk factors of MTSS report a significant relationship between a higher BMI and MTSS risk.10–13 Our study found an OR of 2.29 (95% CI 1.02 to 5.16) for the group with an above-average BMI, which is consistent with the literature. An explanation for this finding is that a higher body weight relative to body height causes a relatively high mechanical loading to the tibia during weight-bearing activities.12 When this frequently occurs during a prolonged period, the body is unable to recover appropriately, producing bony overload and adhering complications.12

Our study found conflicting results with the literature regarding age as a risk factor for developing MTSS. The literature consistently reports that age is not associated with an increased risk.10 Our study found an OR of 0.31 (95% CI 0.13 to 0.76) for the above-average age group. Observations from practice are in line with this result. Therefore, this might be a specific risk factor in our PETE population. A rationale for this finding is that older students are more likely to sustain an acute lower extremity injury compared with younger students.31 Older and injured students may be less actively involved in the sports programme and therefore less susceptible to developing MTSS. However, we lack data to support this rationale.

All the other risk factors in our study did not have a significant relationship with the development of MTSS in our PETE students. These include height, weight, fat percentage, running performance, sports participation, insole use, use of supportive shoes, hip exorotation ROM, hip endorotation ROM, hip adduction and abduction strength, hip adduction-abduction strength ratio, navicular drop, squat knee angle, shin pain at palpation, and shin oedema.

Strengths and limitations

Our study has some limitations that need to be addressed. First, our study used a self-evaluation injury registration. This method may overestimate the actual incidence of MTSS. We attempted to minimise this effect by using a validated questionnaire and a precise definition. Second, our study managed to reach a response of 87%. The 13% missing may bias the results, underestimating the incidence of MTSS and the significance of the risk factors. Third, we did not control for sports exposition. Fourth, the physical screening test results may have limited reliability. Most of the conducted tests are well documented in the literature and have good reliability. However, we did not analyse inter-rater reliability during the training sessions. Thus, the results should be treated with care.

This study also had some strengths. First, we used a relatively large population compared with previous prospective studies, and we included both men and women. Second, we used a relatively extended follow-up period in comparison with most previous studies on MTSS.