|Home | About | Journals | Submit | Contact Us | Français|
The purpose of this study was to compare isometric subtalar inversion and forefoot adduction strength in subjects with Stage II posterior tibial tendon dysfunction (PTTD) to controls.
Twenty four subjects with Stage II PTTD and fifteen matched controls volunteered for this study. A force transducer (Model SML-200, Interface, Scottsdale, AZ) was connected with a resistance plate and oscilloscope (TDS 410A, Tektronix, Beaverton, OR) to the foot. Via the oscilloscope, subjects were given feedback on the amount of force produced and muscle activation of the anterior tibialis (AT) muscle. Subjects were instructed to maintain a plantar flexion force while performing a maximal voluntary subtalar inversion and forefoot adduction effort. A two-way ANOVA model with the factors including, side (involved/uninvolved) and group (control/PTTD) was used.
The PTTD group on the involved side showed significantly decreased subtalar inversion and foot adduction strength (0.70 ± 0.24 N/Kg) compared to the uninvolved side (0.94 ± 0.24 N/Kg) and controls (involved side = 0.99 ± 0.24 N/Kg, uninvolved side = 0.97 ± 0.21 N/Kg). The average AT activation was between 11–17% for both groups, however, showing considerable variability in subjects with PTTD.
These data confirm a subtalar inversion and forefoot adduction strength deficit by 20% to 30% in subjects with Stage II PTTD. Although isolating the PT muscle is difficult, a test specific to subtalar inversion and forefoot adduction demonstrated the weakness in this population.
Posterior tibial tendon dysfunction (PTTD) is characterized by insidious onset of swelling and pain along the posterior tibial (PT) tendon that can lead to adult acquired flat-foot deformity.12 Subjects have progressed to Stage II when signs of tendinopathy are coupled with a flexible flatfoot deformity.12 Progressive loss of muscle function in Stage II PTTD is suggested by abnormal gait patterns and difficulty with heel rise tasks, with little direct evidence of alterations in the muscle.4,21,30 Yet, new treatments for subjects with Stage II PTTD are targeting the effects of exercise on muscle/tendon recovery.4,16 Documenting muscle recovery is beneficial when assessing the effects of these new treatments on the PT muscle.28 Yet, because all deep compartment muscles (flexor hallicus longus (FHL), flexor digitorum longus (FDL) and PT) work synergistically to contribute to subtalar inversion, it is not always clear which muscle to attribute weakness to. Therefore, the challenge was to design a test that is sensitive to PT muscle weakness to mark the progression and/or recovery during the course of PTTD.
A forefoot adduction strength test may be sensitive to PT weakness due to muscle properties and joint mechanics. A key measure of the ability of a muscle to produce force is physiologic cross sectional area (PCSA).18 The PCSA takes into account muscle anatomic cross sectional area and the pennation angle of muscle fibers, providing a measure of the muscle fibers in parallel. A review of in-vitro3,7,25,31 and in-vivo8 estimates of the PCSA of leg muscles suggests the PCSA of the PT muscle is two to four times that of the FDL and FHL (Table 1). In addition to muscle PCSA, in-vitro studies suggest the moment arm of the PT muscle is significantly larger than the FDL and FHL for subtalar inversion.6,10,14 Unfortunately, the moment arms of the deep compartment muscles at the talonavicualar joint and other forefoot joints are not well understood.6 Flemister et al. demonstrated that forefoot abduction/adduction movements were associated with moment arm estimates of greater than 2 cm for the PT muscle, with muscle shortening occurring during forefoot adduction.6 In support of the importance of forefoot movement, an in-vivo study noted that subtalar inversion and forefoot adduction tasks have greater task specificity than 3 other tasks for isolating the PT muscle from other leg muscles.15 Therefore, because of PCSA, joint moment arms and task specificity, changes in PT muscle function are expected to overshadow the synergistic roles of the other deep compartment muscles when performing a strength test with a subtalar inversion and forefoot adduction task.
Another key element of strength testing of the PT muscle is co-contraction of the anterior tibialis (AT) muscle. Ankle plantarflexion is advocated to inhibit the synergistic contribution of the AT muscle when attempting to detect weakness.13 Ankle plantarflexion is thought to minimize the contribution of the AT muscle on subtalar inversion and forefoot adduction either through task specificity or by placing the muscle in a lengthened, and presumably mechanically disadvantaged position.13 However, no studies were found that investigated the AT muscle activation of subjects with PTTD during an ankle plantarflexion, subtalar inversion and forefoot adduction strength test. Effectiveness of this clinical strategy to detect PT muscle strength partly relies on the ability to minimize the contribution of the AT muscle, suggested by low activation during maximal efforts.
The purpose of this study was to compare isometric subtalar inversion and forefoot adduction strength in Stage II PTTD subjects to controls. The subtalar inversion and forefoot adduction task was hypothesized to be sensitive to PT muscle weakness because of muscle properties, joint mechanics and task specificity in subjects with Stage II PTTD. Ankle plantar flexion was expected to minimize the activation of the AT muscle. However, the potential for subjects with PT muscle weakness to compensate with AT activation was anticipated. A goal of nominal AT activation of less than 10% was set a priori for both subjects with Stage II PTTD and controls.
Twenty-four subjects with Stage II PTTD (18 females, 6 males) and 15 controls (12 females, 3 male) volunteered after giving informed consent. The subjects with PTTD were representative of the population, who frequently are middle aged or older, female, and overweight (Table 2). The control subjects were matched for age and body mass index of the first 10 PTTD subjects enrolled. The control subjects were required to have no known foot pathology and be free of lower extremity pain for the last 6 months. Inclusion criteria for the PTTD group required signs of 1) tendon pathology and 2) unilateral flexible flatfoot deformity. Signs of tendon pathology included pain and/or swelling along the medial ankle. Signs of flatfoot deformity were determined by clinical assessment with any one of the following, 1) greater unilateral hindfoot eversion, 2) greater forefoot abduction and/or 3) lower medial longitudinal arch height. To verify the subjects with Stage II PTTD on average had abnormal foot posture, the arch index was used (Table 2).32 The arch index is the height from the floor to the top of the foot at 50% of foot length divided by truncated foot length. Truncated foot length is defined as the distance between the most posterior aspect of the heel to the center of the first metatarsal phalangeal joint. The reliability and validity of this method has been reported in a previous study.32 Exclusion criteria included inability to walk greater than 15 meters or a history of previous foot or ankle pathology. A statistical power analysis suggested that sample sizes above 10 subjects per group were required to achieve 80% power with an expected effect size of 0.30 N/Kg and standard deviation of ±0.25 N/Kg.
To capture maximal isometric efforts, a force transducer (Model SML-200, Interface, Scottsdale, AZ) was connected in series with a resistance plate and oscilloscope (TDS 410A, Tektronix, Beaverton, OR). Calibration of the force transducer with known weights suggested low errors (r2 = 0.997, root mean square error = ±1 N). The resistance plate was mounted on ball bearing tracks to allow for movement in the medial/lateral direction. Thus, force exerted against the resistance plate was detected by the force transducer and displayed on the oscilloscope. A small pad, covered in moleskin, that fit the general shape of the medial forefoot was used to distribute pressure over the skin, allowing subjects to exert maximal efforts against the resistance plate with little discomfort. To assist subjects in maintaining a consistent ankle plantarflexion and leg position, an air stirrup brace (Aircast, Inc.) was utilized (Figure 1). However, the air stirrup brace was not rigidly fixed to the uprights, thus requiring subjects to achieve proximal stabilization through active muscle control. The role of the air stirrup brace was to assist the subject maintain a constant ankle plantarflexion angle and foot posture. Inability to invert the hindfoot or raise the medial longitudinal arch of the foot invalidated the test. The air stirrup brace was set so the heel was approximately 10 cm above the resistance plate, resulting in varying degrees of plantar flexion from 30 to 45 degrees depending on foot size. Because the ankle plantarflexion angle has little influence on PT muscle length, the influence of ankle plantarflexion on isometric strength was assumed to be negligible.6,14,20
Visual feedback of the amount of force exerted and muscle activity of the AT muscle for each effort was provided via an oscilloscope. After cleaning and abrading the skin, a surface electrode (DE-2.1, Delsys, Inc., Boston, MA) was placed in a standardized location (one hand width distal to the tibial tuberosity) over the skin of the AT muscle. The surface electrode was connected to a 2-channel EMG system (Bangoli-2 EMG System, Delsys, Inc., Boston, MA) and the gain was adjusted from 1 to 1000. The force and muscle activity were displayed on the oscilloscope to provide feedback to the examiner and subjects (Figure 1). Readings were taken directly from the oscilloscope which has a digital display that samples data at 1000 Hz. Peak to peak EMG amplitudes in volts were recorded for the maximum voluntary contraction (MVC) and strength testing trials (Figure 1).
The procedures included determining 10% AT activation and peak isometric subtalar inversion and forefoot adduction. Prior to performing a maximum inversion and forefoot adduction effort, three maximal ankle dorsiflexion efforts were recorded using manual resistance. As a benchmark, 10% of the highest peak to peak EMG signal of three MVC trials for the AT muscle was used as a target for examiners and subjects during the trials. To perform the ankle inversion and forefoot adduction effort, the subject’s foot was placed in the ankle stirrup, the foot plate was adjusted to align the metatarsal head with the medial malleoli, which placed the subject in a neutral or slightly abducted foot position (Figure 1). After being properly positioned, subjects performed five practice submaximal efforts and five maximal efforts. During each effort, subjects were instructed to maintain a plantar flexion moment while performing a subtalar inversion and forefoot adduction effort. Subjects were required to maintain neutral foot posture throughout the test. Visually this meant maintenance of the medial longitunal arch as judged by the examiner. This eliminated efforts that involved simply dragging the foot across the foot plate with proximal musculature. During the submaximal efforts, if AT muscle activity was high (exceeded 10% of MVC), subjects were coached to plantarflex more, and/or push medially with their forefoot, diminishing the contribution of the AT muscle. These commands were individualized to each subject, with the goal of minimizing AT activation. Once the practice trials were completed, visual feedback of the force and verbal encouragement were used to motivate subjects to exert a maximal effort. Peak force and peak to peak EMG amplitude for each maximum subtalar inversion and foot adduction effort was read from the oscilloscope and manually recorded. For all subjects the sequence of which side was tested was randomized.
Eleven subjects from the control group volunteered to perform the test again on a separate day. The second session occurred less than a week from the first session. This interval of time was sufficiently short to expect the same isometric strength measures for both sessions. All the procedures above were followed during each session. To minimize the chance for examiner bias, the examiner was not allowed to review the results of session one prior to performing the second session. One examiner tested all subjects.
The first analysis addressed reliability, while the second addressed the comparison of isometric strength between the controls and PTTD subjects. To assess the reliability of the subtalar inversion and foot adduction peak force, an intraclass correlation statistic (model 3, 1) was applied to the session one and session two data. For the first hypothesis, the dependent variable was the peak isometric subtalar inversion and foot adduction force of the 3 maximal efforts. All force values were normalized to body mass. A two-way mixed effects ANOVA model was used to test for differences between groups (between groups factor = PTTD vs. control) and sides (within groups factor = involved vs. uninvolved). For the control subjects, the right side was considered the involved side and the left the uninvolved side. Preliminary analyses suggested no side to side differences in control subjects justifying this approach. A significant interaction, indicating lower peak isometric force of the involved side of the PTTD subjects compared to the uninvolved side of subjects with PTTD and controls (both sides) was consistent with the main hypothesis. In the presence of an interaction, pairwise comparisons were used to test for differences across groups and sides. Additionally, to mirror clinical practice, where weakness is commonly assessed by comparing to the opposite side, the ratio of involved to uninvolved peak isometric subtalar inversion and foot adduction force normalized to mass was compared between the Stage II PTTD subjects and controls using a Student’s t-test.
Second, to assess whether the average AT muscle activation during the peak isometric subtalar inversion and foot adduction test varied, the dependent variable was the peak to peak EMG activation normalized to MVC (% MVC). Similarly, a two-way mixed effects ANOVA was used to test for differences between groups (between groups factor = PTTD vs. control) and sides (within groups factor = involved vs. uninvolved). A significant interaction, indicating higher % MVC on the involved side of the PTTD subjects compared to the uninvolved side of the PTTD subjects and controls (both sides) would suggest compensation with the AT muscle. In the absence of an interaction, a main effect for group, with the PTTD group showing higher AT muscle activation than the controls would indicate substitution not unique to muscle weakness. For all analyses (two-way mixed effects ANOVA’s and Student’s t-test), an alpha level of 0.05 was used for significance.
The between session reliability of the normalized peak isometric subtalar inversion and foot adduction test was high (Table 3).
There was a significant interaction between group and side (p = 0.003), that resulted from a lower peak isometric force of the involved side of the PTTD group (Figure 2). Normalized peak isometric subtalar inversion and foot adduction force for the PTTD group was significantly lower (p = 0.001) at 0.70 ± 0.24 N/kg compared to the uninvolved side at 0.99 ± 0.24 N/Kg. The involved side for the PTTD group was also significantly lower (involved p < 0.001 and uninvolved p = 0.001) as compared to both sides of the control group (Table 4). A post hoc analysis revealed a significant correlation (r = −0.55, p < 0.001) between age and normalized peak isometric subtalar inversion and foot adduction force. Because the groups showed a small difference in age of 5.5 years (Table 2), that was close to significant (Table 2, p = 0.07), the analysis was repeated with age as a covariate. With age entered as a covariate the significant interaction remained (p = 0.012) and no pairwise comparisons were changed. The ratio of the involved to the uninvolved side was also statistically (p = 0.003) lower in the PTTD group compared to the controls (Figure 3). Post hoc analysis suggested that the ratio of the involved to the uninvolved side was not related to age (r = −0.263, p = 0.10) so no further analysis was pursued.
The EMG activation of the AT muscle did not demonstrate a significant interaction (p = 0.314) or main effect for group (p = 0.324) (Table 4). However, the high standard deviations in the PTTD group compared to the control group caused us to examine AT activation more closely. In the PTTD group the percent of subjects with less than 10% MVC of the AT muscle was 54% and 46% on the involved and uninvolved sides respectively. Similar percentages were observed in the control group with 60% of subjects on both sides achieving less than 10% MVC of the AT muscle. The highest AT muscle activation of the control group on either side was 27% MVC. In contrast, three subjects in the PTTD group used an AT activation that was higher than 27% MVC on both sides.
The findings of this study suggest an isometric subtalar inversion and forefoot adduction weakness in subjects with Stage II PTTD as compared to controls. The results of this study show a subtalar inversion and forefoot adduction strength deficit of 20% to 30% in subjects classified with Stage II PTTD. While isolated testing of the PT muscle is difficult, a specific test based on muscle properties (Table 1), joint mechanics and task specificity was sensitive to weakness of the PT muscle. However, co-activation of the AT muscle was high despite targeting lower co-activation levels. It is assumed that pain intensity (which was not assessed in this study) during the testing may have also influenced maximum efforts. Subject positioning and proximal stabilization may also influence the specific strength test proposed. The reliability data (Table 3) suggest consistency in control subjects; however, consistency in subjects with PTTD needs to be improved for better clinical interpretation.
Four factors specific to muscle function including 1) PCSA, 2) moment arms, 3) muscle activation and 4) pain may contribute to the subtalar inversion and forefoot adduction test being sensitive to PT weakness in subjects with PTTD. The PCSA is a key morphologic measure of muscle that reflects the ability of a muscle to produce force. Of the deep compartment muscles (FDL, FHL and PT), the PT muscle accounts for 60% of the PCSA (Table 1). A study of subjects with PTTD just prior to surgery (an average of 40 months of failed treatment) documented fatty infiltration in 3/12 PT muscles and deficits in muscle volume of 2% to 17% compared to the uninvolved side.30 This study supports the influence of PTTD on muscle volume, and hence, PCSA. Also, in vitro studies document moment arms in cadaver subjects but not in subjects with PTTD or foot pronation.6,10,14 Nikki et al. documented the failure of the PT muscle to raise the foot to the same position after inducing flatfoot.19 One explanation for this decreased effectiveness of the PT muscle is a change in moment arms at either the subtalar or talonavicular joints. Studies of subjects after immobilization suggest weakness is a combination of activation failure and muscle atrophy.26 Some degree of activation failure may also exist in subjects with PTTD. Also, pain during the isometric strength test may inhibit maximal efforts, contributing to activation deficits. Collectively, these studies suggest that the subtalar inversion and foot adduction isometric strength deficit of 20% to 30% observed in subjects with PTTD is due to a combination of PT muscle weakness and/or pain. Using a subtalar inversion and foot adduction isometric strength test may provide an early indication of PT muscle weakness in subjects with PTTD.
The greater activation of the AT muscle by the PTTD group on the involved side suggests difficulty isolating the PT muscle during the task. The average AT activation was not statistically different between the control and PTTD groups. However, only 60% of the subjects in the control group and 40% to 50% of the subjects in the PTTD group achieved the target of less than 10% activation. The 3 subjects in the PTTD group with activation greater than 30% MVC showed severe weakness (ratio involved/uninvolved side ranged from 0.50 to 0.60), suggesting they may have been attempting to substitute for a weak PT muscle. For this analysis, AT activation was presumed to have the potential to mask weakness of the PT muscle. However, the amount of AT activation did not correlate to the deficit in PT isometric strength (r = −0.31, p = 0.14). This lack of correlation, and the observation that those with the highest AT activation demonstrated significant weakness (ratio involved/uninvolved side ranged from 0.50 to 0.60), suggests AT activation may not mask weakness for this task. Because EMG only indicates neural drive and not mechanical effects, this is possible. For example, the AT muscle is significantly lengthened in the test position decreasing its ability to generate force. Also, the AT muscle may not have a significant moment arm for forefoot adduction, neutralizing the mechanical effect of its activation. For example, the AT moment arm for rearfoot inversion has been estimated as 20% of the moment arm for the PT muscle.14 Nevertheless, less AT activation would indicate a more isolated measure of the PT muscle, which was not achieved across subjects with PTTD in this study. Other strength tests that use a dorsiflexed foot position, such as isokinetic testing, are speculated to induce even higher AT activation.
Recent studies are exploring the potential for exercise to alter the course of PTTD. Alvarez et al. reported an observational study that demonstrated promising outcomes using an aggressive exercise regime in subjects with Stage I and II PTTD.4 Others reported the design of a clinical trial that is underway to distinguish the benefits of eccentric versus concentric exercise for subjects with PTTD.16 These exercise programs partially base their approach on the outcomes observed with active treatment of the Achilles tendon.1,2 However, unlike Achilles tendinopathy, PTTD has been associated with damage to major foot ligaments5,9 and abnormal walking mechanics.11,22,23,27 Further, in-vitro studies suggest that the effectiveness of the PT muscle in controlling foot posture partially depends on the integrity of foot ligaments.19 Because of this dependence of the effectiveness of the PT muscle on foot ligaments, restoring the strength of the PT muscle may have positive but limited therapeutic benefits. To assess the effect of exercise programs, such as those proposed above, methods to quantify PT muscle recovery are needed. This study presents one testing method that appears sensitive to PT muscle weakness. However, future research is needed to determine test properties such as a clinical meaningful change and responsiveness to improve clinical interpretation. Clinicians are also advised that isolating the PT muscle fully from AT muscle activation is difficult, with a high percentage of subjects exceeding 10% activation.
The ability to truly isolate the PT muscle from other synergistic muscles is difficult. The air stirrup brace may have inhibited or enhanced the contribution of the PT muscle. Siegler et al.24 showed in-vivo that angular stiffness (degrees/Nm) increased markedly toward the end range of subtalar inversion and less so in the midrange. Because the task is isometric, the midrange stiffness is more relevant. The increased midrange stiffness of the air stirrup brace, although small, may have limited inversion motion, inhibiting the full contribution of the subtalar inverters. On the other hand, the air stirrup brace may have improved the positioning of the hindfoot in subjects with PTTD, increasing the moment arm of the PT muscle at the subtalar joint.15,19 Both in-vitro19 and in-vivo15 studies document greater contributions of the PT muscle when foot position is toward neutral. Future studies may consider comparisons across different testing positions. For example, Valderrabano et al.28 used a sidelying position when evaluating subjects with PTTD post operatively. The sidelying position also may provide greater proximal stabilization of the tibia. Further, without a direct comparison, it is unclear if isokinetic strength testing as used by Alvarez et al.4 is sufficient to capture PT muscle weakness. Alternative positioning and stabilization strategies may also assist with inhibiting the AT muscle. Because the AT muscle is an invertor, subtalar eversion with plantarflexion may further inhibit this muscle.
Other limitations relate to reliability testing, the sample used and limb dominance. A key limitation is that the reliability testing was only performed on control subjects. Reliability and responsiveness testing of subjects with PTTD is necessary to establish the utility of this test clinically. In addition, the clinical meaningfulness of a strength deficit is not clear. Future studies focusing on repeated measurements of isometric strength over time are needed. Another issue is the sample characteristics of the PTTD and control groups. Stage II PTTD is inclusive of a wide spectrum of disease severity.17 As new stages are defined, the PTTD group sample characteristics in Table 2 will be important to consider when comparing across studies. Also, the use of a matched control group on three key variables including gender, BMI, and age may influence the results, as demonstrated by the correlation of age and isometric strength. The effect of limb dominance was not assessed in this study. Some studies suggest that lower extremity dominance may account for some of the variance when comparing side to side.29 However, side-to-side differences due to dominance are unlikely to fully explain the 20% to 30% differences between the PTTD subjects and controls observed in this study.
The authors would like to acknowledge the funding received from the American Orthopedic Foot and Ankle Society and National Institute of Arthritis And Musculoskeletal And Skin Diseases (Grant Number R15AR054507) that was used in the study referred to here.
Although none of the authors has received or will receive benefits for personal or professional use from a commercial party related directly or indirectly to the subject of this article, benefits have been or will be received but are directed solely to a research fund, foundation, educational institution, or other nonprofit organization with which one or more of the authors is associated.