Ultrasound measures of intrinsic foot muscle size and motor activation following lateral ankle sprain and chronic ankle instability. A cross-sectional study

Design

A laboratory-based, descriptive cross-sectional study was performed. The independent variable was group (Control, Coper, LAS, CAI) and the dependent variables were IFM cross-sectional area at rest and motor activation ratios during non-resisted and resisted contraction. The relationships between IFM resting size, motor activation, and clinical measures of clinical measures of foot morphology, posture, and strength were also assessed.

Participants

A convenience sample of 80 recreationally-active individuals (Control: n=22, Coper: n=21, LAS: n=17, CAI: n=20) aged 18-35 with and without history of a LAS and CAI were recruited from a public university for participation from September 2016 to May 2017. The participants in this study were part of larger study of multisegmented foot function, with demographic and injury history have been previously reported.⁵ Table 1 details group demographic, injury history, and patient-reported outcome measures. There were no significant group differences in age, anthropometrics, or foot phenotype. “Recreationally-active” was defined as participation in some form of physical activity for at least 20-minutes per day, at least thrice weekly. Participants were included in the Control group if they did not have any history of ankle or foot sprain. Individuals who sustained an inversion sprain that affected function 2-8 weeks prior to consent were included in the LAS group. Participants with a history of at least one LAS at a minimum of 12-months prior to the study who did not experience perceived or episodic giving way and scored Identification of Functional Ankle Instability (IdFAI) ≤10, Foot and Ankle Ability Measure (FAAM) activities of daily living subscale (ADL) ≥99 and FAAM-Sport ≥97 were included in the Coper group.² Individuals with a history of at least one LAS at a minimum of 12-months prior to the study who experienced continued perceived or episodes of giving way and scored IdFAI >10, FAAM-ADL <90 and FAAM-Sport < 85 and did not sprain their ankle in the past 8-weeks were included in the CAI group.¹⁴ Individuals were excluded if they had any self-reported history of fracture or surgery in the leg or foot, self-reported disability due to neuromuscular impairment in the lower extremity, neurological or vestibular impairment that affected balance, diabetes mellitus, lumbosacral radiculopathy, a soft tissue disorder such as Marfan or Ehlers-Danlos syndrome, any absolute contraindication to ankle or foot joint manipulation, or were pregnant. Participants who met inclusion criteria provided informed consent. Recruitment was stratified to ensure relative equity in the distribution of males and females in each group. Figure 1 details the study flow sheet from recruitment to analysis. Data was collected in the university’s sports medicine laboratory. The study was approved by a university’s Institutional Review Board.

• Download figure
• Open in new tab

Figure 1.

Study flow sheet from recruitment to analysis. LAS, Lateral ankle sprain; CAI. Chronic Ankle Instability; FDB, flexor digitorum brevis; QP, quadratus plantae; FHB, flexor hallucis brevis.

Procedures

Following consent, participants provided demographic information, health and injury history, and completed the FAAM ADL¹⁵ and Sport subscales,¹⁶ Identification of Functional Ankle Instability (IdFAI),¹⁷ the Patient Reported Outcomes Measurement Information System (PROMIS) General Health Questionnaire,¹⁸ the 11-item Tampa Scale of Kinesiophobia (TSK-11),¹⁹ and the Godin Leisure-time Exercise Questionnaire.²⁰ Predicted EUROQOL (EQ-5D) quality of life scores were calculated using methods described by Revicki and colleagues.²¹ Height, mass, leg length, and foot morphology were measured and foot posture index assessed.²²

Demographic information, medical history, physical examination of the non-clinical group, and ultrasound imaging were performed by the primary author who was a physical therapist and was a board-certified orthopaedic clinical specialist with 15-years of clinical experience and one-year experience performing USI. Physical examinations of the LAS and CAI groups were performed by either an athletic trainer with three-years clinical experience or a physical therapist with two-years clinical experience who were blinded to the participants’ medical history and functional status. USI measurement was performed by two laboratory assistants familiar with the IFM anatomy and were blinded to the participant’s clinical history and state of contraction. In the LAS and CAI groups, the involved limb was assessed. In the case of participants with a history of bilateral injury, the self-reported worse limb was examined. The limbs of healthy controls were matched to the limbs assessed in the clinical groups.

IFM Muscle Size and Motor Function

USI of the abductor hallucis (AbdH), flexor digitorum brevis (FDB), quadratus plantae (QP), and flexor hallucis brevis (FHB) cross-sectional area during rest, contraction without external resistance at the end range of motion (non-resisted contraction), and maximum voluntary contraction against isometric manual resistance (resisted contraction) were performed as described by Fraser and colleagues.¹¹ Figure 2 depicts the transducer location and measured images of the IFM. USI was performed using a Siemens Acuson Freestyle US system with a wireless 8-Mhz linear transducer (Siemens, Mountain View, CA). Each muscle was imaged thrice at rest, during active contraction, and with resistance. Images were measured by two trained assessors, who were blinded to injury status and physical examination findings, using ImageJ version 1.50f (National Institutes of Health, Bethesda, MD). The average measurement of the three images were utilized for analysis in each condition. Measures of active and resisted contraction were normalized to resting measures and reported as an activation ratio,^11,12,27 calculated using the formula:

• Download figure
• Open in new tab

Figure 2.

Site of transducer placement during ultrasound scans of the 1. abductor hallucis (Abd. H.), 2. flexor digitorum brevis (FDB), 3. quadratus plantae (QP), and 4. flexor hallucis brevis (FHB). MT, metatarsal; TFL, total foot length. Adapted from Fraser JJ, Mangum LC, Hertel J. Test-retest reliability of ultrasound measures of intrinsic foot motor function. Phys Ther Sport. 2018;30:39-47. doi:10.1016/j.ptsp.2017.11.032. Used with permission of the publisher.

An activation ratio > 1.00 indicates an increase in muscle size and < 1.00 indicative of a decrease in muscle size in comparison to the muscle size at rest.

Statistical Analysis

The level of significance was p ≤0.05 for all analyses. A priori sample size estimation of 10 participants per group were needed to demonstrate large effects based on the minimal detectable change in IFM activation ratio of 0.21, a=.05, and =.20.¹¹ Descriptive statistics were calculated for group demographic and clinical outcome measures. Ultrasound images that were of poor quality were omitted from analysis.

A preliminary analysis was conducted to assess the effects of group alone on IFM size and activation, without consideration of potential mediating factors. Prior to group comparison analysis, Levene’s test of homogeneity of variance were performed. Variables that had significant homogeneity were assessed using a one-way analysis of variance (ANOVA). Variables that failed to achieve significance with Levene’s test were assessed using Welch’s ANOVA. For significant findings, post hoc Tukey’s Honestly Significant Differences and Cohen’s d effect size (ES) estimates were calculated for each clinical group compared to healthy controls. ES were interpreted using the scheme proposed by Cohen:²⁸ <0.2 equates to a trivial ES, 0.2-0.49 small, 0.5-0.79 moderate, and >0.8 large. Comparisons where the 95% CI around ES point estimate did not cross the ‘0’ threshold were considered statistically significant.

Correlation analysis comparing group, sex, anthropometric, foot composite variables to IFM size at rest and during contraction was performed to inform the development of a regression model. Associations between hallux and lesser toe flexion normalized strength variables and IFM were assessed to evaluate the potential utility of these measures as clinical surrogates. Relationships between continuous variables were evaluated using Pearson’s correlation coefficients. Assessment of relationships with categorical variables (group, sex) were performed using Spearman’s rho. Correlations were interpreted ≥.75 as good to excellent, .50-74 as moderate to good, .25-.49 as fair, and 0-.25 as little to no association.²⁹

To assess the contribution of the factors on resting, non-resisted contraction, and resisted contraction measures, three multiple linear regression analyses were performed for each IFM:

Data was analyzed using R Version 3.5.1 (The R Foundation for Statistical Computing, Vienna, Austria). Cohen’s d effect sizes, proportion estimates, and 95% confidence intervals (CI) were calculated using Excel for Mac 2016 (Microsoft Corp., Redmond, WA).

Resting Measures

In the assessment of injury history on resting IFM cross-sectional area measurements, group was not a significant factor in this study. This was somewhat unexpected as Feger and colleagues¹⁰ reported large decreased adductor hallucis obliquus and FHB muscle volume in individuals with CAI compared to healthy controls. It is plausible that methodological differences in how the IFM were measured is, in part, responsible for the disparity. Feger and colleagues¹⁰ used MRI image slices to reconstruct a three-dimensional image of each IFM used to calculate a volumetric measure. The FHB has a medial and lateral head, so it is plausible that MRI is better suited to capture volumetric loss in geometrically complex muscles. It is also plausible that disparity in self-reported ankle-foot function was also a factor. The participants in the study conducted by Feger and colleagues¹⁰ reported substantially higher disability on the FAAM. It is plausible that the participants included in our study had less neuromotor impairment compared to the sample studied by Feger and colleagues.¹⁰ Since LAS and CAI are incurred by a diverse population and are highly heterogenous conditions as it pertains to scope and severity of impairment,⁴ it is possible that there are other important covariates such as sex, body mass, and foot phenotype^9,13 that have not been previously considered.

Sex had fair associations with all resting IFM measures. These results are consistent with those reported by Abe and colleagues⁹ who assessed the relationships of IFM cross-sectional measures to toe strength. They found that men had significantly larger AbdH and FDB muscles compared to women.⁹ It is highly plausible that sex hormones, such as androgens, may also influence the size of the IFM. Males have higher testosterone levels, a primary androgen hormone, which is a primary factor for sex-related related disparities in muscle mass.³⁰

Both sex and foot phenotype were significant predictors of AbdH and FHB muscle size. Foot phenotype also contributed to resting FHB measures, with fair associations observed. It is plausible that these findings are related to sex-related differences in foot structure and function. Fukano and colleagues³¹ found that females had greater foot flexibility and deformation under load. Additionally, greater foot deformation has been associated with participants with higher BMIs.³² BMI was found to be a significant predictor for both AbdH and FDB muscle cross-sectional area, two muscles that have an important role in support of foot architecture. Therefore, it is important to consider the covariates of sex, body mass, and foot phenotype in evaluation of resting IFM size following ankle-foot injury.

Clinical Implications

These findings, when synthesized together, can be contextualized using the theoretical framework postulated by McKeon and colleagues.³⁵ In the “foot core” paradigm, there is an interplay of active and passive foot subsystems, with the active subsystem represented by the IFM and the passive subsystem provided by the multitude of joint capsules and ligamentous in the foot. The external moments imposed by body mass during foot deformation, which is modulated by sex-related joint laxity, is resisted by the IFM. Since single-point resting measures provide of IFM cross-sectional measures insight to associated atrophy or hypertrophy, it appears that resting measures are more a function of foot phenotype and external loads rather than injury history. The third facet of the “foot core” triad presented by McKeon and colleagues³⁵ is the neural subsystem. Clinical group was a significant factor for QP neuromotor function, represented by non-resisted and resisted activation.

Based on the findings of this study, routine examination of the IFM at rest and during activation using USI following LAS and CAI cannot be recommended at this time. However, clinicians may consider using these measures as a clinical correlate if the patient is suspected of having neuromotor impairment following injury (i.e. tarsal tunnel neuropraxia).

Limitations and Strengths

This study was not without limitations. The authors employed a cross-sectional design in the assessment of clinical group on resting IFM size and motor activation. As a result, causation cannot be implied. The sample consisted of young adult, nulliparous women and men with and without LAS and CAI, which limits external validity to similar populations. Since ankle-foot impairment varies in scope and depth following LAS and in CAI, heterogeneity of clinical presentation is likely. This study also had strengths worth mentioning. This is the first study to assess IFM size and neuromotor function and identify findings using a clinically accessible imaging modality. This study also considered multiple contributory factors during analysis, which provides greater insight to the clinical outcomes employed in this investigation.