The effect of repetitive exercise on tensile changes in the infraspinatus muscle evaluated with ultrasound elastography

Abstract

Background

The ability to identify muscle strain accurately is a critical aspect in managing exercise-induced muscle damage. Using differences in muscle stiffness as a proxy for muscle damage and recovery can better inform the decision to continue exercise versus pursue recovery.

Hypothesis/purpose

To identify changes in rotator cuff stiffness after a single bout of exercise as measured by ultrasound elastography.

Study design

10 healthy subjects, average age of 23.7, were enrolled in this prospective interventional trial. Each subject had a baseline scan then completed a standardized exercise program followed by another scan and daily evaluations until the results returned to within 10% of baseline.

Methods

Elastography measurements were collected around the infraspinatus tendon, and the deltoid was used as a baseline standard.

Statistical analysis used: Summary statistics, t-test, and regression analysis were completed for the data.

Results

Average strain was 0.5 at baseline, 0.7 immediately post-exercise, 0.6 on follow-up day 1, and 0.4 on follow-up day 2. There were statically significant differences between every timepoint. baseline-post 0.24, (p < 0.0001), post-day1 (decrease of 0.14, p < 0.0001), day1-day2 (decrease of 0.15, p < 0.0001), and BL-day2 (decrease of 0.05, p < 0.005).

Conclusions

These results demonstrate that the infraspinatus muscle stiffness increases immediately after exercise and takes up to 48 h to return to within 10% of baseline. We also found that at day 2 there continued to be an overcorrection with the infraspinatus remaining softer at this time point.

Background

Resistance exercise has been regularly shown to induce muscle damage. Eccentric exercise applied to a muscle results in muscle damage; reflected in a loss of maximal voluntary contraction, decreased torque and range of motion, and increased muscle stiffness [1]. While some degree of muscle damage is considered important for optimal muscle adaptation, excessive muscle damage can impair performance and further training for athletes [2]. There is an extensive body of research on the underlying mechanisms of muscle damage [1, 3]. Muscle stiffness has been shown to be elevated after exercise leading to muscle damage and a decrease in muscle function [4]. However, prior studies often measure stiffness indirectly through joint torque stiffness, which can be impacted by other factors outside of muscle stiffness (tendons, ligaments, joint capsule). More recently, ultrasound elastography has been used to analyze muscle elastic properties [5].

Ultrasound elastography measures the elastic properties of tissue and expands diagnostic options by allowing evaluation of muscle and other soft tissue stiffness and ability to respond to outside stresses [6, 7]. With the addition of elastography to traditional ultrasound, we may be able to further delineate the elastic properties of the tissue giving us insight into potential areas that are rested, stressed, recovered, or overloaded. While ultrasound elastography has recently been used to determine when muscle damage has occurred, an important area that has not been rigorously studied has been the use of ultrasound elastography in understanding muscle recovery after exercise [5, 8–10].

The use of elastography to examine recovery has been supported in animal studies using the rat model. A 2019 study found that muscles initially expressed increased stiffness in the first two days after injury which was followed by a steady decline in muscle hardness and a return to baseline at day 10 [11]. Another study detected a direct correlation between muscle injury and subsequent recovery using elastography [12]. An injury group was compared to a recovery group and a control group, and the injury group showed elastography changes that followed the typical pattern of muscle stiffness peaking at day 3 and normalizing by day 10 [12]. Furthermore, the injury group had greater detectible tendon changes (increased collagen) when compared to the recovery and control groups [12]. Preliminary studies to determine feasibility of a human model for recovery were successful but demonstrated significantly different recovery periods [11–13]. In the human model, recovery appears to peak at day 2 and resolve by day 4 [13]. However, this study looked at a small sample of adults and lacked a control group or comparison to baseline [13].

The ability to identify muscle strain early and accurately is a critical aspect in managing exercise-induced muscle damage. Using differences in muscle stiffness as a proxy for muscle damage and recovery can better inform the decision to continue exercise versus pursue recovery [5]. Using this same principle to evaluate when muscle stiffness has returned to baseline provides more data in understanding muscle recovery and could be invaluable for injury prevention.

Our study sought to use ultrasound elastography to quantify muscle recovery in the shoulder using the infraspinatus muscle as a model. This study evaluated infraspinatus muscle strain values after a single standardized workout and tracked changes in muscle elastic properties until muscle recovery was achieved.

Hypotheses and aims

Aim: To analyze change in infraspinatus (IS) stiffness after a bout of exercise using compression elastography.

Hypothesis for Aim: We hypothesized that the IS will become stiffer after exercise and return to normal in two (2) days.

Methods

Study design

This is a longitudinal descriptive research study.

Participants

Ten participants were selected by the primary investigator from a group of volunteers. Both shoulders were included in the study. Exclusion criteria included anyone below the age of 18 or above the age of 80, individuals with a previously diagnosed rotator cuff injury, previous rotator cuff surgery, or self-reported shoulder injury were excluded.

Procedures

This study was approved by the Office of Human Research Ethics (Study # 20–1042, Reference ID 350285) on February 04, 2022. Written consent was obtained from all participants prior to intervention. Both shoulders were scanned obtaining baseline measurements. Subsequently, participants were led through an approved standardized exercise program lasting 18 min. Participants were taught the exercise program at the Orthopedics Sports Medicine Clinic. Each participant performed three sets of 12 repetitions of the following exercises:

1. Resistance band face pulls

2. Resistance band adductions

3. Resistance band pull apart

4. Weighted shoulder circles

5. Resistance band isometric external rotations with elbows at ribs

6. Weighted adduction with elbows bent and forearm perpendicular to floor

The researcher directly observed participants to ensure full compliance. Participants participated in a single bout of the exercise program that immediately followed their baseline elastography scan. Participants were scanned at baseline, immediately after exercise, and each day after until the participants’ elastography scans returned to within 10% of baseline.

Conventional ultrasound and strain elastography were performed using a linear 8–14 MgHz probe and the Philips Affiniti 70G ultrasound machine.

Position of shoulder for US evaluation:

Participants rested their forearm across their lap with their palm facing upward. With the US held in long axis and the probe parallel to the ground, the probe was placed on the posterior shoulder. Conventional ultrasound was used to identify the humeral head, glenoid, and infraspinatus in the longitudinal axis. Following the infraspinatus from its humeral attachment medially, the thickest portion of the infraspinatus muscle was identified. While maintaining the thickest portion of the IS within the frame, the probe was rotated 90 degrees (probe long axis now oriented vertically) to obtain a short axis view of the IS. The infraspinatus tendon was identified within the body of the muscle.

Strain elastography was used to obtain mean strain values cranial, caudal, medial, and lateral to the infraspinatus tendon. The deltoid was also measured and used as a reference value for comparison given its consistency when compared to variations in subcutaneous tissue amounts based on body habitus.

Data analysis

Using the Phillips Affinity 70G ultrasound machine, a recording with overlying elastography analysis was obtained. A 30-cine clip was selected that reflected the most consistent and low-pressure readings on the compression bar. Using the recording allowed for more comprehensive comparison and reduced researcher bias. Once the video was obtained, 5 mm x 5 mm squares were placed cranial to the infraspinatus tendon, caudal to the infraspinatus tendon, medial to the infraspinatus tendon, and lateral to the infraspinatus tendon (Fig. 1 and 2).

Fig. 1.

Examples of ultrasound strain analysis before exercise

Fig. 2.

Examples of ultrasound strain analysis after exercise

Figure 1 | The image on the left represents the view of the shoulder obtained prior to use of elastography. The image on the right represents the five regions of interest. The yellow, green, blue, and orange boxes are the four measurements directly around the tendon. These are the cranial, caudal, medial, and lateral elastography values of the infraspinatus muscle. The multi-colored box in the most superior position is the deltoid measurement that serves as a baseline for comparison. The bar on the right side of the screen shows the color interpretation of elastography with red being the hardest muscle values and blue being the softest. This ensures that the deltoid can be used as comparison because it is predominantly blue, ensuring correct positioning of the arm.

Figure 2 | The image on the left represents the view obtained prior to use of elastography but after exercises had been performed. The image on the right represents the five regions of interest. The two orange, green, and lower blue boxes are the four measurements directly around the tendon. These are the cranial, caudal, medial, and lateral elastography values of the infraspinatus muscle. The blue box in the most superior position is the deltoid measurement that serves as a baseline for comparison. The bar on the right side of the screen shows the color interpretation of elastography with red being the hardest muscle values and blue being the softest. This ensures that the deltoid can be used as comparison because it is predominantly blue, ensuring correct positioning of the arm.

These locations served as our regions of interest and were compared to the reference point of the deltoid. Effort was made to ensure the standardized squares were not placed directly below the reference point in the deltoid. This was done to avoid any non-experimental changes in elastography of the reference range based on its direct proximity to infraspinatus tissue under strain. A video clip and image were saved for each participant for qualitative analysis and for use in presenting the data.

Mean strain values for each of the four areas of interest were recorded. An average of the four reference areas of interest was computed. Mean strain of the area of interest in the deltoid was also recorded. Five ratios were calculated. One ratio was calculated comparing each of the areas in the infraspinatus muscle to the reference area in the deltoid. One ratio compared the average of all four reference areas within the infraspinatus to the reference area in the deltoid.

Data was analyzed including demographic information, summary statistics, a regression analysis, and a graphical representation. Each of the ten participants provided data for each of their shoulders, twenty measurements were analyzed as unique sets of values. Significance was deemed to be a p-value less than 0.05.

Results

Demographics

Table 1 represents the demographic information of participants included in the study. Male participants were 70%of the sample. The average age of participants was 23.8. On average, participants worked out 3.9 times per week.

Table 1:

Participant Demographics

Qualitatively, each participant was asked to describe the types of other workouts that they participated in. Table 2 represents the various responses and frequency. Importantly, there was no cap on the number of qualitative responses, so participants gave anywhere from zero to three responses. Sixteen participants participated in weightlifting weekly, five participants running weekly, and a broader mix of other cardiovascular-focused activities (soccer, swimming, and basketball) than being reported by one to four participants.

Table 2:

Other Activities Performed by Participants

Summary statistics

Table 3 represents the summary statistics including mean, standard deviation, median, and range of strain values. These values represent the average value of all four locations on the infraspinatus muscle divided by the baseline deltoid measurement to achieve a ratio value. These values were reported at four separate times: baseline, immediately after the exercise intervention, at follow up one day later, and at follow up two days later.

Table 3:

Summary Statistics

Regression analysis

Regression analysis was performed with the linear regression model as follows:

$$E\left[Y_{\mathit{ij}}\right]={\beta }_0+{\beta }_{1}1\left(Visi{t}_{\mathit{ij}}=POST\right)+{\beta }_{2}1\left(Visi{t}_{\mathit{ij}}=FU1\right)+{\beta }_{3}1\left(Visi{t}_{\mathit{ij}}=FU2\right)$$

where $\beta =\left\{{\beta }_0,..,{\beta }_3\right\}$ are regression parameters, the average value is defined as Y_ij for subject i at visit j. The explanatory variables in the models are as follows Vist_ij is the visit label of subject i at j, Slide_ij is the side subject i at the jth visit (abnormal vs normal).

Considering that observations from the same person may be correlated over time, the working correlation matrix is assumed to be exchangeable, with observations from different clusters (study participants) assumed to be independent. To handle longitudinal data, generalized estimating equations (GEE) is used to estimate regression parameters and their empirical standard errors. The regression results are summarized in Table 4 where strain values represent the average value of all four locations on the infraspinatus muscle compared to the deltoid. Adjusted for within subject effect, mean average ratio found post-exercise is 0.24 higher than that at baseline; mean average ratio at the first follow-up is 0.11 higher than that at baseline but 0.14 lower than that at post-exercise; mean average ratio at the second follow-up is 0.05 lower than baseline and 0.15 lower than that at first follow-up. Each comparison of ratios was found to be statistically different at a level of ɑ = 0.05.

Table 4.

Regression Results

Graphical representation of strain ratios

Table 5 represents the different ratios used to visualize strain of the infraspinatus muscle compared to the deltoid. The average ratio graph combines all four measurements divided by the strain on the deltoid. Other findings include differences in peak values with the Average and Medial Ratio graphs peaking at above 0.7 and 0.8, respectively. The other three ratios appear to peak at 0.7. Finally, every ratio except the Caudal Ratio finds the infraspinatus muscle softer than baseline at Follow-Up visit 2.

Table 5:

Mean ratios for the 20 measured strain values (10 participants, bilateral shoulders):

Discussion

Ultrasound elastography is becoming an increasingly used modality to assess muscle damage in real-time and in a point-of-care setting rather than relying on more invasive measures including muscle biopsy. Ultrasound elastography allows investigators the ability to calculate muscle stiffness changes without the need for estimations that use less specific data including joint torque stiffness calculations that may be influenced by other factors outside of the muscle. For this reason, it is important to have a framework for analyzing muscle strain via ultrasound elastography that can be easily replicated, performed quickly and reliably, and can be used to analyze different aspects of muscle damage, namely muscle recovery. As such, it was the goal of this study to provide a clear method for assessing muscle strain on the infraspinatus muscle using ultrasound elastography. This study also expands upon pre-existing research to create a method for trending muscle stiffness immediately after a single bout of shoulder-focused exercise and until recovery is achieved. Understanding how muscle stiffness changes immediately after exercise and continues to change in the following days may be a key factor in identifying early muscle strain and preventing more serious injury in individuals who perform repetitive tasks e.g. athletes.

Our study found a significant difference in muscle stiffness with muscles being initially stiffer after exercise, progressively softening, and returning to within 10% of baseline by follow-up day 2. Interestingly, while participants were within 10% of their baseline values two days after exercise, their infraspinatus muscles were significantly (P < 0.0001) softer than baseline. This suggests that there may be a “rubber-banding” effect where the muscle stiffness over-corrects and becomes temporarily softer at some time after exercise. These results were corroborated by linear regression and by visual analysis of line graphs plotting average strain over time.

Significance

There is a lack of current research literature on measuring muscle recovery with point-of-care ultrasound elastography. Our study supports a novel way to trend muscle strain on the infraspinatus muscle allowing for better visualization of recovery from muscle damage in an outpatient setting with ultrasound capability. The fact that the average ratio, which was used for data analysis in this study, appears to visually correlate with the ratios of each individual measurement suggests that our method of averaging four measurements prior to comparing to the deltoid baseline is an appropriate method for approximating infraspinatus muscle strain. These findings support the validity of using a standardized approach to measuring infraspinatus muscle strain over several days and are easy to follow and replicate with small amounts of ultrasound training. The ability to establish a baseline, follow muscle strain after activity, and to predict recovery using strain ratios is a novel and important tool that could be used by healthcare providers of all types to more accurately predict muscle injury risk. Most importantly the finding that requires further investigation is the over correction we identified at day two. It is unclear what the clinical significance of this is and further research into this and muscle injury is needed.

Limitations

There are a few limitations to be noted in this study. There was no way to guarantee that participants avoided exercise or extra activity during the days between scans while participating in the study. Researchers attempted to control for this variability by ensuring that each recording used in the study was done by one trained researcher. Additionally, the study was limited in sample size and needs to be repeated with a larger sample and a control group.

Future directions

The most important future directions that could be guided by this research would be the creation of studies that correlate elastography values and clinical relevance. This study finds statistically different strain values after exercise and during recovery, but we are unable to comment on if this difference results in different clinical outcomes.

Building on the results of this study, it would be important to replicate this study and follow participants after they return to within 10% of baseline to appropriately catch participants whose muscle stiffness over-corrected. We find that many participants’ infraspinatus muscles are softer than baseline but were unable to capture how long it takes to return to baseline given that we stopped following participants after two follow-up visits. Lastly, a future study could compare muscle strain during muscle recovery between a single bout of exercise and exercise that spans several days to more accurately mimic athletic practices.

Conclusion

Elastography remains promising in being able to visualize changes in muscle. The results of this study show that muscles get soft immediately after exercise and then progressively get harder until recovery. This is important both for expanding the body of research surrounding muscle elastography as well as in its practical applications relating to sports and injury.

Funding

This project was provided by the Carolina Medical Student Research Program. This work was supported by a short-term research training (STRT) grant from the National Institute of Diabetes, Digestive and Kidney diseases (NIDDK) grant number T35-DK007386 from the National Institutes of Health.

Data availability

Data can be accessed via request to corresponding author.

Declarations

Conflict of interest

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Ethical approval

This study was approved by the UNC Chapel Hill Office of Human Research Ethics (Study # 20–1042, Reference ID 350285) on February 04, 2022. All participants provided written informed consent prior to participating.