Sex-Based Differences in Tensiomyography as Assessed in the Lower Erector Spinae of Healthy Participants: An Observational Study

Abstract

Background:

Although there is mounting evidence on sex-linked differences in paraspinal muscle function, it is unknown whether sex-based variations in mechanical and contractile characteristics of the lumbar erector spinae (LES) can be monitored noninvasively in healthy participants at rest using tensiomyography (TMG).

Hypothesis:

Sex-specific effects in muscle displacement (Dm) and velocity of muscle deformation (Vd) will be observed via TMG assessed in the LES.

Study Design:

Observational study.

Level of Evidence:

Level 3.

Methods:

LES response was measured in a relaxed state in 40 healthy adults (20 females). Possible differences between the conditions were investigated using mixed-model analyses of variance. Two-stage hierarchical linear regression analyses were performed to predict the outcome of TMG Dm and Vd based on participant sex.

Results:

There were significant main effects of sex with large effect sizes for both TMG parameters, resulting from lower mean values in women compared with men (Dm, P < 0.01; Vd, P < 0.01). In contrast, neither the main effect of side (left vs right LES) nor the interaction between the side and sex reached significance (all P > 0.3). Introducing the sex variable in stage 2 of the regression analyses significantly improved the prediction of the TMG parameters (all ∆R² ≥ 0.18; all P < 0.01; all f² ≥ 0.29).

Conclusion:

Sex-based differences in muscle stiffness and contractile characteristics could be observed by TMG on LES muscles in healthy individuals at rest. The data suggest that these disparities are not exclusively attributable to anthropometric measures but may be linked to intrinsic sex-based differences in skeletal muscle characteristics.

Clinical Relevance:

We recommend implementing TMG in a clinical setting using the obtained results as a basis to factor for the patient’s biological sex when assessing effects of therapeutic/exercise regimens aiming at the optimization of myofascial tissue regeneration and performance.

Women and men differ in muscle function, such as muscular performance capacity, fatigability, and recovery.^15,19,52 One point of interest lies in the investigation of sex-based differences in morphology and biomechanical characteristics of the lumbar paraspinal muscles.^3,6,20,32,51 The lower paraspinal muscles (eg, lumbar erector spinae [LES]), which are embedded in the epaxial myofascial compartment of the fascia thoracolumbalis,⁵³ substantially contribute to dynamic control and stability of the lumbar spine.¹⁴ Therefore, understanding the morphofunctional characteristics of this complex myofascial junction, including a consideration for sex-specific differences, is important for optimizing therapeutic approaches and interventions for patients⁴⁴ and athletes.⁴¹ In this context, studies indicate distinct variations between the sexes in LES contractile characteristics⁵⁵ as well as lumbar myofascial stiffness.³⁷ This has been shown in healthy participants^16,24,37 and also in patients with low back pain.²⁴

It is conceivable that these differences have an influence on adaptive capacities to varying training interventions or in the clinical praxis.^6,15,19 Therefore, appropriate and validated measurement devices are required to identify potential sex-based effects on training interventions or treatments that aim to have an impact on muscle stiffness and contractile function. To quantitatively evaluate the contractile properties of LES muscles, a special variant of mechanomyography (MMG) called tensiomyography (TMG) may constitute an easy-to-manage approach. TMG records mechanical and temporal characteristics of the radial muscle displacement of superficial skeletal muscles in response to a single twitch provocation under isometric conditions.³¹ The device has been used as an indirect measure of muscle stiffness^5,11,30,40 and in the determination of fiber type distribution in select limb muscles.^8,9,46

Although TMG studies are increasingly being published, only limited information is available regarding sex-specific effects on the instrument’s variables.²⁸ We hypothesized that sex-specific effects in muscle stiffness–associated and contractile characteristics can be monitored via TMG.

Methods

Participants

Healthy female (n = 20; mean ± SD age, 39.2 ± 12.1 years) and male participants (n = 20; mean ± SD age, 36.3 ± 9.4 years) were recruited from the Faculty of Psychology and Human Movement Science, University of Hamburg. All volunteers were informed about the experimental setup of the study and any potential unpleasantness that might occur during the measurement procedure. All attendees signed an informed consent form. We used the following inclusion criteria: healthy male and female participants aged between 18 and 65 years without acute and chronic injuries to the musculoskeletal system as well as lumbopelvic dysfunction or treatment within the previous 6 weeks. Exclusion criteria were specific spinal pathologies, nerve root pain, systemic diseases, previous spine surgery, history of malignancy, heart conditions, implanted biomedical device (cardiac pacemaker), use of corticosteroids during the past 3 months, skin alignments, body mass index (BMI) >30 kg/m², and pregnancy.^1,23,38

Experimental Approach and Study Design

This observational study was based on the recommendations of the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement,¹² and the investigation was conducted in accordance with the guidelines of the Declaration of Helsinki.⁵⁴ The study protocol was approved by the local ethics committee and registered with the National Trial Register.

The participants’ bilateral lower LES muscles were assessed using TMG to evaluate the mechanical and contractile properties of the muscle. The sequence in which the measurements were carried out (right side first, left side first) was randomized by drawing lots. The room temperature was maintained constant at 23°C (±1°C). All participants were asked to abstain from caffeine intake for 2 hours preceding the assessment³⁹ and instructed to avoid fatiguing exercise for 48 hours prior to assessment to prevent possible confounding. The test participants laid in a prone position on an examination table with the arms resting at the side of the body and a foam pad placed proximal to the ankle joint to ensure 5° of knee flexion. The measurements were performed by a single qualified osteopath with more than 3 years of experience in TMG assessments. Additionally, the participants’ age, weight, height, and BMI were assessed, and their habitual physical activity (PA) level was recorded by oral request and grouped into 3 categories: sedentary (<1 hour cumulative time per week), moderate (1.5-3 hours cumulative time per week), and high (>3 hours cumulative time per week).²⁵

Tensiomyography Testing Protocol

A highly sensitive digital displacement transducer with a spring constant of 0.17 N/mm (GK 40; Panoptik) was positioned perpendicularly onto the muscle belly of the LES muscles at the level of the L3-L4 interspace. The location of the thickest part of the muscle bulk (approximately 2 cm lateral to the dorsal midline) was identified via palpation and visual inspection.^10,42 Two self-adhesive electrodes (Pals Platinum neurostimulation electrodes, model 895220, 50 × 50 mm; AxelGaard Manufacturing Co) were placed symmetrically in a distal and proximal distance of 1.5 cm from the sensor (interelectrode distance, 3 cm). After a single square wave monophasic 1-ms pulse was applied by use of an electrical stimulator (TMG-S1; TMG-BMC Ltd) with an initial stimulation current of 30 mA, TMG-OK 3.0 software (TMG-BMC Ltd) simultaneously recorded and displayed the muscle displacement graph.

The recordings were carried out under static and relaxed conditions. To avoid coactivation of neighboring muscles, the sensor position was adjusted according to the participant’s individual muscle dimensions at the outset of data collection, if necessary.¹⁷ The stimulation current was progressively raised by 10 mA until the individual maximal twitch response amplitude or the maximal stimulator output (110 mA) was reached.⁴⁵ To prevent effects of fatigue and potentiation, interstimulus intervals of ≥10 seconds were chosen between successive measurements,⁵⁰ and the state of relaxation between the stimuli was ensured through verbal instruction and visual inspection. The 2 highest twitch responses on the displacement graph of each participant were recorded and averaged for subsequent analyses.^45,50 The following parameters were extracted from the displacement graph: muscle displacement (Dm [mm]), defined as the maximal radial deformation during muscle contraction, and velocity of deformation (Vd [mm/s]), a relative measure of muscle contraction of deformation that was calculated as the rate of contraction between 10% and 90% of maximal Dm.^26,30 The included parameters have been found to be highly reliable (intraclass correlation coefficients, 0.96-0.99) as assessed for the LES at the L3-L4 level.²⁶

Statistics

Data analyses were performed using SPSS (Version 22.0; IBM Corp). Descriptive statistics were calculated for all variables of interest (age, weight, height, BMI, PA level, Dm, Vd, right and left side). In addition, 95% CIs around the mean were calculated for the TMG variables Dm and Vd. All quantitative data were normally distributed (Kolmogorov-Smirnov test; all P > 0.05), and all variances homogeneous (Levene test; all P > 0.05 except for BMI [P = 0.02]). Independent-samples t tests (Welch t test for BMI) were calculated to evaluate differences between men and women for age, weight, height, and BMI. The Pearson chi-square test was used to compare the distribution of PA levels between sexes. The TMG parameters Dm and Vd were investigated using mixed-model analyses of variance (ANOVAs), crossing the factors sex (male vs female) and side (left vs right LES). Partial eta-squares (h2p) were calculated as effect sizes of the ANOVA effects. To rule out the possibility that sex-based effects in the TMG variables might be solely attributable to other sex-specific differences in characteristics such as height and BMI, a 2-stage hierarchical multiple regression was calculated for each assessed TMG parameter. To evaluate the improvement of the prediction by addition of the variable sex, change in R² (∆R²) from stage 1 to stage 2 was tested for significance. Cohen f² was calculated to determine the effect size of the change. Statistical significance was set at P < 0.05 for all statistical analyses.

Results

Anthropometric Characteristics

The anthropometric characteristics for male and female participants are outlined in Table 1. Men and women differed significantly in height, weight, and BMI but not in age or PA level.

Table 1.

Participant characteristics

	Female (n = 20)				Male (n = 20)
	Mean	SD	Min	Max	Mean	SD	Min	Max	P
Age, y	39.2	12.1	24	65	36.7	9.4	25	55	ns
Weight, kg	63.6	9.2	50	80	81.5	7.2	68	94	<0.01
Height, m	1.69	0.04	1.61	1.75	1.83	0.05	1.73	1.94	<0.01
BMI, kg/m2	22.0	2.4	18.6	27.7	24.4	1.4	21.2	26.6	<0.01
Physical activity level, %									ns
High	35	(n = 7)			30	(n = 6)
Moderate	65	(n = 13)			70	(n = 14)
Sedentary	0	(n = 0)			0	(n = 0)

BMI, body mass index; Max, maximum; Min, minimum; ns, not significant.

Descriptive Statistics

The descriptive statistics of all TMG variables are displayed in Table 2. Figures 1 and 2 illustrate the average Dm and Vd, respectively, for both sexes.

Table 2.

Descriptive statistics of TMG Dm and Vd for all participants

	Female (n = 20)				Male (n = 20)
Variable	Mean (95% CI)	SD	Min	Max	Mean (95% CI)	SD	Min	Max
Dm, mm, L	3.09 (2.94-3.70)	1.28	1.34	5.29	4.78 (3.93-5.62)	1.81	2.20	9.47
Dm, mm, R	3.00 (2.37-3.64)	1.35	1.16	5.75	4.89 (4.06-5.72)	1.77	1.79	9.17
Vd, mm/s, L	136.77 (105.97-167.57)	65.82	49.96	268.19	234.06 (194.34-273.77)	84.85	96.17	399.37
Vd, mm/s, R	131.21 (101.72-160.71)	63.02	36.20	252.47	243.14 (201.16-285.12)	89.70	107.59	413.68

Dm, muscle displacement; L, left; Max, maximum; Min, minimum; R, right; TMG, tensiomyography; Vd, velocity of deformation.

Figure 1.

Mean maximal muscle displacement (Dm) in male and female participants of left and right erector spinae (ES). Error bars correspond to SEs.

Figure 2.

Mean velocity of deformation (Vd) in male and female participants of left and right erector spinae (ES). Error bars correspond to SEs.

Sex-Based Differences in the TMG Variables

The mixed-model ANOVA revealed a significant main effect of sex, with large effect sizes for both parameters, resulting from significantly lower mean values in women compared with men (Dm: F_{(1, 38)} = 14.03; P < 0.01; h2p = 0.27; Vd: F_{(1, 38)} = 20.44; P < 0.01; h2p = 0.35). Neither the main effect of side nor the interaction between side and sex reached significance for either parameter (all P > 0.3).

The results of the regression analyses provided further confirmation for the research hypothesis. Introducing the variable “sex” in stage 2 of the regression model significantly improved the prediction of all TMG parameters by showing significant changes in R² from stage 1 to stage 2 (Table 3). An alternative calculation methodology of Vd as established by Loturco et al²⁹ yielded analogous results for all analyses—ANOVA and regression.

Table 3.

Summary of hierarchical regression analyses (N = 40)

	Stage 1 a		Stage 2 b		Change in R2
TMG Variable	R 2	P	R 2	P	∆R2	P	Effect Size f2
Dm, mm, L	0.16	0.04	0.35	<0.01	0.20	<0.01	0.29
Dm, mm, R	0.20	0.02	0.38	<0.01	0.18	<0.01	0.29
Vd, mm/s, L	0.19	0.02	0.42	<0.01	0.23	<0.01	0.40
Vd, mm/s, R	0.21	0.01	0.45	<0.01	0.24	<0.01	0.44

Dm, muscle displacement; L, left; R, right; Vd, velocity of deformation.

^aStage 1 predictors: height and body mass index.

^bStage 2 predictors: height, body mass index, and sex.

Discussion

As hypothesized, we were able to observe that muscle contractility of the bilateral LES, as assessed via TMG, differs between healthy male and female participants. The observed lower mean values of Dm in women compared with men may appear plausible when considering the lower muscle density accompanied by a smaller cross-sectional area (CSA) of the LES muscles.^20,49,51 The greater CSA of the paraspinal muscles in men^20,33 possibly leads to higher LES displacement in response to the electrical stimulus in TMG application. On the other hand, based on previous findings, it might seem paradoxical that the Dm amplitude was lower in women than in men, since a higher value of this measure is considered indicative for less rigidity and stiffness of a muscle.⁴⁰ Accordingly, the scant studies investigating sex-specific effects on TMG measures reported lower mean values for this indirect measure of muscle stiffness in men than in women.^34,43 It must be noted, however, that available data rely mostly on investigations of lower limb muscles, whereas the current study investigated the LES. Considering that each human muscle is unique due to varying muscle fiber distribution, the response to an external stimulus or load varies between muscles.¹⁹ Moreover, it should be kept in mind that men and women differ in regional adipose tissue distribution.⁴⁷ Consequently, it is conceivable that a higher percentage of body fat, with different patterns of distribution, played a crucial role in the cited studies by causing larger noncontractile tissue oscillations after contraction in the female participants.

The assessed sex-based differences in Vd may be mediated by sex-linked variations in muscle fiber composition¹⁸ since type I muscle fibers have a slower contraction speed in comparison with type II fibers and it is well-established that men exhibit larger fibers in the LES, whereas women have a greater proportional area of fiber type I.³² Moreover, the monitored disparities in Vd seem reasonable in view of the aforementioned aspect that men exhibit larger CSA of LES,^20,33 characterized by greater force generation capacity,²¹ which may in turn have led to the observed higher velocity of deformation in males. Interestingly, in a comparable MMG study,⁴⁸ the authors did not observe any sex-based differences for the analyzed MMG parameters in the LES. Given the different methodological approaches (eg, varying starting position, younger population described as “just being healthy,” different interelectrode distance, variations between the measurement devices), a straightforward comparison between the 2 investigations may, however, not be appropriate.

To summarize, the present study was able to observe significant sex-linked effects on muscle stiffness and velocity of deformation of the LES muscles as assessed via TMG in healthy individuals. Our findings are in accordance with previous studies showing significant differences in LES muscle stiffness and contractile characteristics between men and women.^16,25,37,55 Considering the fact that alterations in muscle stiffness and structure are associated with symptoms of low back pain,^14,24 overuse injuries, and performance capacity,⁵⁶ the results appear all the more significant. They may provide a basis for future studies examining potential sex-specific effects of therapeutic interventions in a clinical setting commonly used to improve myofascial tissue regeneration and performance.^2,27

Several limitations of this investigation need to be highlighted. TMG measurements were limited to the LES muscles, and the results obtained cannot be generalized to other lumbar vertebrae segments with differing biomechanical function and morphology.^7,35 Nevertheless, the importance of considering sex as an independent variable in scientific research still stands,³⁶ even if sex-based differences determined in this and other studies point in diverging directions based on the anatomical region or muscle under investigation. Although anthropometry (eg, height and BMI) was taken into account, it is conceivable that lumbar subcutaneous tissue thickness or percentage of body fat and the participant’s level of hydration may have contributed to the monitored differences.^4,13,22 The final limitation relates to the small sample size investigated. Nonetheless, the obtained results appear all the more promising, having reached significance even in such a small sample.

Conclusion

The present study underlines the crucial importance of sex-based differentiation in research design and analysis when determining reference values for biomechanical properties of the musculoskeletal system. The findings indicate that these differences are not exclusively attributable to anthropometric measures (height and BMI) but may stem from intrinsic sex-based differences in skeletal muscle characteristics.