Multiple measures of muscle function influence Sorensen Test performance in individuals with recurrent low back pain

Abstract

BACKGROUND:

Sorensen Test time-to-task-failure (TTF) predicts several low back pain (LBP) clinical outcomes, including recurrence. Because the test is described as a measure of trunk extensor (TE) muscle endurance, LBP rehabilitation programs often emphasize endurance training, but the direct role of TE muscle function on Sorensen Test-TTF remains unclear.

OBJECTIVE:

To assess the discriminative and associative properties of multiple markers of isolated TE performance with regard to Sorensen Test-TTF in individuals with recurrent LBP.

METHOD:

Secondary analysis of baseline measures from participants in a registered (NCT02308189) trial (10 men; 20 women) was performed. Participants were classified by Sorensen Test-TTF as high, moderate or low risk for subsequent LBP episodes, and compared to determine if classification could discriminate differences in TE function. Correlations between Sorensen Test-TTF and isolated TE performance, anthropometrics and disability were investigated.

RESULTS:

Individuals at risk of subsequent LBP episodes had greater perceived disability and fat mass/TE strength ratios ( $P\leqslant$ 0.05) than those not at risk. Modest, significant ( $r=$ 0.36–0.42, $P\leqslant$ 0.05) associations were found between Sorensen Test-TTF, TE endurance and fat mass/TE strength. Exploratory analyses suggested possible sex-specific differences related to Sorensen Test-TTF.

CONCLUSIONS:

Isolated TE muscle endurance is only one of several factors with similar influence on Sorensen Test-TFF, thus LBP rehabilitation strategies should consider other factors, including TE strength, anthropometrics and perceived disability.

Keywords

Fatigue spine body composition magnetic resonance imaging

1. Introduction

Low back pain (LBP) was recently recognized as the number one cause of disability worldwide [1, 2]. Although self-limiting in most cases, LBP is the most common reason for medically-certified sick leave and lost workdays in many countries, with a small fraction of affected individuals experiencing severe, persistent disability [2]. Further complicating this significant health problem is the fact that the vast majority of LBP cannot be attributed to any patho-anatomic cause, and is thus “non-specific,” limiting rational targets for treatment [3]. However, the Sorensen Test has been found to discriminate between those with and without current or previous LBP, and to predict disability 12 months following a subacute episode of LBP [4, 5, 6], suggesting a potential mechanistic link between the Sorensen Test and LBP.

The Sorensen Test [7] is a prone, isometric, position-holding task. Because the test involves maintaining a position until failure (i.e., time-to-task-failure (TTF)), it has been suggested that it is an index of endurance of the trunk extensor (TE) muscles [8, 9, 10]. Thus, the discriminative power of the Sorensen Test is interpreted to indicate that poor TE muscle endurance contributes to the onset, duration, recurrence and associated impairments of LBP [4, 5, 9, 11, 12]. As exercise is one of the most recommended interventions for LBP prevention and rehabilitation [13], categorizing the Sorensen Test as a measure of endurance may influence the design of exercise programs for LBP, which frequently emphasize muscle endurance over strength [11, 14]. Even “core strengthening” or stabilization exercises typically involve high-repetition, sub-maximal activation that is more in line with endurance vs. strength training [15, 16, 17].

However, the demands of the Sorensen Test on an individual may vary widely. For example, for individuals with high trunk mass and weak TE muscles it represents a much more difficult task (greater percentage of maximum capacity) than for those with low trunk mass and strong TE muscles. Indeed, trunk mass has been identified as a negative predictor of Sorensen Test-TTF [18, 19]. Thus, the Sorensen Test might be as much a test of strength as of endurance. Identifying the factors (i.e., strength, fatigability, body mass) contributing to Sorensen Test performance is of clinical interest, as this information could aid in developing interventions aimed at improving test performance. Presumably such programs could also reduce negative sequelae of LBP such as disability, recurrence and, most significantly, the development of chronic LBP. For example, if the Sorensen Test is truly a test of endurance, and this poor endurance contributes to LBP severity, duration, first-time occurrence or recurrence, then strengthening interventions might be of limited use. Supporting this view, one study of exercise in a LBP cohort suggests that high intensity TE muscle training improves strength, but not clinical outcomes, more than low-intensity exercise [20]. On the other hand, a later trial indicates that general aerobic and lumbar-specific exercise programs produce similar improvements in Sorensen Test performance in individuals with chronic LBP [21]. Thus, strength and endurance may both contribute to Sorensen Test performance.

Accordingly, the purpose of this study was to explore the potential contributions of isolated TE strength and endurance to Sorensen Test performance in individuals with a history of LBP via secondary analysis of baseline data from a recently published clinical trial [22]. Because the load on the TE muscles during the test is a function of trunk mass, we also assessed anthropometric factors (e.g., trunk mass, body composition). Finally, psychological variables such as self-efficacy have been found to influence Sorensen Test-TTF [18, 19], so we also measured self-reported disability, which has been shown to be elevated when self-efficacy is low in LBP populations [23, 24]. We used two approaches to evaluate the role of these factors: First, we examined group differences after classifying subjects as being at high, moderate and low risk for subsequent episodes of LBP based on Sorensen Test-TTF to determine if any of the TE muscle or anthropometric parameters could discriminate among groups with clinically-meaningful differences in Sorensen Test-TTF. Second, we directly assessed bivariate correlations between Sorensen Test TTF and these factors to determine the strength of association between them. Although the overall sample was not large, the variety of types of data available were greater than what is available in most studies of Sorensen Test performance in clinical populations. Consistent with an exploratory study, this allowed us to use a broad, if not deep, approach to evaluation. Finally, because of the well-known differences in muscle fatigue between men and women [25], we also conducted exploratory evaluations of sex-associated differences in the data.

2. Method

2.1 Subjects

Thirty-two total participants, ages 18–50 (20 women (W) and 12 men (M)), were recruited for a registered randomized controlled trial (RCT; NCT02308189 [26]) examining the effectiveness of blood-flow restricted exercise for improving trunk muscle function in individuals with LBP. Patients were recruited over a two-year, rolling enrollment period, via methods detailed in the published study protocol [26]. Two participants (both M) were removed from the study after randomization (For CONSORT diagram, see [22]) giving a final study $n=$ 30. All participants provided written informed consent for all experimental procedures and the study protocol was approved by the Institutional Review Board of Ohio University. For the present study, a secondary analysis of baseline data collected in this trial was conducted.

Participant eligibility is detailed in the published study protocol [26]. Briefly, all participants had recurrent LBP, defined as at least two episodes of LBP in the past year, at least one of which caused restriction of work or leisure activity and were sedentary, per the International Physical Activity Questionnaire [27]. To minimize the effect of pain on testing procedures, subjects were excluded for a current report of $\geqslant$ 4 on the Numeric Pain Rating Scale for LBP (NPRS) [28]. All participants had Sorenson Test TTF $<$ 176 s. This TTF is at or below a value shown to discriminate individuals with recurrent LBP [7]. This choice as made with the goal of identifying individuals with impaired muscle function, as the purpose of the parent RCT was evaluate an intervention to improve trunk muscle performance in recurrent LBP. Data presented here were collected prior to treatment group assignment. The published results of the trial are related to pre- vs. post-intervention changes in muscle performance and size variables. Thus this study represents a new analysis that does not overlap with previously published results.

2.2 Data collection

Subjects participated in two separate sessions, separated by $\geqslant$ 3 days. On Day 1, subjects completed informed consent, all screening procedures related to inclusion/exclusion [26], basic medical screening (including height and weight) and finally the Sorensen Test as previously described [29]. Sorensen Test-TTF was the time the subject was able to maintain their upper body parallel to the ground despite verbal encouragement to maintain the position. On Day 2, subjects underwent muscle size and body composition testing (magnetic resonance imaging (MRI) and Dual x-ray absorptiometry (DXA)), followed by pain and disability ratings, then muscle testing. Knee extensor (KE) muscle testing was completed first, followed by TE testing after 20–25 minutes. Tests were always conducted in the same order.

2.3 Anthropometrics and body composition

Whole body and regional lean and fat mass were assessed by DXA per manufacturer (Hologic Discovery QDR model Series, Waltham, MA, USA) guidelines. For purposes of this study, we examined both total and upper body (UB) fat and lean mass. Total mass was calculated by adding fat and lean mass. Muscle cross-sectional area (CSA) was determined by MRI for the TE and KE. Transverse T-1 weighted spin echo images were acquired on a 0.25 Tesla Esaote G-Scan Brio System (Genoa, Italy) as 10 mm slices between L2 and L5. Separate CSA was calculated for the quadratus lumborum, iliocostalis lumborum/longissimus thoracis and multifidus muscles from an average of three slices obtained from the center of the respective muscles, and then added for total TE CSA. Similar methods were used to determine KE cross-sectional area from MRI of the quadriceps femoris muscle group. For both muscle groups, intramuscular fat (IMF) was measured and subtracted from to give muscle CSA [30].

2.4 Pain and disability

Participants completed a series of self-report instruments on pain (a Numeric Pain Rating Scale (NPRS) of 0–10), disability (24-item Roland-Morris Disability Questionnaire (RMDQ)) and movement-related fear (Tampa Scale for Kinesiophobia (TSK)).

2.5 Muscle testing

Maximum isometric torque of the KE and TE muscle groups was determined by subjects performing maximum voluntary contractions (MVCs) of the TE and KE (bilateral and unilateral) muscles on MedX Dynamometers (Ocala, FL, USA), designed for the back and knee, respectively. The extensive restraint system of the MedX back extension makes it well suited to isolating the TE muscles and has been found to be reliable for both strength and endurance testing [31]. Analysis of the KE data allowed us to evaluate a potential generalized, systemic impairment of muscle function in the participants. The MedX systems were custom-instrumented with load cells and interfaced with BioPac software, allowing collection of raw torque signal at 500 Hz. Muscle quality for the TE group was calculated by dividing the TE MVC by the fat-corrected TE CSA, as described above. In addition, we use regional lean, total and fat mass data from the DXA measurements to determine the ratios of regional mass to TE strength (MVC).

Table 1
Participant characteristics

Measure	Mean (SD)		Mean (SD)		Mean (SD)
	Total ( $n=$ 30)		W ( $n=$ 20)		M ( $n=$ 10)
Weight (kg)	73.0	(15.5)	67.3	(13.7)	84.4	(12.6) ${}^{*}$
Height (m)	1.69	(0.09)	1.64	(0.07)	1.78	(0.05) ${}^{*}$
Age (yrs) ${}^{U}$	29.5	(9.5)	27.7	(10.3)	33.3	(6.7) ${}^{*}$
BMI	25.4	(4.6)	24.8	(4.5)	26.6	(4.7)
TSK	32.5	(5.6)	32.6	(5.9)	32.4	(5.3)
RMDQ	4.8	(2.7)	5.1	(2.7)	4.1	(2.6)
NPRS ${}^{U}$	2.8	(1.1)	2.8	(1.1)	2.8	(1.1)
Sorensen Test TTF (s)	82.4	(34.9)	87.1	(37.1)	72.9	(28.6)
TE TTF (s) ${}^{U}$	329.9	(191.7)	320.6	(212.2)	347.8	(154.2)
TE MVC (N $\cdot$ m)	158.7	(63.8)	130.9	(46.8)	214.4	(57.7) ${}^{*}$
TE CSA (cm ${}^{2}$ )	20.65	(3.60)	19.40	(3.11)	23.28	(3.23) ${}^{*}$
IMF (mm ${}^{2}$ )	147.4	(118.1)	142.6	(94.5)	157.7	(163.9)
TE muscle quality (N $\cdot$ m/cm ${}^{2}$ )	7.6	(2.8)	6.5	(2.1)	9.9	(2.8)
Total UB mass (Kg)	40.43	(10.04)	38.01	(9.41)	45.03	(10.04)
UB fat mass (Kg) ${}^{U}$	13.58	(6.20)	13.45	(6.56)	13.83	(5.78)
UB % fat ${}^{U}$	30.9	(8.9)	32.2	(9.1)	28.4	(8.5)
Total UB mass: TE MVC (Kg/N $\cdot$ m) ${}^{U}$	0.225	(0.084)	0.248	(0.078)	0.184	(0.086) ${}^{*}$
UB fat mass: TE MVC (Kg/N $\cdot$ m)	0.093	(0.050)	0.104	(0.050)	0.071	(0.045)
KE MVC (N $\cdot$ m)	666.9	(202.4)	592.8	(126.5)	815.1	(248.3) ${}^{*}$
KE TTF (s) ${}^{U}$	183.4	(108.8)	185.2	(125.5)	179.7	(69.4)

${U}$ $=$ W vs M comparison by Mann-Whitney Test; $*$ $=$ Significant ( $P\leqslant$ 0.05) difference W vs. M. BMI: body mass index, CSA: cross-sectional area, IMF: intramuscular fat, KE: knee extensor muscle, MVC: maximum voluntary contraction, NPRS: Numeric Pain Rating Scale, RMDQ: Roland-Morris Disability Questionnaire, TE: trunk extensor muscle, TSK: Tampa Scale for Kinesiophobia, TTF: time-to-task-failure, UB: upper body.

The same MedX and BioPac equipment was used to determine isolated muscle TTF for the trunk (TE-TTF) and knee extensors (KE-TTF). The fatigue task involved maintaining a sustained isometric contraction at a target of 20% MVC, a common submaximal fatigue task [25, 32, 33]. As with testing MVC of the TE, the TE-TTF was tested in sitting with the trunk midway between the full flexion and extension range of motion available on the MedX [34]. For KE-TTF, subjects were seated with the knee flexed to 60 ${}^{\circ}$ . In both instances, subjects pushed into the appropriate arm of the MedX system while receiving visual feedback regarding their torque output relative to the target (20% MVC) on a computer monitor as well as verbal encouragement maintain the position torque signal. Participants maintained the torque until they could no longer continue (i.e., voluntarily ended the test). The force signal was later analyzed offline and TTF for each muscle group was determined as the first time at which the torque signal fell below the target force line by 5% of target force (1% MVC) for $>$ 3 seconds or when the participant stopped the test, whichever occurred first.

2.6 Statistical analysis

Descriptive statistics for subject characteristics are presented as means and standard deviations, unless noted otherwise. Comparisons between women (W) and men (M) were made with unpaired $t$ -tests where Shapiro-Wilk testing confirmed normality and by Mann-Whitney tests where it did not. Subjects were categorized post-hoc as being at high, moderate or low risk of a subsequent episode of LBP according to published sex-specific cut points of Sorensen Test-TTF (104 $>$ s (M), $>$ 110 s (W); 58–104 s (M), 58–110 (W); $<$ 58 s, respectively [5]). Differences among these groups were evaluated by one-way ANOVA. In the event of a significant main effect of group, post-hoc testing was performed by the LSD test. Bivariate associations were assessed via Pearson correlations where Shapiro-Wilk testing confirmed normality or by Spearman Rank correlations where it did not. Factors identified as having significant discriminative (i.e., ANOVA) or associative (i.e., correlation) effects on Sorensen Test-TTF were used in stepwise linear regression modeling for Sorensen Test performance.

3. Results

Participant characteristics and sex differences are presented in Table 1. Generally, the only differences were those related to body and muscle size and muscle strength, which are typically different between W and M.

Sorting participants by Sorensen Test-TTF into groups with high, moderate and low risk for subsequent episodes of LBP resulted in group sizes of 9, 11 and 10, respectively. Analysis of the risk groups by ANOVA revealed trends ( $P=$ 0.087) for an effect of group on RMDQ and UB fat mass (Fig. 1A and C, respectively) and a significant effect of group on UB fat mass: TE MVC and total UB mass (Fig. 1B and D, respectively).

Figure 1.

A) RMDQ, B) UB fat mass: TE MVC, C) UB fat mass and D) UB mass (all means $+$ SE) for LBP risk groups based on Sorensen Test TTF. Dashed bar $=$ trend for an effect of group; $P=$ (0.05 $<p<$ 0.09). Solid bar $=$ main effect of group; $P<$ 0.05. * $=$ group value significantly greater than the other two groups.

Associations between Sorensen Test-TTF and the muscle, anthropometric, and pain and disability measures are presented in Table 2. Consistent with the description of the Sorensen Test as a measure of TE endurance, TE-TTF was showed a trend for association with Sorensen-TTF, but TE MVC and TE CSA did not. A somewhat more robust, and significant, association between the ratio of upper body fat mass to TE MVC (UB fat mass: TE MVC) and Sorensen Test-TTF was found. Sorensen Test-TTF was also significantly negatively associated with RMDQ, indicating that low Sorensen Test-TTF was associated with greater report of disability. Associations with Sorensen Test-TTF were also somewhat different when W and M were considered separately. Specifically, the ratios of UB total and fat mass to TE MVC were more strongly (negatively) associated with Sorensen Test-TTF in women. In contrast, the negative association between Sorensen Test-TTF and disability was stronger in men, as was the association with KE-TTF. In addition, the association between TSK and Sorensen Test-TTF was qualitatively quite different (i.e., negative in M, but not W), though neither was statistically significant.

Table 2

Associations with Sorensen Test-TTF

	Group
Associated variable	Total	W only	M only
	$r$ -value	$r$ -value	$r$ -value
TE-TTF ${}^{S}$	0.320 ${}^{\dagger}$	0.440 ${}^{{\dagger}}$	0.309
TE MVC	0.022	0.292	$-$ 0.028
Height	$-$ 0.062	0.131	0.079
Weight	$-$ 0.289	$-$ 0.202	$-$ 0.291
BMI	$-$ 0.292	$-$ 0.255	$-$ 0.299
Age ${}^{S}$	$-$ 0.169	$-$ 0.051	$-$ 0.139
TE CSA	$-$ 0.156	$-$ 0.026	$-$ 0.270
IMF	0.281	0.144	0.619 ${}^{{\dagger}}$
TE muscle quality	0.113	0.412	$-$ 0.070
UB mass	$-$ 0.363 ${}^{*}$	$-$ 0.299	$-$ 0.355
UB fat mass ${}^{S}$	$-$ 0.234	$-$ 0.230	$-$ 0.103
Trunk % fat ${}^{S}$	$-$ 0.281	$-$ 0.415 ${}^{{\dagger}}$	0.091
UB mass: TE MVC ${}^{S}$	$-$ 0.181	$-$ 0.433 ${}^{{\dagger}}$	$-$ 0.006
UB fat mass: TE MVC	$-$ 0.365 ${}^{*}$	$-$ 0.577 ${}^{**}$	$-$ 0.141
RMDQ	$-$ 0.417 $*$	$-$ 0.366	$-$ 0.777 ${}^{**}$
TSK	$-$ 0.025	0.069	$-$ 0.324
NPRS ${}^{S}$	$-$ 0.206	$-$ 0.109	$-$ 0.453
KE MVC	$-$ 0.083	$-$ 0.134	0.252
KE-TTF ${}^{S}$	0.157	$-$ 0.026	0.736 ${}^{*}$

$S$ $=$ coefficient from Spearman Rank Correlation. $*$ $=$ Significant ( $P\leqslant$ 0.05) association; ${**}$ $=$ Significant ( $P<$ 0.01) association; ${{\dagger}}$ $=$ Trend (0.05 $<P<$ 0.10) for association; BMI: body mass index, CSA: cross-sectional area, IMF: intramuscular fat, KE: knee extensor muscle, MVC: maximum voluntary contraction, NPRS: Numeric Pain Rating Scale, RMDQ: Roland-Morris Disability Questionnaire, TE: trunk extensor muscle, TSK: Tampa Scale for Kinesiophobia, TTF: time-to-task-failure, UB: upper body.

While bivariate correlations did identify significant associations between several parameters and Sorensen Test-TTF, the strength of these associations was only moderate. To determine if a combination of factors could show a more robust relationship, we used stepwise multiple linear regression modeling. Five factors were tested in the model; those found to exhibit at least a trend for an association (Table 2) with Sorensen Test-TTF and sex. All of the factors were retained in the model ( $r^{2}=$ 0.568, $P<$ 0.001). However, collinearity diagnostics suggested potential problems when the UB total mass variable was entered. As it was the last variable entered, we chose to omit UB total mass and limit the model to the other four factors. The final weighted regression model gave an adjusted $r^{2}$ of 0.465 (Table 3).

Table 3

Regression for factors related to Sorensen Test TTF

	Factor	Unstandardized beta	SE	Standardized beta	$t$	Adj. $r^{2}$	$p$ -value
Model						0.465	0.001
	(Constant)	95.74	15.87	–	6.03		$<$ 0.001
	RMDQ	$-$ 5.82	1.88	$-$ 0.45	$-$ 3.10		0.005
	Sex	32.71	10.47	0.462	3.12		0.005
	UB Fat: TE MVC	$-$ 0.32	0.13	$-$ 0.36	$-$ 2.40		0.025
	TE TTF	0.06	0.03	0.32	2.22		0.036

Weighted for sex. Abbreviations are the same as for Table 1.

4. Discussion

Consistent with the description of the Sorensen Test as a test of trunk muscle endurance, isolated TE-TTF, but not isolated TE MVC, exhibited a trend for association with Sorensen Test-TTF [8]. Others have also reported associations between TE endurance, but not strength, and Sorensen Test-TTF in healthy individuals, though these involved isokinetic TE testing [35, 36]. The present study builds upon these earlier works by evaluating a LBP population and by addressing body composition and anthropometrics. Indeed, while isolated TE MVC was not associated with Sorensen Test-TTF, a significant association was present for the UB fat mass: TE MVC ratio. This finding is consistent with our recent studies reporting that trunk mass is a significant contributor to Sorensen Test-TTF in individuals with recurrent LBP [18, 19]. Moreover, the strength of association between Sorensen Test-TTF and UB fat mass: TE MVC was greater than that seen for TE-TTF, and achieved the threshold of statistical significance. These findings indicate that the Sorensen Test is at least as much of a test of TE strength (or relative strength) as one of endurance. Thus, TE exercise programs for prevention or rehabilitation of LBP should ideally target both muscle endurance and strength and possibly be weight. We base this on the observations that ratio of TE strength to upper body fat mass and TE-TTF were independent contributors to Sorensen Test-TTF (Table 3) and were not correlated with each other. These associative findings were confirmed when testing for differences among groups of individuals classified by Sorensen-TTF as being at different levels of risk for subsequent episodes of LBP (Fig. 1). Again, the effect of UB fat mass: TE MVC ratio, but not TE-TTF, was significant. In addition, several other measures of upper body mass and relative strength also discriminated among the groups. This could indicate that, while isolated trunk endurance contributes to Sorensen Test-TTF, TE strength relative to upper body composition and mass is a more important factor in the risk for LBP. Further studies designed to specifically test this hypothesis are warranted, as they could direct the focus of interventions aimed at rehabilitation and prevention of LBP. In particular, larger studies targeting the muscle performance parameters related to Sorensen Test-TTF are needed to determine if improvements in these factors can reduce disability and improve clinical outcomes in different LBP populations.

The RMDQ has been found to be an important predictor of long term outcomes, including return to work, in individuals with LBP [37, 38]. We observed a significant negative association between Sorensen Test-TTF and self-reported disability per the RMDQ, consistent with that reported for other measures of disability [12]. Although our study group did not have high ratings of disability or current pain, there was also significant main effect of Sorensen Test-TTF classification on RMDQ score. Moreover, the difference between the high- and low-risk groups was within the range of minimal clinically important differences for RMDQ [39, 40]. We have recently found that self-efficacy contributes significantly to Sorensen Test-TFF [18, 19]. Our finding that RMDQ predicted Sorensen Test TTF is consistent with these findings, as it has been reported that greater disability is associated with both pain- and exercise-related self-efficacy [23, 24]. Interestingly, Sorensen Test-TTF was significantly associated with RMDQ, neither TE-TTF nor strength were. It might be that Sorensen Test-TTF is driven by perceived disability, directly or mediated via self-efficacy, rather than low Sorensen Test-TTF reflecting physical impairments that contribute to perceived disability. Nevertheless, the present findings do support the reported discriminative and prognostic properties of the Sorensen Test related to LBP [4, 5, 6].

Finally, because W frequently exhibit less muscle fatigue than M (for review, c.f., [25]), including during the Sorensen Test [41], we explored the data for sex-specific differences, though the study was not specifically designed for this. No significant differences between W and M for either Sorensen Test- or TE-TTF were observed, though slightly greater ( $\sim$ 15%), non-significant Sorensen Test-TTF in W is similar to what has been reported elsewhere [18]. No group X sex interactions were observed in the comparisons among the risk of recurrence groups. However, associative properties of Sorensen Test-TTF did vary between W and M for several factors (Table 2), despite similar Sorensen Test-TTFs. Generally, anthropometrics and body composition (e.g. UB fat mass and % fat) were more strongly associated with Sorensen Test TTF in W. In contrast, stronger associations for Sorensen Test-TTF with RMDQ and TSK were found in M, though the association with TSK was not significant. An unexpectedly strong association between KE-TTF and Sorensen Test-TTF was observed in M. This latter finding could indicate that M in the study experienced a general deconditioning with their LBP, whereas the W experienced a trunk-specific loss of muscle function.

These sex-related differences should be viewed cautiously, as this study was not specifically powered to assess sex-related differences and the sample contained twice as many W as M. These data support no definitive conclusions regarding sex-specific mechanisms contributing to Sorensen Test-TTF performance. However, the observed differences are intriguing enough to warrant further specific investigation. If confirmed, such differences could suggest that interventions for LBP may need to vary depending on sex of the patient. For example, a rehabilitation program involving a general increase in activity might be sufficient for men, whereas programs for women might benefit from the inclusion of TE-specific exercises.

In addition to the limitations related to sex-specific analyses, other limitations of the study must be acknowledged. First, this was a secondary analysis of an RCT. Sample sizes were not optimized for the specific analyses of this study; thus results should be viewed as preliminary. The patient population was narrowly defined and may not reflect other LBP groups (e.g., chronic or acute LBP), with differing levels of pain or disability. Methodologically, the similar relationship between MVC and CSA in the KE muscle group ( $r=$ 0.587, $P=$ 0.001, data not shown) and MVC and CSA in the TE muscles ( $r=$ 0.556, $P=$ 0.002, data not shown), suggests that the MedX testing effectively tested isolated the TE muscles. However, hip extensor muscle is likely to contribute to Sorensen Test-TTF [29, 42] and we did not assess isolated hip muscle parameters. As our regression model explained $\sim$ 56% of the variance in Sorensen Test-TTF, inclusion of these factors might improve its explanatory power. Finally, task type differed across the Sorensen and isolated muscle tests. The isolated KE- and TE-TTF involved force-matching, while the Sorensen Test is a position-matching task. Fatigue differs in position- and force-matching tasks within a muscle group [32, 33], and we have previously reported that the TE muscles exhibit a different pattern of task-specificity in this regard than appendicular muscles [43]. Future studies examining isolated TE-TTF during position-matching may find a more robust relationship between TE- and Sorensen Test-TTF.

5. Conclusions

The expected association between Sorensen Test-TTF and isolated TE-TTF was observed, but other factors exhibited comparable associations with Sorensen Test-TTF, suggesting the test represents a more global assessment of TE muscle performance than a simple index of endurance. Consistent with this view, a regression model incorporating TE-TTF, TE strength relative to fat mass and RMDQ score explained 56% of the variance in Sorensen Test-TTF. Intriguing differences in M and W that could potentially direct sex-specific intervention strategies were suggested by exploratory analyses.

Footnotes

Acknowledgments

The authors thank Rachel Clift for providing study coordination and management and Kwasi Ampomah for assistance with data processing. Portions of these data have been presented in abstract form at the 2019 Combined Sections Meeting of the American Physical Therapy Association. This work was supported, in part, by a grant from the National Institutes of Health (NIAMS R21AR063909 to Brian Clark).

Conflict of interest

Brian Clark has received research funding from the National Institutes of Health, Regeneron Pharmaceuticals, Astellas Pharma Global Development, Inc., RTI Health Solutions, Ohio Department of Higher Education, and the Osteopathic Heritage Foundations. In the past 5 years Brian Clark has received consulting fees from Regeneron Pharmaceuticals, Abbott Laboratories, and the Gerson Lehrman Group. Additionally, Brian Clark is co-founder with equity and scientific director of AEIOU Scientific, LLC. In the past five years, David Russ has received research funding and consulting fees from Abbott Nutrition. The other authors declare no conflicts of interest.

References

DALYs

GBD

Collaborators

. Global, regional, and national disability-adjusted life-years (DALYs) for 315 diseases and injuries and healthy life expectancy (HALE), 1990-2015: a systematic analysis for the Global Burden of Disease Study 2015. Lancet. 2016; 388(10053): 1603-58.

Hartvigsen

Hancock

Kongsted

Louw

Ferreira

Genevay

, et al. What low back pain is and why we need to pay attention. Lancet. 2018; 391(10137): 2356-67.

Maher

Underwood

Buchbinder

. Non-specific low back pain. Lancet. 2017; 389(10070): 736-47.

Rossi

Morcelli

Cardozo

Denadai

Goncalves

Navega

. Discriminant analysis of neuromuscular variables in chronic low back pain. Journal of Back and Musculoskeletal Rehabilitation. 2015; 28(2): 239-46.

Alaranta

Luoto

Heliovaara

Hurri

. Static back endurance and the risk of low-back pain. Clin Biomech (Bristol, Avon). 1995; 10(6): 323-4.

Enthoven

Skargren

Kjellman

Oberg

. Course of back pain in primary care: a prospective study of physical measures. J Rehabil Med. 2003; 35(4): 168-73.

Biering-Sorensen

. Physical measurements as risk indicators for low-back trouble over a one-year period. Spine (Phila Pa 1976). 1984; 9(2): 106-19.

Conway

Behennah

Fisher

Osborne

Steele

. Associations between trunk extension endurance and isolated lumbar extension strength in both asymptomatic participants and those with chronic low back pain. Healthcare. 2016; 4(3).

Demoulin

Boyer

Duchateau

Grosdent

Jidovtseff

Crielaard

, et al. Is the Sorensen test valid to assess muscle fatigue of the trunk extensor muscles? Journal of Back and Musculoskeletal Rehabilitation. 2016; 29(1): 31-40.

10.

Vining

Minkalis

Long

Corber

Franklin

Gudavalli

, et al. Assessment of chiropractic care on strength, balance, and endurance in active-duty U.S. military personnel with low back pain: a protocol for a randomized controlled trial. Trials. 2018; 19(1): 671.

11.

McGill

. Low back exercises: evidence for improving exercise regimens. Phys Ther. 1998; 78(7): 754-65.

12.

Handrakis

Friel

Hoeffner

Akinkunle

Genova

Isakov

, et al. Key characteristics of low back pain and disability in college-aged adults: a pilot study. Arch Phys Med Rehabil. 2012; 93(7): 1217-24.

13.

Foster

Anema

Cherkin

Chou

Cohen

Gross

, et al. Prevention and treatment of low back pain: evidence, challenges, and promising directions. Lancet. 2018; 391(10137): 2368-83.

14.

Welch

Moran

Antony

Richter

Marshall

Coyle

, et al. The effects of a free-weight-based resistance training intervention on pain, squat biomechanics and MRI-defined lumbar fat infiltration and functional cross-sectional area in those with chronic low back. BMJ Open Sport & Exercise Medicine. 2015; 1(1): e000050.

15.

Sung

. Disability and back muscle fatigability changes following two therapeutic exercise interventions in participants with recurrent low back pain. Medical Science Monitor: International Medical Journal of Experimental and Clinical Research. 2013; 19: 40-8.

16.

Macedo

Latimer

Maher

Hodges

McAuley

Nicholas

, et al. Effect of motor control exercises versus graded activity in patients with chronic nonspecific low back pain: a randomized controlled trial. Phys Ther. 2012; 92(3): 363-77.

17.

Hides

Stanton

McMahon

Sims

Richardson

. Effect of stabilization training on multifidus muscle cross-sectional area among young elite cricketers with low back pain. J Orthop Sports Phys Ther. 2008; 38(3): 101-8.

18.

Applegate

France

Russ

Leitkam

Thomas

. Determining physiological and psychological predictors of time to task failure on a virtual reality sorensen test in participants with and without recurrent low back pain: exploratory study. JMIR Serious Games. 2018; 6(3): e10522.

19.

Applegate

France

Russ

Leitkam

Thomas

. Sorensen test performance is driven by different physiological and psychological variables in participants with and without recurrent low back pain. J Electromyogr Kinesiol. 2019; 44: 1-7.

20.

Helmhout

Harts

Staal

Candel

de Bie

. Comparison of a high-intensity and a low-intensity lumbar extensor training program as minimal intervention treatment in low back pain: a randomized trial. Eur Spine J. 2004; 13(6): 537-47.

21.

Shnayderman

Katz-Leurer

. An aerobic walking programme versus muscle strengthening programme for chronic low back pain: a randomized controlled trial. Clin Rehabil. 2013; 27(3): 207-14.

22.

Ampomah

Amano

Wages

Volz

Clift

Ludin

AFM

, et al. Blood flow-restricted exercise does not induce a cross-transfer of effect: a randomized controlled trial. Med Sci Sports Exerc. 2019; 51(9): 1817-27.

23.

La Touche

Grande-Alonso

Arnes-Prieto

Paris-Alemany

. How does self-efficacy influence pain perception, postural stability and range of motion in individuals with chronic low back pain? Pain Physician. 2019; 22(1): E1-E13.

24.

Mannion

O’Riordan

Dvorak

Masharawi

. The relationship between psychological factors and performance on the Biering-Sorensen back muscle endurance test. Spine J. 2011; 11(9): 849-57.

25.

Russ

. Sex differences in muscle fatigue. In: Williams

Ratel

, editor. Human Muscle Fatigue. New York: Routledge; 2009. pp. 135-63.

26.

Amano

Ludin

Clift

Nakazawa

Law

Rush

, et al. Effectiveness of blood flow restricted exercise compared with standard exercise in patients with recurrent low back pain: study protocol for a randomized controlled trial. Trials. 2016; 17: 81.

27.

Craig

Marshall

Sjostrom

Bauman

Booth

Ainsworth

, et al. International physical activity questionnaire: 12-country reliability and validity. Med Sci Sports Exerc. 2003; 35(8): 1381-95.

28.

Farrar

Young

, Jr. LaMoreaux

Werth

Poole

. Clinical importance of changes in chronic pain intensity measured on an 11-point; numerical pain rating scale. Pain. 2001; 94(2): 149-58.

29.

Champagne

Descarreaux

Lafond

. Back and hip extensor muscles fatigue in healthy subjects: task-dependency effect of two variants of the Sorensen test. Eur Spine J. 2008; 17(12): 1721-6.

30.

Manini

Clark

Nalls

Goodpaster

Ploutz-Snyder

Harris

. Reduced physical activity increases intermuscular adipose tissue in healthy young adults. Am J Clin Nutr. 2007; 85(2): 377-84.

31.

Udermann

Mayer

Graves

Murray

. Quantitative assessment of lumbar paraspinal muscle endurance. J Athl Train. 2003; 38(3): 259-62.

32.

Yoon

Hawe

Hunter

. Variation in limb support influences the time to task failure for a postural contraction. J Mot Behav. 2009; 41(5): 393-5.

33.

Hunter

Yoon

Farinella

Griffith

. Time to task failure and muscle activation vary with load type for a submaximal fatiguing contraction with the lower leg. Journal of applied physiology. 2008; 105(2): 463-72.

34.

Mayer

Childs

Neilson

Chen

Koppenhaver

Quillen

. Effect of lumbar progressive resistance exercise on lumbar muscular strength and core muscular endurance in soldiers. Mil Med. 2016; 181(11): e1615-e22.

35.

Juan-Recio

Lopez-Plaza

Murillo

Garcia-Vaquero

Vera-Garcia

. Reliability assessment and correlation analysis of 3 protocols to measure trunk muscle strength and endurance. Journal of Sports Sciences. 2018; 36(4): 357-64.

36.

Mayer

Gatchel

Betancur

Bovasso

. Trunk muscle endurance measurement. Isometric contrasted to isokinetic testing in normal subjects. Spine (Phila Pa 1976). 1995; 20(8): 920-6; discussion 6-7.

37.

Helmhout

Staal

Heymans

Harts

Hendriks

, de Bie

. Prognostic factors for perceived recovery or functional improvement in non-specific low back pain: secondary analyses of three randomized clinical trials. Eur Spine J. 2010; 19(4): 650-9.

38.

Chen

Hogg-Johnson

Barbado Smith

. The recovery patterns of back pain among workers with compensated occupational back injuries. Occup Environ Med. 2007; 64(8): 534-40.

39.

Roland

Fairbank

. The roland-morris disability questionnaire and the oswestry disability questionnaire. Spine (Phila Pa 1976). 2000; 25(24): 3115-24.

40.

Team

UBT

. United Kingdom back pain exercise and manipulation (UK BEAM) randomised trial: cost effectiveness of physical treatments for back pain in primary care. BMJ. 2004; 329(7479): 1381.

41.

Clark

Manini

The

Doldo

Ploutz-Snyder

. Gender differences in skeletal muscle fatigability are related to contraction type and EMG spectral compression. Journal of Applied Physiology. 2003; 94(6): 2263-72.

42.

Plamondon

Trimble

Lariviere

Desjardins

. Back muscle fatigue during intermittent prone back extension exercise. Scand J Med Sci Sports. 2004; 14(4): 221-30.

43.

Russ

Ross

Clark

Thomas

. The effects of task type on time to task failure during fatigue: a modified sorensen test. J Mot Behav. 2018; 50(1): 96-103.

Multiple measures of muscle function influence Sorensen Test performance in individuals with recurrent low back pain

Abstract

BACKGROUND:

OBJECTIVE:

METHOD:

RESULTS:

CONCLUSIONS:

Keywords

1. Introduction

2. Method

2.1 Subjects

2.2 Data collection

2.3 Anthropometrics and body composition

2.4 Pain and disability

2.5 Muscle testing

Table 1 Participant characteristics

3. Results

5. Conclusions

Footnotes

Acknowledgments

Conflict of interest

References

Table 1
Participant characteristics