A sample of 24 individuals (12/24, 50% male) with a history of recurrent LBP (LBP) and 24 individuals (12/24, 50% male) with no history of LBP (Healthy) matched for age, sex, and body mass index (BMI) were recruited from the Ohio University student population and surrounding community for this comparative study. Table 1 summarizes the participants’ characteristics. Individuals with a history of hip arthroscopy or spine surgery, known neurological, visual, or orthopedic impairments, depression, ongoing drug or alcohol problems, elevated resting blood pressure (>135/>90 mmHG), or BMI of >35 were excluded from the study. We defined LBP history as having experienced more than one episode of LBP with symptoms occurring in the past 6 months and a previous consultation regarding their LBP symptoms with a health care provider; participants reporting moderate to severe pain (numerical pain rating scale of >3) within the past 6 weeks or those who did not meet the classification of category 1 (LBP that does not radiate) through category 3 (LBP that radiates beyond the knee but without neurological signs) on the Classification System of the Quebec Task Force on Spinal Disorders were excluded from participation. The protocol was approved by the Ohio University Institutional Review Board for human subjects research, and all individuals provided written consent prior to participation.

Participants completed 2 separate testing sessions. Participants in this study were first included in an assessment of performance on the classic Sørensen test [34] and were invited to participate in the virtual reality Sørensen test 3-14 days later. During the first testing session, participants completed the psychological surveys, maximal strength assessments, and the classic Sørensen test. During the second testing session, participants completed the virtual reality Sørensen test.

The maximal voluntary contraction procedure was completed as described previously [34]. As illustrated in Figure 1, subjects were situated on the custom fatigue table with the anterior superior iliac spine aligned with the edge of the table and the torso supported by the platform positioned such that the trunk center of mass was centered over the 6 DOF load cell. In addition, the torso was secured to the platform, the pelvis was secured to the table, and the lower legs were secured at 33% of hip height by a padded bar connected to the single DOF load cell. Bracing of the feet was inhibited with a foam roll placed below the ankles. The trunk mass was measured in this position. EMG was measured as previously described and the custom LABVIEW (Version 13; National Instruments) program collected EMG and load cell measurements.

For the trunk extension trials, participants were instructed to pull their torso up into the back restraint. Three submaximal trunk extension attempts of increasing intensity were followed by 3 maximal trunk extension attempts. Then, participants were instructed to extend their legs up against the stationary leg restraint; 3 submaximal hip extension attempts of increasing intensity were followed by 3 maximal attempts. We provided 2 minutes of rest between each attempt. Verbal encouragement and visual and audio feedback were provided via the custom LABVIEW (Version 13; National Instruments) program.

The participants performed the virtual reality Sørensen test on a standard table with belts across the pelvis and calves at 33% of hip height, the anterior superior iliac spine aligned with the edge of the table, and the upper body unsupported, as seen in Figure 2. Subjects wore an Oculus Rift head mounted display (Oculus Rift Developers Kit 2), as shown in Figure 2. During the test, the Oculus Rift displayed a sky environment in which the participant attempted to “fly” through hoops, as seen in Figure 3. Extending and flexing the trunk appeared to make the subject fly higher and lower, respectively. The hoops were positioned such that the participants were encouraged to maintain a horizontal position for as long as possible. Immediately following a brief practice attempt of <5 seconds, participants completed the self-efficacy questionnaire.

Participants then attempted to maintain the task position until failure while receiving audio and visual feedback through the virtual reality video game; a tone played when the participant’s position was >2° beyond the target position in either direction, which was visually represented by flying above or below the hoops. The trial was terminated when participants fell out of the range (±2°) for >3 consecutive seconds.

In this study, independent-sample t tests were used to evaluate differences in participant demographics. A two-way analysis of variance was used to determine group and sex differences in TTF on the virtual reality Sørensen test. In addition, stepwise linear regression analyses were performed to determine which physiological factors were related to TTF on the virtual reality Sørensen test in each group. The second set of linear regression analyses was performed with the significant physiological factors entered into the first block and psychological factors entered stepwise in the second block to determine which psychological factors were related to TTF on the virtual reality Sørensen test in each group. All analyses were performed in SPSS (IBM Corp.), and results are reported as mean (SE) unless otherwise stated.

The independent-sample t test revealed no significant differences between Healthy and LBP groups with one exception; the depression scores were significantly higher in the LBP group than in those the healthy group (Table 2).

A 2 Group (Healthy, LBP) x 2 Sex (Male, Female) two-way analysis of variance revealed no significant differences in group (F_1,44=0.00, P=.99) or group by sex (F_1,44=0.33, P=.57; Table 3); however, there was a marginal effect of sex (F_1,44=3.89, P=.06), which reflected a tendency toward the longer TTF in female versus male participants.

Simple correlations were run between TTF and each of the physiological and psychological factors that were entered into the linear regression analyses. Table 4 displays the Pearson correlations.

This study aimed to examine the performance on a variation of the Sørensen test using a virtual reality video game in individuals with and without a history of recurrent LBP. To the best of our knowledge, this is the first study to use a virtual reality video game in conjunction with the Sørensen test. Contrary to much of the published literature, we did not find a significant difference in TTF between the groups. In the first longitudinal study, males with a short TTF on the Sørensen test were most likely to experience LBP in the following year, identifying the test as a predictor of first-time LBP [36]. The test was later recognized as a discriminator of those with and without LBP; individuals who had no prior LBP experience exhibited a markedly longer TTF than those with any LBP experience [37]. Many studies have reported consistent findings; however, others have failed to find a difference in performance between those with and without LBP. Many physiological factors, including the BMI, trunk mass, MPF slopes of the trunk and hip extensors, and maximal trunk and hip strength, have been found to influence the performance on the task; these factors have been discussed previously [34].

Although our sample of individuals with recurrent LBP performed just as well on the virtual reality Sørensen test as those without LBP, several interesting findings regarding the factors associated with TTF emerged. In the healthy group, it appears that TTF was driven primarily by trunk mass and the MPF slopes of the trunk and hip extensors, and this is consistent with our previous work [34]. In addition, trunk mass was a predictor of TTF in the LBP group. Other studies also demonstrated the marked effects of anthropometrics on performance. The workload of the task is governed by the weight of the body above the hips. It is obvious that an individual with a heavier trunk mass will not be able to maintain the test position for as long as another individual with the same strength capacity but lighter trunk mass. The effects of anthropometrics tend to be consistent in both individuals with and without LBP. Previously, marked associations have been identified between body mass, BMI, and the MPF slope of ERS in males and females with and without LBP [38]. Moreover, a marked association between TTF and torso mass in females with and without LBP has been demonstrated [39]. Trunk mass is an important factor in Sørensen test TTF, especially in those with a history of LBP, and should be considered when assessing performance. A variation in the Sørensen test that normalizes the workload to a consistent percentage of maximal strength would account for differences in trunk mass and strength to allow for a more objective assessment of endurance.

In this study, self-efficacy emerged as an important factor in Sørensen test performance. Motivation has long been recognized as a consequential factor in Sørensen test performance [22,36,40-43]; however, to the best of our knowledge, we are the first to attempt to manipulate it through the use of a virtual reality video game. The self-efficacy measure was created specifically for the Sørensen test task, which likely explains its strong association with TTF in the LBP group. Interestingly, self-efficacy was not predictive of performance in the healthy group. Thus, in this sample of individuals without a history of LBP, it appears that self-efficacy did not drive performance. Alternatively, our sample of individuals with a history of recurrent LBP performed better on the virtual reality Sørensen test if they reported higher ratings of confidence in their capacity to perform the task; this is consistent with our previous findings on the standard Sørensen test [34] as well as those obtained by others who found that performance was predicted, in part, by self-efficacy [22]. This would suggest that self-efficacy may be a worthwhile target for cognitive behavioral interventions for LBP.

In addition, there was a significant effect of kinesiophobia in our LBP group; those who had lower TSK scores maintained a longer TTF on the virtual reality Sørensen test; this would suggest that TSK is predictive of performance in individuals with recurrent LBP when provided with a distraction element. On the classic version of the Sørensen test [34], TSK was not predictive of performance in this same group of individuals with recurrent LBP. The virtual reality video game may have actually exacerbated fear cognitions by blocking the participant’s view of the real world, reducing their sense of control, and instead redirecting focus toward their pain-related fear. Moreover, it is possible that attentional resources were reduced in response to the game. Future research could benefit from investigating the response to different types of games to determine whether certain games are more effective or whether games are not effective in any form to counteract pain-related fear.

Previous research has demonstrated an effect of sex on performance on the Sørensen test. Females tend to maintain the test position for longer than males [43,44]; however, several studies have found no sex differences [45,46], and others have found that males maintain the test position for longer than females [47-49]. We did identify a trend toward a sex difference on TTF with females maintaining the position slightly longer than males (P=.06). Others have reported that healthy females maintained an isometric trunk extension task markedly longer than healthy males [50]. The authors attributed their results to the muscle mass and strength hypothesis, which describes the relationship among the total muscle mass, vascular compression, and the demand for oxygen. According to this hypothesis, because females typically have lower muscle mass, the vasculature is less compressed during isometric exercise, and the demand for oxygen to the active muscles is lower [51,52]. There is some evidence that females have a greater ratio of type I oxidative muscle fibers in the trunk extensors [53], which would have a greater concentration of beta-2 adrenergic receptors, enhancing vasodilation [54]; there is also evidence that females have a greater degree of capillarization in some muscles [55], enhancing perfusion. However, others have demonstrated that intramuscular pressure may not be associated with a shift in MPF during isometric trunk extension exercises [56,57]. Although muscle mass was not measured in this sample, it is possible that TTF was influenced by perfusion.

As with any study of human subjects, this study is not without its limitations. Our LBP group consisted of individuals with mild, recurrent LBP, which may have also restricted the sample to individuals with low levels of disability and pain-related fear. Individuals with higher disability and pain-related fear have poorer rehabilitation outcomes and typically perform more poorly on the Sørensen test. Thus, significant pain-related fear associations may have emerged in a sample of individuals with more severe kinesiophobia and disability symptoms. In addition, this group was primarily young, fit, college-aged students; future studies will benefit from measuring physical activity levels because cardiorespiratory fitness is likely associated with performance on any endurance task, such as the Sørensen test.

This study demonstrates that individuals with and without mild, recurrent LBP perform similarly on a variation of the Sørensen test using a virtual reality video game, but the underlying mechanisms driving performance vary between the groups. Performance on this variation of the Sørensen test in healthy individuals is driven primarily by physiological factors, including trunk mass and the MPF slopes of ERS and BF. In addition, trunk mass is an important factor of performance in individuals with a history of recurrent LBP; however, levels of self-efficacy and kinesiophobia also appear to be important predictors of TTF on this virtual reality Sørensen test.