This is the first study to show increases in activities of the erector spinae and multifidus during arm- and leg-ergometer exercises with increased workload. The results suggest that although the metabolic responses were higher in the arm exercise relative to the leg exercise, the activities of the paraspinal muscles were similar in both exercises. Furthermore, the timing of activation of the erector spinae was almost similar to that of the ipsilateral biceps, i.e., agonist muscles, and that of the multifidus was activated oppositely during arm-ergometer exercise, suggesting that these paraspinal muscles increased their activities to support rotational moment of the trunk. The activation time windows of the two paraspinal muscles were opposite to those of the ipsilateral vastus lateralis during the leg-ergometer exercise, suggesting that these paraspinal muscles increased their activities to prevent the lateral bending moment of the trunk on the tilted side.
Our results showed differences in resting muscle activities of the erector spinae muscle with other muscles (Fig. 2), while the increases in sEMG signals of the erector spinae and multifidus were comparable during arm- and leg-ergometer exercises. Previous studies showed that the activity of the erector spinae changes with the angle of pelvic tilt, and this may be the reason for the difference at rest in the present study [24–25].
In the arm-ergometer exercise, the trunk rotated during pulling an object with one arm, and the activity of the ipsilateral erector spinae increased in order to increase the rotation moment, which was similar to the observation reported by Lavender et al. [26]. In addition, the multifidus is reported to support contralateral trunk rotation [27]. The above previous findings suggest that the ipsilateral erector spinae and the contralateral multifidus were active because of slight trunk rotation during pulling the handle with ipsilateral arm. In the leg-ergometer exercise, tilting of the seated person on coronal plane was associated with the appearance of lateral bending moment of the trunk on the tilted side, and activation of the contralateral erector spinae and multifidus, which allowed stabilization of the trunk [28]. The ipsilateral bending moment of the trunk seemed to have occurred during stepping on the pedal with one leg in the present study. The results suggest that simultaneous contractions of the contralateral erector spinae and multifidus during the leg-ergometer exercise opposed the ipsilateral bending moment of the trunk, thus ensuring maintenance of trunk balance. Furthermore, the strength of the erector spinae and multifidus seems to be depended on the magnitude of the rotation moment and of the lateral bending moment due to the increase in workload during the arm- and leg-ergometer exercises, respectively.
Also, in agreement with previously data [29], our results showed increased HR and oxygen uptake with the increased workload during the arm- and leg-ergometer exercises, but the increases in these two parameters at any given increase in workload were higher during the arm-ergometer exercise than the leg-ergometer exercise [29]. Thus, the present study suggests that the stress on the erector spinae and multifidus themselves during the arm-ergometer exercise is similar to that during the leg-ergometer exercise, although the energy efficiency is worse during arm-ergometer exercise compared with to leg-ergometer exercises.
Limitations
The present study did not include motion analysis during the two forms of exercises. One-minute submaximal arm-ergometer exercise in healthy participants was reported to be accompanied by trunk rotation [30]. Also, anterior-posterior changes in the center of gravity were described during 30 minutes of arm- and leg-ergometer exercises at 50% of maximum [31]. Trunk movement similar to that mentioned above could have occurred in the present study, though this is speculative at present.
The present participants were able to complete the ergometer exercise at 50W and 100W, workloads, which are relatively high for frail elderly subjects. Furthermore, the present study involved only 2 minutes of exercise, whereas the recommended training time for aerobic exercise in patients with Parkinson's disease is a minimum of 20 min, for example [22]. Therefore, the effect of prolonged or repeat arm- and leg-ergometer exercise on paraspinal muscles remains unclear.
Perspectives
Patients with Parkinson's disease experience more falls than age-matched healthy subjects [32] and have shown higher baseline levels of paraspinal muscle activity during walking, which is attributed to reduced muscle strength [19]. In such patients, exercise-based interventions can improve motor dysfunction and imbalance [33], which cannot be controlled by pharmacological and surgical interventions [34]. Current rehabilitation recommendations to improve posture in these patients include core trainings, which promote activities of trunk muscles [35]. Because prolonged repeated contractions of the erector spinae and multifidus involuntarily would be more beneficial for the patients than just performing core-building trainings, we think that a prospective study is needed to assess the effects of arm- and leg-ergometer exercises in elderly persons and/or patients with Parkinson's disease.