Abstract
Introduction
Upon acquiring the ability to walk erect, human beings evolved to freely use their hands, but erect posture places a considerable weight burden onto the lower abdominopelvic region [1]. Over 90% of lower back pain is recovered, regardless of cause, but 50% of the latter gets chronic disability due to constant lower back pain [2]. This disability causes medical, social, and economic loses to both individuals and the overall economy [3].
The lower back must not only support the upper body, but some people spend one third to one half of their working time sitting down [4]. Maintaining constant postures for a long time can cause low back pain and deterioration [4]. It can lead to musculoskeletal problems from specific muscle overload combined with abdominal muscles relaxation [5]. Moreover, musculoskeletal damage due to muscle overloading can cause spinal instability resulting from muscle weakness near the spine, damage in trunk soft tissue, decreases in muscle endurance, and intervertebral disc degeneration [6]. Continued spinal instability can lead to interbody damage and, over time, muscle near the spine shows decreased cross-sectional area, worsening lower back pain [7, 8].
Trunk muscle activity is required for spinal stability, and recent literature addressing lower back pain focuses on changes in postural muscle activity that may contribute to lower back pain [9–11]. Different postural positions can be characterized by the activation and relative contributions of different postural muscles, and may variously contribute to the recovery from or worsening of chronic lower back pain [12]. O’Sullivan et al. [13] compared seated postures by investigating trunk muscle activity and differences between large and small muscle group activities in relation to spinal curvatures. These muscles play a very important role in lumbo-pelvic stabilization and postures related to lower back pain. In particular, lower back pain or other ailments arise among students or workers who maintain poor postures for long time periods. Regardless of postural differences, pressure applied to the posterior spine when seated is seven times greater than that when lying down and three times greater when standing up [8]. This stress is greater upon the lumbar spine when sitting than when standing, increasing the risk for lower back pain [8].
The sitting lifestyle is common in eastern Asia, and considerable work is performed in seated postures. However, studies on strain applied to spine or trunk muscles in relation to different seated postures, are lacking. In South Korea, where sitting in various postures is part of the broader culture, research studies describing biomechanical effects of sitting on the human body, and studies on the relationship between posture and pain are lacking.
Hence, we hypothesized that there are differences between trunk muscle activities depending on sitting postures. The present study compares trunk muscle activities in four types of seated postures: cross-legged, long, side, and W-shaped.
Methods
Participants
This was a cross-sectional study. The purpose and methods of the study were posted in the schoolbulletin to recruit potential participants. Ten volunteers were recruited and screened according to the following criteria: healthy adults in their 20 s, absence of back or lower extremity problems such as spinal disk herniation, no history of previous spine operations, and the ability to maintain a seated posture for 10 minutes or longer. Volunteers with congenital back deformities or orthopedic or neurological problems were excluded. Two volunteers were excluded and eight participated in the study. There were 4 male and 4 female participants, with average age of 22.9 (0.99) years, weight 62.3 (12.26) kg, and height 169.6 (8.11) cm.
All participants provided signed informed consent after receiving an explanation of the study purpose and procedures. The study was approved by the Kyungnam University Institutional Review Board.
Procedures
General characteristics (sex, age, weight, and height) of the participants were collected through a brief interview. Trunk muscle activity was measured using surface electromyography (sEMG). Maximal voluntary isometric contraction (MVIC) was recorded for four muscles, the external oblique (EO), rectus abdominis (RA), latissimus dorsi (LD), and erector spinae muscles (ES) on both sides of the body. Recording electrode locations were based on the SENIAM recommendations [14]. Electrode pads were attached about 3 cm above the iliac crest for EO, 3 cm lateral to the navel for RA, 3 cm lateral to T10 for LD, and 5 cm lateral to L4 for ES. Skin regions were shaved and cleaned with alcohol swabs before attachment. Activity of each muscle was collected for 10 seconds, in three trials, with a two minute break between trials. MVIC was calculated based on 5 seconds of collected data (between 3 and 7 seconds). Participants were shown how to assume the sitting positions including the cross-legged, long, side, and W sitting postures, and allowed to practice them (Fig. 1). After a 10 minute break, trunk muscle activities of the EO, RA, LD, and ES on both sides of the body were collected in the four sitting postures. The participants were instructed to assume the four positions in any order. The participants were instructed to sit, shoulder width apart, with arms relaxed at their sides. In addition, they were instructed to focus straight ahead at a designated point, as much as possible, to maintain an upright posture. The upright position was defined as a slight anterior rotation of the pelvis to achieve neutrallordosis of the lumbar spine and relaxation of the thorax, as described by O’Sullivan et al. [10]. Each posture was maintained for 10 minutes, before a 10-minute break between postures to reduce influence of the preceding posture on muscle activity. The participants had rest periods with their comfortable posture such as lay down. sEMG signals in each muscle were collected for 10 minutes while maintaining each posture. The data were collected as mean of the 3rd and 8rd minute and were extracted for each posture and used for statistical comparisons.
Electromyography
sEMG (Tringo™ Wireless, Delsys, USA) was used to collected muscle activities of the EO, RA, LD, and ES. Data were processed using standard filtering and rectifying methods. The sampling rate was 2000 Hz, and a 60 Hz high-pass filter and 10 Hz low-pass filter were applied (all filters, zero-lag 4th order Butterworth). For normalization of the data, the root mean square (RMS) value of the raw EMG data was calculated. The EMG data of each muscle was normalized by calculating the RMS of the five second MVIC of the muscle. The EMG data as mean of the 3rd and 8rd minute collected during the sitting postures were also normalized by calculating the RMS, and then expressed as a percentage of the MVIC value.
Statistical analysis
SPSS 18.0 software (SPSS, Inc., USA) was used for statistical analysis. General characteristics of the participants were analyzed using descriptive statistics, and one-way analysis of variance was used to compare trunk muscle activities between the different sitting postures. The level of statistical significance was set at α= 0.05.
Results
There were no significant differences in electromyographic muscle activity of EO, RA, LD, and ES in the four postures (p > 0.05). However, in the W-shaped sitting posture, the left LD showed the greatest electromyographic muscle activity, followed by the right LD and left EO, respectively. The right and left LD in the long sitting posture and left ES in the side sitting posture showed greater electromyographic muscle activity than those of other muscles. In cross-legged sitting, electromyographic muscle activities were less than in other postures (Table 1).
Discussion
In the present study, trunk muscle activities in the four types of sitting postures, including the cross-legged, long, side, and W sitting postures, were investigated using EMG. The results revealed no significant differences in trunk electromyographic muscle activities of the EO, RA, LD, and ES recorded for each of the four sitting postures, inferring that the various seated postures do not differentially affect trunk muscle activity.
Kang et al. compared lumbar flexion angle and relative muscle activities between participants under constrained hip movement and those with unconstrained hips while sitting cross-legged [15]. Trunk muscle activities of the RA, EO, and internal oblique were analyzed by EMG. Lumbar flexion angle showed a significant difference between the two groups, however electromyographic activities of trunk muscle showed no significant differences. This study showed that postural change, while maintaining a seated, cross-legged sitting posture can change lumbar flexion angle without affecting muscle activity. The present study similarly shows that changes in leg posture without spinal change does not significantly influence muscle activity. The results of the present study are supported by the following: Hong reported significant differences in muscle activity, with postural changes, including tilting 10° to the left, 5° to the left, 10° to the right, 5° to the right, and sitting straight [16]. O’Sullivan et al. [10] compared muscle activation when sitting with an upright spinal posture and that when sitting with a slumped backward spinal posture. The internal oblique, lumbar multifidus, and spinalis thoracis of the erector spinae showed significant activity differences between an upright and slumped backward spinal posture. Slumping backward increased thoracic extension, decreased lumbar extension and posterior pelvic tilt, and resulted in significantly increased muscle activation of the spinalis thoracis and external oblique [13]. These studies focused on comparisons of muscle activation in various trunk positions. The fact that our study did not involve changes in trunk position can explain why our results differed from those of previous studies. The differences in results may also reflect the changes in trunk muscle activation in response to changes in position.
The present study suggests that there are no significant differences in trunk electromyographic muscle activation in cross-legged, long, side, and W sitting postures. However, we found the greatest electromyographic muscle activity in the left LD, followed by the right LD and left EO, respectively, in the W sitting posture. The right and left LD in the long sitting posture and left ES in the side sitting posture showed greater electromyographic muscle activity than that of other muscles. To our knowledge, this study is the first to investigate trunk muscle activity in postures associated with seated activities common in Asia, including Korea. Thus, the present study is of value in determining that various sitting postures do not show appreciable differences in trunk muscle activity. It should be noted that previous studies [8, 16] were conducted with participants sitting on a chair, as they might find it difficult to engage in cross-legged, long, side, and W sitting.
The present study has several limitations. There was small sample size. Only one direction was preferred by the participants in the side sitting posture, and the activity of the multifidus muscle, which is reportedly the most active muscle in the seated posture, was not investigated. Also, muscle activities were not investigated in relation to the spinal angle in each sitting posture. In the Future, further study will be needed.
Conflict of interest
The authors have no conflict of interest to report.