The Effect of an Orthotic Device for Balancing Neck Muscles During Daily Office Tasks

Introduction

Neck pain is a common medical condition that represents a major public health concern in terms of personal health, overall well-being (Cote, Cassidy, & Carroll, 2001; Fejer, Kyvik, & Hartvigsen, 2006), and medical costs (Bassols, Bosch, & Banos, 2002; Borghouts, Koes, Vondeling, & Bouter, 1999). In Europe and North America, one-third of adults experience neck pain in the course of one year (Bovim, Schrader, & Sand, 1994; Cote et al., 1998; Croft et al., 2001). A large number of researchers pointed out that prolonged sedentary work in computer workers and office administrators is one of the major risk factors to developing neck pain (Brandt et al., 2004; Dong, 2005; Eriksen, Natvig, Knardahl, & Bruusgaard, 1999; Korhonen et al., 2003). Sedentary work potentially leads to changes in body posture such as forward head posture (Bokaee et al., 2016), which is characterized by greater flexion of the lower cervical spine, greater extension of the upper cervical spine, and higher scapular protraction and elevation than those commonly observed in a correct posture (Szeto, Straker, & Raine, 2002). This posture, assumed for a prolonged time, induces a creep response in the tissues (Harms-Ringdahl, Ekholm, Schuldt, Nemeth, & Arborelius, 1986; Twomey & Taylor, 1982) and increases compressive loading in the cervical spine as well as neck muscle activation (Schuldt, Ekholm, Harms-Ringdahl, Nemeth, & Arborelius, 1986). It has been shown that patients with neck pain experience higher activation of neck flexor (Lindstrom, Schomacher, Farina, Rechter, & Falla, 2011; Zito, Jull, & Story, 2006) and neck extensor (Johnston, Jull, Souvlis, & Jimmieson, 2008; Lindstrom et al., 2011; Szeto, Straker, & O’Sullivan, 2005) muscles. An excessive activation of neck muscles could threaten the balance between agonist and antagonist muscles, resulting in a lower stability of the head. Thus, assuming that a higher activation of neck muscles can lead to developing neck pain, new strategies should be adopted to reduce neck muscles activation. For this reason, the effect of wearing orthotic devices that could help in reducing activation of neck muscles represents a major goal of ergonomics.

In recent years, it was claimed that the use of counterweights applied in the occipital region of the cranio-cervical area was effective in reducing neck extensors activation (Dal Monte & Giannini, 2016; Giombini et al., 2013; Harrison et al., 2007; Pavan, Frigo, & Pedotti, 2013). The basic idea was that a slight load applied to the back of the head may reduce the tension of the neck muscles required to counterbalance the head weight. This strategy has been recently investigated in helicopter pilots, aiming at reducing the incidence of neck pain due to the use of night vision goggles (M. F. Harrison et al., 2007). In recent years, a special cap, referred to as neck balance system (NBS), which includes a small mass of 0.3/0.5 kg over the occipital region, has been proposed by scientists (Dal Monte & Giannini, 2016; Giombini et al., 2013; Pavan et al., 2013; Quinzi, Giombini, Pigozzi, & Macaluso, 2018) to reduce the activation of neck extensor muscles and possibly avoid the onset of neck pain in populations at risk. Patients with chronic neck pain who adopted NBS improved their posture (Pavan et al., 2013) and experienced lower neck pain during activities of daily living (Giombini et al., 2013) as well as a reduction of neck muscle activation during car driving (Menotti et al., 2015). The assumption is that the mass applied over the occipital region reduces the activation of neck extensor muscles without increasing the activation of flexor muscles by rebalancing head weight distribution on the neck (Menotti et al., 2015).

The head posture and the activation of neck muscles is also influenced by the position of other spinal segments (Menotti et al., 2015). For instance, an increased lumbar flexion is related to a higher activation of neck extensors (Caneiro et al., 2010). For this reason, some investigators looked at the effects of a lumbar support showing that this device leads to lower prevalence of low back pain in regular lumbar support users, such as taxi (Chen, Dennerlein, Chang, Chang, & Christiani, 2005) or bus (Leinonen, Kankaanpaa, Vanharanta, Airaksinen, & Hanninen, 2005) drivers, and it is useful in reducing the activation of low back muscles (B. J. Andersson & Ortengren, 1974; Kingma & van Dieen, 2009) during sitting postures (B. J. Andersson, Ortengren, Nachemson, Elfström, & Broman, 1975). It has also been shown that the combined use of NBS and a lumbar support reduces activations of both neck extensor and low-back muscles in urban drivers (Menotti et al., 2015).

Although the beneficial effects of wearing the NBS on reducing the activation of neck muscles have been shown by a number of investigations (Dal Monte & Giannini, 2016; Giombini et al., 2013; Pavan et al., 2013), no study looked at the effect of NBS during ordinary static activities, like working with a computer. This is a task that is commonly carried out by a particular category of workers who are prone to developing neck pain.

This study aims at understanding the effect of NBS on the activation of neck muscles during simple static tasks executed in a sitting position in front of a computer screen, thus reproducing common office work activity. Furthermore, the combined effect on neck and lower back extensor muscles of wearing the NBS and lumbar support also was investigated. It was hypothesized that the combined use of the NBS together with a lumbar support leads to a reduction in the activation of neck and back extensor muscles compared with using no device.

Materials and Methods

Participants

Twenty healthy individuals (age = 33.3 ± 11.4 years, body mass = 67 ± 10 kg, height = 1.72 ± 0.07 m, physical inactivity = 6.75 ± 2.84 h/d) volunteered to participate in the study. Volunteers were divided into two groups according to their gender and matched for age and physical inactivity levels (Table 1).

OPEN IN VIEWER

Table 1:

Anthropometric Characteristics of Participants Recruited in the Present Study

	Males (N = 10)	Females (N = 10)
	Mean (SEM)	Mean (SEM)
Age (years)	33.3 (2.54)	33.2 (2.68)
Mass (kg)	75.05 (1.55)	59.33 (1.54)
Height (m)	1.77 (0.009)	1.67 (0.01)
Physical Inactivity (h/d)	7.2 (0.63)	6.56 (0.66)

Individuals were recruited from students and employees of the local university. Volunteers were included in the study according to the following criteria: (1) age between 20 and 60 years, (2) no symptoms of low back and cervical pain, and (3) no symptoms of neuromuscular diseases. After a thorough explanation of the procedures, all volunteers signed a written informed consent. This research complied with the Declaration of Helsinki, and the experimental protocol of the investigation was approved by the local Institutional Review Board.

Experimental Procedure

Each experimental session consisted of three tasks that volunteers performed in a sitting and static position in front of a computer screen. These tasks were: reading, typewriting, and following a cue moving up and down the computer screen at a pace of 25 bpm. Each task lasted 3 minutes. Each task was repeated in four experimental conditions: (a) with the neck balance system (NBS), (2) with the neck balance system plus the lumbar support (NBS + LS), (3) with the lumbar support only (LS), and (4) using no devices (ND). Tasks and experimental conditions were performed in a random order.

Before the experimental conditions were performed, all volunteers were asked to perform the maximal voluntary contractions (MVC) of neck flexors and extensors and lower back extensors. Following a standardized warm up, which consisted of flexing and extending the head for 5 minutes, participants were asked to lay on an examination table and extend their neck as forcefully as possible. A belt placed at inion height was used to connect the head of the participants through a load cell to an immovable load (Figure 1, Panel A). For the MVC of neck flexors, volunteers were asked to lay in a supine position on the examination table and perform a maximal contraction of their neck flexor muscles (Figure 1, Panel B). Their head was connected to an immovable load by means of an inelastic belt placed on the forehead of volunteers. The MVC of back extensor muscles was performed by asking the volunteers to lay on the examination table in a prone position and maximally extend their back against an immovable resistance applied by the experimenter. During each MVC, participants were verbally encouraged to achieve their maximum and maintain it for at least 3 seconds before relaxing. Three MVC attempts for each muscle group were performed, interspersed by 5-minute rest (Menotti et al., 2012).

OPEN IN VIEWER

Figure 1.

Volunteer positioning for the assessment of the maximal voluntary contractions (MVC) of (A) neck extensors and (B) flexors.

Workstation

The workstation was chosen for allowing subjects to be seated in the correct position while performing office tasks. Following ergonomics guidelines (Triano, 2010), a desktop system was chosen so that the screen could be adjusted to account for individual anthropometric characteristics. A comfortable, fully adjustable chair was chosen for the subject to sit in, with the backrest reclined at 100° to 110°, as suggested by ergonomics guidelines (Triano, 2010). Thus, at the beginning and during the test, participants were encouraged to respect the following criteria: (a) The computer monitor had to be directly centered in front of the subject, (b) the monitor had to be at a comfortable height in order to point the eyes directly at the top third of the screen, (c) the monitor had to be placed at a comfortable horizontal distance for viewing, and participants (d) had to rest their hands on the computer keyboard with elbows angle at 90°, (e) position their upper arms and elbows as close to the body and as relaxed as possible for typing, and (f) keep feet flat on the floor with the thighs parallel to the floor.

Data Processing

All data recorded by the portable sEMG recording system were analyzed offline using OTBioLab. To remove possible high-frequency noises, sEMG signals were digitally band-pass filtered between 20 and 300 Hz. According to the literature, the high-pass cutoff frequency at 20 Hz was used to reduce motion artifacts with minimal impact on the total sEMG power (Clancy, Morin, & Merletti, 2002; De Luca, Gilmore, Kuznetsov, & Roy, 2010; Pinheiro, dos Santos, & Chaves, 2016). In fact, as shown by De Luca et al. (2010), 20 Hz offers the best compromise for optimizing the informational content of the sEMG signal compared to 10 and 30 Hz.

During the MVCs, the amplitude of the sEMG of the semispinalis capitis, sternocleidomastoids, and erector spinae was quantified by computing the root mean square (RMS) over 1-s windows (Laudani et al., 2013). Subsequently, the window showing the maximal RMS amplitude was chosen for further analysis (RMS_Max).

During each daily office task, the amplitude of the sEMG signal of semispinalis capitis, sternocleidomastoids, and erector spinae was quantified by computing the RMS of the sEMG signal over 1-s epochs. Similarly, the median frequency of the power spectrum was computed over adjacent 1-s windows. Muscular activations recorded during the three tasks and four experimental conditions were expressed as a percentage of the maximum RMS observed during the MVC of each muscle (RMS_Max).

OPEN IN VIEWER

Figure 2.

Volunteer and orthotic devices positioning during the three tasks investigated. The upper box depicts the neck balance system (NBS), showing the positioning of the small masses in its posterior portion (see inset in the upper right part). In the lower box, the lumbar support (LS) is depicted.

Results

In the four experimental conditions and across the different tasks, muscle activation, expressed as a percentage of MVC, was very small, ranging from approximately 1.5% to approximately 5%, depending on the task to be performed and the muscle considered (Table 2). In general, across tasks and conditions, the semispinalis capitis showed the highest activation compared to the other muscle investigated, followed by the ERS, whereas the SCMD showed the lowest activation across tasks and conditions.

OPEN IN VIEWER

Table 2:

Muscle Activation in Percent of the Maximal Voluntary Contractions (MVC) Averaged Across Tasks for the Semispinalis Capitis Muscle (SPC), Erector Spinae (ERS), and Sternocleidomastoid (SCMD)

Average Muscular Activation (%MVC)
	Experimental Condition
Muscle	NBS	NBS + LS	LS	ND
SPC	3.97	4.49	4.48	4.61
ERS	2.31	2.26	2.41	2.40
SCMD	1.68	1.46	1.62	1.53

Note. NBS = neck balance system; LS = lumbar support; ND = no device.

Muscle Activation

Semispinalis capitis muscle

Figure 3 depicts the activation of the semispinalis capitis across different conditions and tasks for the right and left sides of the body. The statistical analysis performed on the activation of the semispinalis capitis muscle showed a significant main effect of condition, F(3, 16) = 5.723, p = .007. The post hoc analysis showed a reduced activation of the semispinalis capitis muscle in the NBS condition with respect to the ND condition (p = .001), NBS + LS condition (p = .002), and LS condition (p = .009), whereas no significant differences were observed between the other experimental conditions.

OPEN IN VIEWER

Figure 3.

Activation of the semispinalis capitis (SPC) muscle, expressed as a percentage of the maximal activation recorded during the MVC for reading, typewriting, and nodding across the four experimental conditions tested (NBS, NBS + LS, LS, ND).

In addition, a significant main effect of task was observed, F(2, 17) = 29.296, p < .001. The post hoc analysis revealed that writing elicited greater muscle activation with respect to reading (p < .001) and nodding (p < .001). Reading and nodding showed similar muscle activation (p = .161).

No significant main effect of side was observed, F(1, 18) = 0.760, p = .395. Similarly, the RMANOVA showed no significant main effect of gender, F(1, 18) = 0.002, p = .965.

Erector spinae muscle

In Figure 4, the amplitude of the activation of the ERSL across condition and tasks for the right and left sides of the body is shown. No significant effect of condition was observed for the activation of ERS muscle, F(3, 16) = 0.252, p = .859. Conversely, a significant main effect of task was observed for this muscle, F(2, 17) = 12.819, p < .001. The post hoc analysis showed that writing resulted in an increased activation of this muscle with respect to reading (p < .001) and nodding (p < .001). A tendency toward statistical significance was observed for side, F(1, 18) = 4.24, p = .054, with the right ERS showing slightly greater activation than its homologue (2.63% vs. 2.15%). Also for this muscle, no significant effect of gender was observed, F(1, 18) = 1.39, p = .254.

OPEN IN VIEWER

Figure 4.

Activation of the erector spinae (ERS) muscle, expressed as a percentage of the maximal activation recorded during the MVC for reading, typewriting, and nodding across the four experimental conditions tested (NBS, NBS + LS, LS, ND).

Sternocleidomastoid muscle

The activation of the sternocleidomastoid muscle was not modified by the experimental condition, F(3, 16) = 1.40, p = .250. Similarly, no effect of task, F(2, 17) = 0.032, p = .968, and side, F(1, 18) = 0.507, p = .486, on SCMD activation was observed (Figure 5). Also for this muscle, no significant effect of gender were observed, F(1, 18) = 3.40, p = .081.

OPEN IN VIEWER

Figure 5.

Activation of the sternocleidomastoid (SCMD) muscle, expressed as a percentage of the maximal activation recorded during the MVC for reading, typewriting, and nodding across the four experimental conditions tested (NBS, NBS + LS, LS, ND).

Discussion

The main finding of this study was that NBS reduced the activation of neck extensor muscles, while it had no effect on neck flexor muscles. The experimental analysis also showed that activation of neck and back muscles was not influenced by the use of the lumbar support alone as well as its use combined with the NBS. Furthermore, muscle activation was influenced by tasks but not gender.

Across the four conditions, the NBS condition was associated with a reduction of muscular activation of the semispinalis capitis muscle, whereas the other conditions (ND, LS, and NBS + LS) had no significant effect on semispinalis capitis, sternocleidomastoid, and erector spinae muscles. In keeping with previous studies focusing on the effect of the NBS on neck extensor muscles in car drivers (Menotti et al., 2015) or helicopter pilots (M. F. Harrison et al., 2007), our results showed that such devices are effective in reducing neck muscle activation also during daily office tasks. Neck extensor muscles have the main function of supporting the head, which, due to its anatomical properties, tends to tilt forward. As the cranium is bigger than its underlying vertebrae and the center of mass of the head is positioned forward relative to the fulcrum (C7 vertebrae, point of support), neck extensor muscles need to exert a continuous tonic contraction to generate the moment of force (moment) necessary to prevent the head from tilting forward. In this framework, the slight balancing mass of the NBS, which is applied posteriorly to the fulcrum, generates a supplementary neck extensor moment that sums up to that generated by neck extensor muscles, thus contributing to head balance (De Col, 2008). It is worth noting that no differences across the four experimental conditions (NBS, NBS + LS, LS, ND) were observed for the activation of the sternocleidomastoid muscle. This result is in accordance with previous literature (Dal Monte & Giannini, 2016; Menotti et al., 2015; Pavan et al., 2013) that observed a comparable activation of neck flexor muscles between the combined use of these two ergonomic devices and the use of no device during car driving.

In the present study, no difference in muscle activation was observed between the ND, LS, and NBS + LS conditions. Consistently with the study of Leinonen et al. (2005), in our study, the activation of erector spinae muscle was very small, reaching approximately 2.4% of that observed during the MVC (see Table 2).

Literature concerning the effect of the lumbar support on back extensor muscles activation is inconsistent: Leinonen et al. (2005) failed to show differences in sEMG amplitude of the paraspinal muscles while using or not using a low back support in bus drivers, whereas the effectiveness of this device alone or combined with the NBS in reducing sEMG activity of low back and neck muscles was demonstrated (B. J. Andersson & Ortengren, 1974; Menotti et al., 2015). It has been proposed that this ergonomic device promotes a physiological posture of the lower back, thus distributing the load on the vertebral discs evenly (B. J. Andersson & Ortengren, 1974; G. B. Andersson, Murphy, Ortengren, & Nachemson, 1979). Therefore, even if in our study the use of lumbar support did not result in lower activation of erector spinae or semispinalis capitis, the usefulness of this device would have been suggested for reducing low back pain. For instance, Chen and colleagues (2005) reported a lower prevalence of low back pain in regular lumbar support users versus nonusers. Regrettably, none of these studies (B. J. Andersson & Ortengren, 1974; Leinonen et al., 2005) provided an explanation for the mechanisms underlying the effectiveness or not of the lumbar support. It has been shown that lumbar support enhances lumbar lordosis in a sitting posture (G. B. Andersson et al., 1979), decreases intra-discal pressure (B. J. Andersson & Ortengren, 1974), and rotates the pelvis forward (D. D. Harrison, Harrison, Croft, Harrison, & Troyanovich, 1999). This latter aspect may represent a crucial point to explain the similar muscle activation observed for neck extensors in our study between the NBS + LS, LS, and ND conditions. Different investigations reported changes in the sagittal spinal curvatures associated with pelvis orientation (Boulay et al., 2006; Legaye, Duval-Beaupere, Hecquet, & Marty, 1998). In particular, it was shown that an increased pelvic incidence, defined as the angle between the line perpendicular to the sacral plate at its midpoint and the line between this latter point to the axis of the femoral head (Legaye et al., 1998), relates to increased sagittal curvature of the spine (Boulay et al., 2006). An increased sagittal curvature may alter the physiological loading through the spine as a consequence of a shift in trunk mass, leading to increased flexion moments and compression and shear forces imposed on spine segments (Kiefer, Shirazi-Adl, & Parnianpour, 1998). In addition to increased mechanical loading, changes in spinal posture may compromise back extensor muscles strength, as suggested by Mika, Unnithan, and Mika (2005), and the normal function of paraspinal musculature (O’Sullivan et al., 2002) perhaps due to alterations in length-tension relationships, moment arm lengths, and force vector orientations (McGill, Hughson, & Parks, 2000). In our study, the lumbar support may have modified pelvis incidence, thus exacerbating sagittal spinal curvature. To account for these possible modifications in spine curvatures and maintain their gaze on the central part of the computer screen, participants may have adapted their neuromuscular control strategy by increasing the activation of the neck extensors, thus cancelling the beneficial effect of the NBS when used in combination with the LS.

A tendency toward significant effect of side was observed for erector spinae muscle but not semispinalis capitis and sternocleidomastoid muscles. In particular, the right erector spinae showed slightly greater activation than its homologue muscle during all tasks and conditions (right ERS > left ERS). A possible explanation for the asymmetry in ERS activation can be found in the greater use of the right side of our body due to the utilization of the mouse. In our study, all participants were right-handed and used a right-hand mouse configuration. It could be argued that our participants relied on a greater activation of their right-side muscles to accomplish the tasks proposed in our study. Coherently, Leinoinen at al. (2005) showed an asymmetric sEMG activation of the trapezius muscle, with the left side being more activated than the right side in bus drivers. The authors suggested that the increased left trapezius activity is probably due to the fact that drivers mainly use their left hand to steer the bus. Accordingly, in our study, the higher activation of the right erector spinae may have been related to the characteristic of the tasks performed.

The results presented in our study present some limitations. First, we failed to make a postural analysis of participants during the task. This further analysis would have provided crucial information on the possible modifications of posture observed across experimental conditions. Future investigation should consider evaluating possible modifications in posture during the use of orthotic devices. Another important limitation of this study is related to the assessment of comfort or discomfort rate immediately after performing tasks for each condition. However, previous studies (Giombini et al., 2013) showed that participants using the NBS had a significant reduction in Neck Disability Index (NDI), Neck Pain and Disability Scale (NPDS), and visual analogue scale (VAS) score. Furthermore, these scores were reduced at the end of the 8-week treatment and maintained at 12 weeks after the end of treatment. Similarly, Menotti et al. (2015) found no differences in the level of perceived comfort assessed by the VAS scale between driving a car with the two ergonomic devices and driving without the devices. In the same vein, we failed to assess the performance of the three tasks during the four experimental conditions. Indeed, a thorough quantification of the performance of the participants, for example in terms of lines typewritten or read per minute, could have further supported or not the use of these devices. However, at anecdotal level, none of the participants reported any feeling of impaired performance, nor did we notice any particular difficulty in performing the tasks in any of the experimental conditions. Last, although the use of the NBS induced a reduction of the activation of neck extensor muscles, a slight increase of the compression force over the underlying anatomical structures should be taken into account. Whether a chronic exposure to this increased compression force over cervical structures may represent a risk factor or not for the onset of possible pathologies of the cervical spine has to be ascertained yet.

In conclusion, using counterweights applied at the occipital region of the head reduces the muscular activation of neck extensor muscles in healthy subjects. Thus, ergonomics devices, such as the NBS, may be adopted by computer workers to reduce the activation of neck extensors and possibly reduce the risk of developing neck pain. Future studies are warranted to compare the acute effect of the NBS, alone or combined with a lumbar support, between healthy subjects and subjects with cervical and low back pain by combining electromyographic and motion capture analysis.

Key points