Changes in deep neck muscle length from the neutral to forward head posture. A cadaveric study using Thiel cadavers

Abstract Forward head posture (FHP) is one of the most common postural deviations. Deep neck muscle imbalance of individuals with FHP is of primary concern in clinical rehabilitation. However, there is scarce quantitative research on changes in deep neck muscle length with the head moving forward. This study aimed to investigate changes in deep neck muscle length with different severity levels of FHP. Six Thiel‐embalmed cadavers (four males and two females) were dissected, and 16 deep neck muscles in each cadaver were modeled by a MicroScribe 3D Digitizer in the neutral head posture, slight FHP, and severe FHP. The craniovertebral angle was used to evaluate the degrees of FHP. Quantitative length change of the deep neck muscles was analyzed using Rhinoceros 3D. In slight FHP significant changes in length occurred in four muscles: two shortened (upper semispinalis capitis, rectus capitis posterior minor) and two lengthened (longus capitis, splenius cervicis). In severe FHP all occipital extensors were significantly shortened (10.6 ± 6.4%), except for obliquus capitis superior, and all cervical extensors were significantly lengthened (4.8 ± 3.4%), while longus capitis (occipital flexor) and the superior oblique part of the longus colli (cervical flexor) were lengthened by 8.8 ± 3.8% and 4.2 ± 3.1%, respectively. No significant length change was observed for the axial rotator. This study presents an alternate anatomical insight into the clinical rehabilitation of FHP. Six muscles appear to be important in restoring optimal head posture, with improvements in FHP being related to interventions associated with the occipital and cervical extensors.

the head when watching a computer screen, for example, may facilitate FHP (Neumann et al., 2017). Maintaining an imbalanced position may change the functional resting length of muscles, creating a habitual unnatural posture due to the changed length-tension relationship and proprioceptive feedback (Kendall et al., 2005;Khan et al., 2020;Moustafa et al., 2020).
The deep neck muscles are of primary concern in FHP treatment because of their relatively high muscle spindle density, which underpins the precision of movement and proprioceptive information (Liu et al., 2003). Rehabilitation of the deep neck muscles has been observed to improve stability of head and neck posture, manifested as increased ability to maintain an upright position of the cervical spine (Blomgren et al., 2018;Falla et al., 2007).
There are, however, inadequate quantitative studies on deep neck muscle length changes. Most of the literature provides qualitative descriptions, with only a single study (Khayatzadeh et al., 2017) reporting quantitative data on neck muscle length changes using a computer-generated model based on fresh-frozen cadaveric spines. Cartner et al. (2011) cautioned that the mechanical resistance of fresh-frozen cadaveric specimens decreases with increasing exposure time.
Further quantitative verification is, therefore, needed to inform clinical decision making in the rehabilitation of FHP. Thiel-embalmed cadavers have lifelike characteristics, enabling the biomechanical characteristics of muscle, such as mechanical strain rate effects of tendons and ligaments, elasticity, and stiffness of muscles, to be evaluated (Joy et al., 2015;Liao et al., 2015;Vollner et al., 2017). The craniovertebral angle (CVA) is widely used in FHP evaluation, having high validity and excellent inter and intra rater reliability (Migliarese & White, 2019;Salahzadeh et al., 2014). The present study investigated the length changes in deep neck muscles from the neutral posture to different FHP severity levels using Thiel cadavers, with the aim of identifying the changing lengths in deep neck muscles and the muscles most affected by FHP.

| Specimens and preparation
The study was conducted using six Thiel-embalmed cadavers (four males, two females) with a mean age at death of 86.2 ± 8.7 (71-97) years from the Centre for Anatomy and Human Identification, University of Dundee. Inclusion criteria for the study were that the cadavers had to have intact deep cervical muscles, cervical and thoracic spine; the exclusion criterion was a history of spinal surgery. All cadaveric research was conducted in compliance with relevant anatomical legislation, with all donors having given their consent in accordance with the Anatomy Act (1984) and the Human Tissue (Scotland) Act (2006). This study was approved by the Ethics Committee of the Centre for Anatomy and Human Identification. All donors also gave permission for images to be taken.
Each cadaver was placed in the lateral decubitus position and supported on both sides of the trunk. Wooden "pillows" were used to keep the head, neck and spine in alignment and to maintain a horizontal gaze (i.e., parallel to the ends of the dissecting table) in order to facilitate measurement of the CVA. Regional dissection, by the same investigator, was then carried out in the cervical region to expose the deep cervical muscles. The deep cervical muscles selected for study were 16 muscles in five different functional groups (Table 1).
Primary landmarks (anterior and posterior tubercles of transverse processes of C3, C4, and C5; transverse processes of C7, T1, and T2; spinous process of C7; center point of the anterior surface of the T3 vertebral body) in the cervical region were marked using pins ( Figure 1), enabling the attachments of the deep cervical muscles to be consistently located when measuring muscle lengths.

| FHP simulation
As previously noted, CVA is a valid reliable measure to define neutral head posture and FHP (Migliarese & White, 2019;Salahzadeh et al., 2014). It is the angle between a horizontal line passing through the C7 spinous process and a line extending from the tragus of the ear to the C7 spinous process in the sagittal plane (

| Measurement device and data collection
The lengths of the deep neck muscles were obtained using the Micro-Scribe 3D Digitizer (Immersion Corporation, California) and Rhinoceros 3D software. The MicroScribe 3D Digitizer was placed in the same position relative to the head for each cadaver. The length of each selected muscle was determined using the stylus of the digitizer by measuring the distance between their proximal and distal attachment points. The muscle attachment sites used in the present study were reported by Khayatzadeh et al. (2017), thereby ensuring that their results could be compared directly with the current study, as well as providing a larger sample size for future research. All selected muscles were considered to pass between their attachments in a straight line, with the length and spatial information being input to the Rhino 3D software. Each measurement was taken twice for each head posture, from which the mean was determined. Any change in muscle length between the neutral head posture and slight or severe FHP was expressed as a percentage of the length in the neutral head posture.

| Statistical analysis
Statistical analyses were performed using SPSS v23.0 (IBM SPSS Statistics, Armonk, NY). The normality of the data distributions was assessed using the Shapiro-Wilk test, with test-retest reliability being assessed using the intraclass correlation coefficients (ICC) and 95% confidence intervals (CI): two-way mixed-effects model, absolute agreement and single measurement were selected for the ICC analysis. The reliability criteria were set as <0.50 = poor, 0.5 to 0.75 = moderate, 0.75 to 0.9 = good, and >0.90 = excellent (Koo & Li, 2016

| Muscle length changes in different head postures
The length changes of the deep cervical muscles, the percentage change in length from the neutral posture and results of LSD tests are shown in Table 2 and Figure 3.
F I G U R E 1 Primary landmarks for locating attachments of neck muscles. The primary bony landmarks were the anterior and posterior tubercles of the transverse processes of C3 (pink), C4 (green), and C5 (yellow), the transverse processes of C7, T1, and T2 (black), the spinous process of C7 (green), and the center point of the anterior surface of the T3 vertebral body (not shown)

| Occipital extensors
The occipital extensors showed an overall shortening trend as the head moved forward (Table 2 and Figure 3). From the neutral head posture to slight FHP only two muscles significantly shortened, these being upper semispinalis capitis (À4.1 ± 2.7%, p = 0.022) and rectus capitis posterior minor (À9.3 ± 5.3%, p = 0.01). From the neutral head posture to severe FHP, however, all occipital extensors, except obliquus capitis superior, shortened significantly (p < 0.05): the mean shortening was À10.6 ± 6.4%. The most prominent changes in muscle length were observed in rectus capitis posterior major and minor in both slight and severe FHP (Figure 3).

| Cervical flexors
The three parts of the longus colli exhibited different trends (Table 2 and Figure 3). No significant muscle length changes were observed as the head moved forward to slight FHP. In severe FHP only the superior oblique part of longus colli lengthened significantly (4.2 ± 3.1%, p = 0.025).

| Axial rotator
No significant length change was observed in obliquus capitis inferior in either slight or severe FHP (Table 2 and Figure 3).

| DISCUSSION
The present study investigated changes in deep neck muscle length between the neutral head posture, defined as a CVA of 55 , and two severity levels of FHP, defined by CVA 45 and 35 , using Thiel cadavers. The observations showed that some muscles were subject to significant length changes when placed in the two simulated FHPs, with consistent patterns of change with increasing FHP severity.
These observations provide the basis for developing a new strategy for the rehabilitation of FHP.

| Responses of deep neck muscle in different FHP severities
The deep neck muscles respond differently to slight and severe FHP; in slight FHP, four muscles significantly changed their length ( Figure 3), with two shortening (upper semispinalis capitis, rectus capitis posterior minor), and two lengthening (longus capitis, splenius cervicis). These four muscles also had the highest effect size values (η p 2 > 0.14). As well as the length changes observed in slight FHP, rectus capitis posterior major and semispinalis cervicis also showed significant length changes in severe FHP. Restoration of the length of these six muscles could be conducive to regaining optimal head posture.
The observations of this study suggest that these six muscles should be the target for rehabilitation of FHP; there is also some supporting evidence from previous studies. Fernandez-de-las-Penas, Alonso-Blanco, Cuadrado, Gerwin, and Pareja (2006) reported that FHP was associated with the presence of suboccipital myofascial trigger points, which had similar locations to rectus posterior major and minor. Hallgren et al. (2017) were able to verify this, observing that rectus capitis posterior major and minor had significantly increased electromyographic activity in FHP. In an ultrasonographic study (Goodarzi et al., 2018), it was reported that in individuals with FHP the semispinalis capitis exhibited the smallest thickness changes during voluntary isometric contraction. FHP influences the ability to activate semispinalis capitis effectively.
Furthermore, 4-week deep cervical flexor training was observed to be more effective in improving FHP and neck pain than conventional isometric training (Gupta et al., 2013). Deep cervical flexor training is a low load craniocervical flexion exercise, which mainly strengthens longus capitis and longus colli (Blomgren et al., 2018). A randomized controlled trial investigating the effects of semispinalis cervicis and deep cervical flexor training noted that both improved FHP, neck disability, pain intensity and muscle strength (Suvarnnato et al., 2019). The results of the present study indicate that future FHP rehabilitation research could focus on the six muscles above, that is, upper semispinalis capitis, rectus capitis posterior major/minor, longus capitis, semispinalis cervicis, and splenius cervicis.

| Changing patterns of deep neck muscle length and cervical spine movement
FHP is the result of changes in the alignment of the cervical spine, and although this alignment has substantial cervical subtype variability (Daffin et al., 2019), the present study considered FHP to be associated with upper cervical extension and mid-lower cervical flexion. This The changing patterns of muscle length observed in the current study were similar, but not identical, to the findings of the qualitative studies (Haughie et al., 1995;Simons et al., 1999) and the computerized model study (Khayatzadeh et al., 2017). There is now mounting consensus that during FHP there is shortening of the occipital extensors and lengthening of the occipital flexors. Qualitative descriptions report that the posterior cervical and suboccipital muscles become shortened, and the anterior neck flexors become lengthened (Haughie et al., 1995;Simons et al., 1999), while the computerized-model study of Khayatzadeh et al. (2017)

| Muscle weakness and the length-force relationship
Both shortening and lengthening of a muscle would impair its ability to generate muscle force (Goodarzi et al., 2018;Levangie & Norkin, 2011). Based on the length-force relationship of skeletal muscles, the maximal force generation can only be achieved at the optimal muscle length (Levangie & Norkin, 2011). Consequently, muscles with altered resting lengths may have lower mechanical efficiency and impaired contractile function, reducing their mechanical economy during daily activities (Szczygiel et al., 2020). Goodarzi et al. (2018) reported that maximal voluntary isometric contraction of the neck extensors significantly decreased with increasing FHP severity. Oliveira and Silva (2016) also reported a significant association between neck muscles endurance (deep neck flexor and neck extensor), FHP and neck pain, which support the impact of FHP on muscle contractile function and the length-force relationship.

| New insight into the clinical rehabilitation of FHP
Deep cervical flexor training is a common corrective treatment for FHP by strengthening longus capitis and longus colli (Falla et al., 2007;Gupta et al., 2013;Kim & Kwag, 2016;Lee et al., 2013); however, its clinical efficacy is still controversial. Sikka et al. (2020) reported that a 4-week deep cervical flexor training program did not correct FHP in adolescents. Subbarayalu and Ameer (2017) (Falla et al., 2004), rather than individuals with FHP. Moreover, the relationship between FHP and neck pain remains unclear since confounding factors (e.g., age) may play an important role (Mahmoud et al., 2019). Deep cervical flexor training may, therefore, be more effective in alleviating pain than improving FHP (Sikka et al., 2020).

| Limitations
There are some limitations in the current study. First, the study is a cadaveric study where CVA measures were taken in a side lying position and not in an upright position. Second, the mean age of the cadavers (86.2 ± 8.7 [71-97] years) indicated that degenerative changes would occur in the aging spine (Tetreault et al., 2015). For example, osteoarthritic changes were likely to be present in C5/6 and C6/7 segments (White et al., 2007), which could alter the adjacent cervical mechanics when simulating FHP, thus changing the apparent length of the longer neck muscles. Third, the sample size was relatively small; a few muscles with large effect sizes showed no change in length (

ACKNOWLEDGMENTS
The authors would like to thank cadaver donors for donating their bodies to science so that anatomical research could be performed.
Results from such research can potentially increase humankind's overall knowledge that can then improve patient care. Therefore, these donors and their families deserve our highest gratitude. We would also like to thank Professor Roger Soames for his helpful comments on our draft.