Physical Therapy Korea

Sleep accounts for approximately one-third of a person’s lifetime, and people relieve mental and physical fatigue during the sleep. It plays an important role in recovery, energy storage, memorization, strengthening of the immune system, and emotions regulation [1,2]. Sleep deprivation can lead to various symptoms, such as imbalanced body rhythms, daytime fatigue, decreased concentration, severe emotional dysregulation, increased appetite, and weight gain [1-4]. There is a growing realization that the quantity as well as the quality of sleep is important. Correct sleeping posture, sufficient quantity, and an environment for good quality sleep significantly impact human cognitive, behavioral, and emotional aspects, and are essential for promoting normal activity [4].

The positioning of the head and neck is critical for overall sleep quality [5,6]. Sleeping with cervical alignment beyond the normal range increases the muscle tone in the neck and shoulders and can cause musculoskeletal pain in the head, neck, shoulder, and temporomandibular joint. Therefore, sleeping in an incorrect posture is a major factor that reduces sleep quality [7,8]. Experts suggest that encouraging the natural lordotic curve of the cervical spine is crucial to promote deeper sleep for longer durations [9].

Pillows support the cervical spine and the back of the head and are closely associated with neck pain [10-12]. Previous studies have shown that the filling materials, height, shape, and area of cervical and occipital supports are essential factors of a good pillow for optimal sleep [13-15]. Traditionally, rectangular pillows filled with polyester, duck down or goose down, have been used [16]. Although these pillows offer fluffiness and comfort, they are unable to provide sufficient support for maintaining proper cervical alignment during sleep.

Subsequently, a variety of ergonomically designed pillows made of latex and other memory foam materials have been designed to distribute pressure and provide support for maintaining normal C-shaped lordosis of the cervical spine [12,17,18]. Although the degree of comfort varies depending on the material, hardness, height, and shape of the foam, many users experience discomfort in adapting to the pillows and sleeping for a long time [9]. Especially, contour pillow design with cervical support was effective in reducing contact pressure values and maintaining an appropriate cervical angle [16,17]. Therefore, more recently, hybrid pillows that combine the advantages of conventional cotton pillows and cervical foam pillows have been designed, consisting of a memory foam core and a floppy top layer.

While numerous pillows of various sizes, shapes, and materials are being designed to improve sleep posture, there is still a lack of research on their efficacy on biomechanical characteristics in supine posture. Therefore, we conducted an experiment to guide pillow design and selection, potentially improving sleep quality for manufacturers and users. We compared head-neck posture and subjective comfort levels among three pillows: microfiber-filled, firm neck support foam, and hybrid, in supine and side-lying positions. The null hypothesis of this study posits that there are no differences in head-neck pressure distribution, head-neck alignment, muscle tone, and subjective comfort among all pillow conditions.

1. Subjects

This study included twenty-eight healthy adults (19 males and 9 females) with an average age of 29 years. We used G*Power ver. 3.1 (Kiel University) to determine sample size. With an alpha level of 0.05 and a desired 0.80 power. Choosing a small Cohen’s d effect size of 0.25, the analysis suggested a sample of 28 participants. The exclusion criteria were: (1) neurological and musculoskeletal disorders and (2) surgical history of the cervical spine or shoulders. All participants obtained written informed consent by signing an informed consent form approved by the Institutional Review Board of Yonsei University (IRB no. 1041849-202010-BM-153-02) after fully understanding the purpose and method of the study.

2. Procedure

Participants laid down for supine position 5 minutes in each of the four different pillow conditions: (1) no pillow (NP), (2) neck support foam pillow (NSFP), (3) standard microfiber filled pillow (SFP), and (4) hybrid foam pillow (HFP). The order of pillow application was randomized for each pillow to minimize the potential impact of order effects. Pressure data were collected during the last one minute of the 5 minutes. The two examiners measured the craniovertebral angle (CVA), cranio-horizontal angle (CHA), chin-sternum distance (CSD), and muscle tone at the 4-minute mark under each condition (Figure 1). After all measurements were completed, participants were instructed to lie on each pillow in the side-lying position. Subsequently, they were prompted to assess the subjective comfort level associated with each pillow in both the supine and side-lying postures.

Figure 1. Four different pillow conditions. (A) No pillow, (B) neck support foam pillow, (C) standard microfiber filled pillow, and (D) hybrid foam pillow.

3. Pillow Types

Among the commercialized foam pillows designed to provide firm neck support by maintaining natural cervical curve, the Kobuk memory foam pillow (Balance 9) was selected for this experiment (Figure 2). This all-in-one design pillow is made of soft integral skin memory foam of moderate or more hardness, and the size is 370 × 169 × 90 mm. According to the manufacturer, this pillow was designed to relax and stretch the posterior neck muscles by maintaining a normal neck curve during sleep. Because the neck support pillow is smaller than an ordinary foam pillow, it is large enough to support the lower part of the occipital region and curvature of the cervical spine, rather than providing full support for the entire head.

Figure 2. Two type of pillow. (A) Neck support foam pillow and (B) standard microfiber filled pillow.

A 500 × 700 mm rectangular SFP was selected for this experiment (“microfiber 300 T” pillow, Bazaar) (Figure 2). The pillow has microfiber filling and a high-density microfiber fabric cover (100% polyester). The height and hardness of the pillow could not be measured because its shape was easily deformed upon applying pressure. Moreover, the manufacturer did not clarify the height and hardness of the pillow.

The HFP used in this experiment has the combined advantages of the NSFP and SFP. This pillow consists of a polyurethane memory foam core and a microfiber fabric cover. The upper layer of the cover is filled with fluffy microfiber stuffing (Figure 3). The size of the memory foam core is 400 × 600 mm, and the total size of the pillow with the cover is 500 × 700 × 140 mm. Considering the users’ size and preferences the horizontal edges on both sides of the hybrid pillow are designed in a convex shape with two different heights (60 and 70 mm) to support the natural curvature of the cervical spine. The center of the core has a wide and deep hemispherical concave shape to distribute the pressure more evenly to the entire occipital region. Both sides of the pillow are designed to be higher than the center to provide comfort in the side-lying position. Additionally, the margin of the concave is gradually connected to the surface of both sides of the pillow to allow smooth head turning during posture transition.

Figure 3. Hybrid foam pillow. (A) Memory foam core and (B) complete set of the pillow with cover.

4. Outcome Measures

When the participants laid on each pillow for 5 minutes, the peak pressure under their head and neck area was measured using the X-sensor (X3 PX100, XSENSOR Technology Corporation). The X-sensor is a device measuring 36 × 36 cm in size, consisting of 1,296 individual pressure sensors. It is utilized in various pressure-related research studies [19,20]. Each 10 mmHg was color-coded from blue to red, with a pressure output range of 0 to 256 mmHg. The red and dark blue colors represent the highest pressure and the pressure near zero, respectively. To identify the relative position of the head and neck on the sensor pad, the lowest border of the X-sensor pad was matched to the acromial tip. The pressure distribution displayed on the notebook monitor was captured to analyze the pattern of pressure distribution for each participant in the pillow condition. The highest peak pressure value of the pixel among all pixels was recorded for statistical data processing.

The CVA is the acute angle formed by a straight line connecting the spinous process of C7 to the tragus of the ear and a horizontal line passing through the spinous process of C7 [21]. A smaller angle indicates more forward head posture (FHP). The CHA is the angle between the horizontal line at the tragus of the ear and the line between the tragus and lateral angle of the eye [21]. It is used as an estimated value of the relative head position on the upper cervical spine [21]. The CVA and CHA while lying on a pillow were measured according to the method used by Hida et al. [22] in their research; these values were named as the cranio-vertebral angle in supine (CVAs) and cranio-horizontal angle in supine (CHAs) in this study, respectively. The CSD measured using a tape measure, represents the degree of neck flexion [23]. The lengths from the CVAs, CHAs and CSD were measured to compare the degree of FHP while lying comfortably on three different pillows (Figure 4). Two markers were placed on the left side of the tragus and shoulder tip points before capturing an image. A smart-phone camera was placed on a tripod located 1 m from the participant to capture a steel photograph of the neck and head areas in the sagittal plane. Then, the CVAs and the CHAs were determined using the Image J program (National Institutes of Health).

Figure 4. Measurement of CVAs, CHAs, and CSD. (A) Contact between back and back, (B) tragus, (C) canthus, (D) chin, and (E) sternal notch. CVAs, cranio-vertebral angle in supine; CHAs, cranio-horizontal angle in supine; CSD, chin-sternum distance.

The tone of the sternocleidomastoid muscle (SCM) was assessed using the MyotonPRO (Myoton AS) device with the subject in a supine position, ensuring comfort. Measurements were taken parallel to the muscle fibers, precisely midway between the mastoid process and the clavicle. Myoton has been validated in previous studies as a reliable instrument for characterizing muscle function [24,25]. The manufacturer’s manual defines the muscle tone as the frequency of oscillation (Hz) of a muscle in its resting state without any voluntary contractions. Right SCM was selected for convenience in setting the equipment.

After completing all measurement procedures, each participant was asked to report the subjective comfort of each pillow in supine and side-lying positions on a 5-point Likert scale (1: maximum discomfort, 2: minimum discomfort, 3: moderate, 4: comfort, 5: maximum comfort).

5. Statistical Analysis

All the statistical analyses were performed using IBM SPSS Statistics ver. 26.0 (IBM Co.). One-way repeated measures analysis of variance (ANOVA) was used to compare the mean values of the measured variables in each condition. All statistical significance levels were set at 0.05.

The statistically significant univariate F-statistics were evaluated using Bonferroni for post-hoc analysis in order to account for multiple comparisons. For significant post-hoc results, we computed both 95% confidence intervals and effect sizes. Effect size (d) is defined as a ratio of the mean change score divided by the standard deviation of the baseline scores. According to Cohen [26], an effect size of 0.20 or less represents a small change; 0.50 represents a moderate change; and 0.80 represents a large change.

Additionally, differences in the level of subjective comfort, assessed using the Likert scale, between the pillows were analyzed using Friedman ANOVA. Wilcoxon signed-rank tests were conducted for pairwise comparisons of comfort levels between each pair of the pillow conditions.

The participants’ characteristics are presented in Table 1.

Table 1 . General characteristics of the participants (N = 28).

Variable	Value
Sex (male/female)	19/9
Age (y)	29.3 ± 6.1
Height (cm)	173.4 ± 9.4
Weight (kg)	75.3 ± 16.5
BMI (kg/m2)	24.8 ± 3.6

Values are presented as number only or mean ± standard deviation. BMI, body mass index..

1. Pressure Under the Head and Neck

The pressure under the head and neck was most concentrated on a small area in the NSFP condition, while the pressure was more distributed in the HPF condition (Figure 5). The peak pressure value was highest in the NSFP condition (166.56 ± 7.57 mmHg) and decreased in the order of NP (128.67 ± 9.82 mmHg), SFP (77.71 ± 7.25 mmHg), and HFP conditions (57.03 ± 2.96 mmHg) (Table 2, Figure 6A). The peak pressure in the HFP condition was significantly lower than that in other pillow conditions. Additionally, the peak pressure in the NSFP condition was significantly higher than that in all other pillow conditions.

Table 2 . Head-neck alignment and muscle tone in four pillow conditions.

Variable	(A) NP	(B) NSFP	(C) SFP	(D) HFP	F	p-value (Bonferroni)
Peak pressure (mmHg)	128.67 ± 9.82	166.56 ± 7.57	77.71 ± 7.25	57.03 ± 2.96	59.72	< 0.001 (B>A>C>D)
CVAs (°)	47.51 ± 6.17	39.61 ± 3.16	32.54 ± 3.00	35.29 ± 3.22	97.10	< 0.001 (A>D>C>B)
CHAs (°)	16.55 ± 0.99	10.01 ± 1.00	–0.15 ± 0.94	6.73 ± 0.59	120.50	< 0.001 (D>A>C>B)
CSD (cm)	11.54 ± 0.31	11.32 ± 0.32	9.32 ± 0.32	10.23 ± 0.29	60.99	< 0.001 (A>D>B>C)
Muscle tone (Hz)	13.27 ± 0.18	12.61 ± 0.17	12.37 ± 0.13	12.48 ± 0.15	30.39	0.24

Values are presented as mean ± standard deviation. CVAs, cranio-vertebral angle in supine; CHAs, cranio-horizontal angle in supine; CSD, chin-sternum distance; NP, no pillow; NSFP, neck support foam pillow; SFP, standard microfiber filled pillow; HFP, hybrid foam pillow..

Figure 5. Head-neck pressure distribution under the four pillow conditions from a typical subject. (A) No pillow, (B) neck support foam pillow, (C) standard microfiber filled pillow, and (D) hybrid foam pillow.

Figure 6. Head-neck pressure distribution and alignment values of pillow conditions. (A) Peak pressure, (B) CVAs, (C) CHAs, and (D) CSD. NP, no pillow; NSFP, neck support foam pillow; SFP, standard microfiber filled pillow; HFP, hybrid foam pillow; CVAs, cranio-vertebral angle in supine; CHAs, cranio-horizontal angle in supine; CSD, chin-sternum distance. *p < 0.05, **p < 0.01, ***p < 0.001.

2. Head and Neck Alignment

The CVAs was highest in the NP condition (47.51° ± 6.17°) and decreased in the order of NSFP (39.61° ± 3.16°), HFP (35.29° ± 3.22°) and SFP conditions (32.54° ± 3.00°) (Table 2, Figure 6B). The CVAs in the NP condition were significantly higher than those in other pillow conditions. Furthermore, the CVAs of the SFP condition were statistically lower than those in other pillow conditions.

The CHAs was highest in the NP condition (16.55° ± 0.99°), and it decreased as the NSFP (10.01° ± 1.00°), HFP (6.73° ± 0.59°), and SFP conditions (–0.15° ± 0.94°) (Table 2, Figure 6C); each pair of comparison revealed a significant difference.

The CSD was highest in the NP condition (11.54 ± 0.31 cm) and decreased in the order of NSFP (11.32 ± 0.32 cm), HFP (10.23 ± 0.29 cm), and SFP conditions (9.32 ± 0.32 cm) (Table 2, Figure 6D). Based on post-hoc test results, all pairwise comparisons, except for the comparison between conditions NSFP and NP, showed significant differences.

3. Muscle Tone of the SCM

The muscle tone was highest in the NP condition (13.27 ± 0.18 Hz) and decreased in the order of NSFP (12.61 ± 0.17 Hz), HFP (12.48 ± 0.15 Hz), and SFP conditions (12.37 ± 0.13 Hz). However, there were no significant differences between the pillow conditions (Table 2).

4. Level of Subjective Comfort

A higher subjective comfort score indicates higher comfort. The order of the average ranks of the subjective comfort scales in the supine and side-lying positions was HFP, SFP, and NSFP (Table 3). The Friedman test revealed a significant difference in the average ranks of the subjective comfort scale across the pillow conditions (p < 0.05) in both positions. The level of comfort in the HFP condition (p < 0.001) was statistically higher than that in the NSFP condition in both positions.

Table 3 . Level of subjective comfort.

	Average rank			χ2	df	p-value
	NSFP	SFP	HFP
Supine	1.66	1.98	2.36	7.62	2	0.02*
Side-lying	1.27	2.23	2.50	27.15	2	< 0.001

NSFP, neck support foam pillow; SFP, standard microfiber filled pillow; HFP, hybrid foam pillow. *p < 0.05..

The level of comfort was significantly higher in the SFP condition than NSFP condition (p < 0.001) only in side-lying. However, no statistically significant difference between the HFP and SFP conditions (p = 0.07) in the side-lying position (Table 3).

This cross-sectional experiment was conducted to provide supporting evidence for pillow selection by comparing the biomechanical characteristics and subjective comfort of three types of pillows in healthy young adults. The main findings of this study are as follows: 1) the peak pressure value was highest in the NSFP condition, followed by the NP, SFP, and HFP conditions; 2) the CVAs, CHAs, and CSD were highest in the NP condition, followed by the NSFP, HFP, and SFP conditions; 3) the muscle tone of the SCM was highest in the NP condition, followed by the NSFP, HFP, and SFP conditions; and 4) the participants’ subjective comfort level was highest in the HFP condition, followed by the SFP and NSFP conditions in both the supine and side-lying positions.

This experiment revealed that lying on NSFP concentrated pressure on the lower occipital and upper neck regions. This finding is consistent with that of Shen et al. [27] that the peak pressure is high in pillows with low elasticity and high hardness. In contrast, the pressure was evenly distributed while lying on the HFP, which is due to the specially designed concave shape of the central core of the HFP to maximize the pressure distribution by avoiding the concentration of weight under the convex shaped occiput.

The CVAs, CHAs, and CSD were selected to determine the pillow that maintained the most appropriate cervical spine alignment. The CVAs represents the total neck flexion value, and a smaller CVA indicates a greater FHP [28,29]. The CHAs represents capital flexion, and the CSD represents total neck flexion; a lower value represents greater flexion. The normal CVA of the cervical spine in an upright posture is between 48° and 50° (Abbasi et al. [28]), and the normal angle of the CHA is 16.3° [30]. The CVAs and CHAs in the NP condition were closest to the normal upright values. The magnitudes of CVAs, CHAs, and CSD showed similar tendencies across the three conditions. the capital and cervical flexion were lowest in the SFP condition, followed by the HFP and NSFP conditions.

Good supportive pillows can reduce stress during sleep. If a pillow design provides appropriate support for the head and cervical areas, it could successfully reduce undesirable muscle activation during sleep. Therefore, we expected a certain type of pillow to effectively decreases the muscle tone. However, no significant different was found in the SCM tone among the pillow conditions. A previous study reported that the EMG activity of the SCM was lower with the cervical pillow than with the general pillow [31]. According to another study, there was no significant main effect among pillow types in both participants groups, with and without FHP [32]. Unfortunately, the findings of the current muscle tone study and previous EMG studies are inconsistent in terms of muscle activation in both the muscle tone (which represents a more passive biomechanical property of the muscle) and EMG amplitude (which reflects a more active characteristic of a muscle) of the SCM, and the influence of different pillow designs on SCM muscle activity remains unclear. In our experiment, the muscle tone of each pillow was measured after 5 minutes of use; however, a longer use duration is required to measure the difference in the SCM tone for each pillow more accurately.

Previous studies reported that people initially feel more comfortable with softer pillows, however, their perception of comfort changes over time [9,33]. Similar to previous studies, we also found that the subjective comfort level was the highest in the HFP condition, followed by the SFP, and NSFP conditions after using the pillows for 5 minutes in both the supine and side-lying positions. However, in our experiment, SFP was found to be associated with greater capital and cervical flexion. Therefore, it cannot be concluded that the subjective comfort experienced over a short period guarantees optimal comfort through optimal cervical alignment support during extended periods of sleep.

The neck flexion angle was close to the normal upright value in the NSFP condition, despite the peak pressure being the highest and the comfort level being the lowest in the NSFP condition. Firm pillows feel less comfortable initially, but provide support to the cervical area after adaptation, which is desirable for preventing spinal deformation [9]. Therefore, in cases where alignment correction and myofascial release for the head and neck area is necessary, a firm NSFP type of pillow with the C-shaped alignment support may be more useful, even if it leads to reduced comfort due to inadequate pressure distribution, rather than prioritizing long-term comfort during sleep. Taking into account the overall results of this study in terms of pressure distribution, cervical alignment, and the level of subjective comfort, it is suggested that a hybrid pillow, which combines the benefits of NSFP and SFP, may present a valuable alternative to conventional pillows.

This research has yielded findings that can be useful for pillow selection. However, the small sample size and the imbalance in the number of male and female participants within the sample, which could potentially influence body size, are notable limitations. Furthermore, this study only investigated the specific characteristics of each pillow during a 5-minute lying period. Therefore, as the next step in research regarding pillows, a study is suggested to investigate the impact of biomechanical and sleep science-related characteristics of pillows based on their shape and material on long-term sleep with normal and symptomatic populations.

In conclusion, our research offers valuable insights into the selection of the most appropriate pillow, considering both biomechanical factors and comfort preferences. The findings emphasize the potential advantages of a hybrid pillow that incorporates features from both NSFP and SFP, especially regarding pressure distribution and neck support. However, in situations where achieving optimal cervical alignment and promoting myofascial release for the head and neck region is paramount, it is advisable to opt for a firm NSFP pillow, even if it means sacrificing some comfort. Further investigation is warranted to gain a comprehensive understanding of the long-term impact of various pillow characteristics on sleep quality among diverse populations.

None.

None to declare.

Kyeong-Ah Moon, Ji-Hyun Kim, Ye Jin Kim, Joo-Hee Park, and Hye-Seon Jeon have a patent application pending for the research discussed in this paper. The authors assert that any potential conflicts of interest did not impact the design, execution, or reporting of the research presented herein. The study was conducted impartially, and the findings are communicated with transparency and integrity. Furthermore, the patent application was submitted through “Yonsei University.”

Conceptualization: KAM, JHK, YJK, JHP. Data curation: KAM, JHK, YJK, JHP. Formal analysis: KAM, JHK. Investigation: KAM, JHK. Methodology: KAM, JHK, YJK, JHP. Project administration: KAM, JHK. Software: KAM, JHK. Supervision: HSJ. Validation: HSJ. Visualization: KAM. Writing - original draft: KAM, HSJ. Writing - review & editing: KAM, JHK, HSJ.