Psychological and Physical Risk Factors for Pressure Pain Hypersensitivity of the Upper Trapezius in Food Service Workers With Nonspecic Neck/shoulder Myofascial Pain

Objectives: It is unclear which factors increase the risk of developing pressure pain hypersensitivity, a type of neurophysiological hyperexcitability. The present study investigated the relative contributions of physical and psychological factors to pressure pain hypersensitivity of the upper trapezius for each sex. Methods: In total, 154 individuals with neck/shoulder myofascial pain participated, among 372 food service workers. Participants completed a questionnaire (age, sex, Beck Depression Inventory, and Borg Rating of Perceived Exertion scale) and then were photographed to measure posture. Pressure pain sensitivity, two range of motions (cervical lateral-bending and rotation), and four muscle strengths (serratus anterior, lower trapezius, biceps, and glenohumeral external rotator) were measured by a pressure algometer, iPhone application, and handheld dynamometer. For each sex, forward multivariate logistic regression was used to test our a priori hypothesis among selected variables that a combination of psychosocial and physical factors contributed to the risk for pressure pain hypersensitivity. Results: In multivariate analyses, lower trapezius strength (odds ratio = 0.94, 95% condence interval = 0.91–0.97, p = 0.001) was the only signicant risk factor for pressure pain hypersensitivity in men. Dominant painful ipsilateral cervical rotation range of motion (odds ratio = 0.96, 95% condence interval = 0.92–0.99, p = 0.037) was the only risk factor for pressure pain hypersensitivity in women. Discussion: Lower trapezius strength and dominant painful ipsilateral cervical rotation range of motion could serve as guidelines for preventing and managing pressure pain hypersensitivity of the upper trapezius in food service workers with nonspecic neck/shoulder myofascial pain.

Tenderness over muscles is a common clinical nding in painful conditions of presumed muscular origin [13,14]. Pressure pain sensitivity (PPS) is a quantitative sensory test for assessing pain sensitivity in deep tissues [15]. This neurophysiological test may be suitable to measure PPS and tissue tenderness because these conditions are believed to reduce the test values [15].
Although the pathophysiologic mechanism of MP has not been identi ed, it may involve central sensitization (hyperresponsiveness and hyperexcitability of the central nervous system) [16,17]. However, it is unclear which factors increase the risk of developing pressure pain hypersensitivity (PPH) in terms of neurophysiological hyperresponsiveness and hyperexcitability. Suggested factors include individual factors (e.g., sex) [18], physical factors (e.g., posture) [19], and psychosocial factors (e.g., depression and stress) [20]. Previous studies have investigated the in uences of individual factors on PPS. Thus, there is a need to investigate the combined in uences of multiple factors on the risk for PPS.
Physical risk factors are useful and potentially could be modi ed with interventions such as exercise [21].
Conversely, individual characteristics (e.g., sex and age) cannot be modi ed. To determine a speci c management approach for neck/shoulder pain, Donatelli proposed examination of the following aspects: cervical and shoulder posture, muscle length, rotator cuff muscle strength, and scapular rotator strength [22]. Psychological, biological, and social domains could explain differences in pain severity and perception between men and women with MP [18,23]. Concerning sex differences in pain outcomes, PPS has demonstrated the greatest effect size [24] and women are more sensitive to pressure pain than men [25]. Because of sex differences in pain perception and PPS, as well as differences in food service tasks, the relative contributions of physical and psychological risk factors to PPH should be identi ed for each sex.
Therefore, the present study investigated differences in physical and psychosocial factors between participants according to PPH status for each sex, and the relative contributions of physical and psychological factors to PPH for each sex.

Participants
Participants were recruited through a questionnaire to con rm their experience of neck/shoulder MP as FWs. In total, 154 individuals with neck/shoulder MP participated, from among 372 FWs in a theme park.
A owchart for recruitment of the participants is shown in Fig. 1. Inclusion criteria were non-traumatic neck/shoulder pain, > 6 months of work in food service, presence of neck/shoulder pain for ≥ 3 months, and visual analog scale score > 30 mm. Exclusion criteria were shoulder fractures, a prior diagnosis of shoulder instability, a history of surgery in the shoulder, any systemic disease, untreated psychiatric condition, examination suggesting the presence of neurological diseases or internal diseases, hypertension (resting systolic blood pressure > 150 mm Hg or diastolic blood pressure > 90 mm Hg), and/or pregnancy. Participant characteristics are shown in Table 1. The experimental protocol was established according to the ethical guidelines of the Helsinki Declaration. The study protocol was approved by the Yonsei University Mirae Campus Institutional Review Board (certi cation number: #1041849-201603-BM-005-02). Before assessment, the investigator explained the entire experimental procedure and all participants voluntarily provided informed consent. Instruments, Inc., Greenwich, CT, USA) with a 1 cm diameter rubber tip attached to a strain gauge that displayed values in kg/cm 2 . The tip was applied to the UT at a standardized location containing the midpoint between C7 and the acromion process, in the dominant painful side (intraclass correlation coe cient of inter-rater reliability: 0.91) [26,27]. PPS was de ned as the lowest pressure at which the sensation of pressure turned to slight pain or discomfort [18,26,27]. The mean value of three trials was calculated and used for the main analyses. A 1 min resting period was allowed between each recording. Both the participant and examiner were blind to force readings during the assessment. A standard metronome was also used to control the rate of increase in pressure. Men with pressure pain sensitivity < 2.9 kg/cm 2 in the UT and women with pressure pain sensitivity < 2.0 kg/cm 2 in the UT were presumed to have PPH [28].

Psychological domain
The Beck Depression Inventory (BDI) is widely used to measure depression. The BDI consists of 21 items based on symptoms and attitudes that Beck considered common among patients with depression, but not among non-depressed individuals [29]. Statements were ranked to indicate the range of depression severity from neutral to maximal.

Physical domains
Exertion of work intensity was measured using the Borg Rating of Perceived Exertion (BRPE) scale.
Participants were asked to self-rate their exertion of work intensity on a scale between 6 and 20 [30].
For cervical range of motion (ROM), the dominant painful contralateral cervical side-bending and dominant painful ipsilateral cervical rotation ROM were measured using an iPhone with Clinometer and Compass applications [31]. Using a belt strap, participants were blocked from performing trunk and shoulder movements during measurements of cervical-lateral bending and rotation movement. The measurements of cervical ROM were made for the total range (i.e., difference between initial and nal measures). The mean value of three trials was calculated and used for the main analyses. handheld dynamometer force sensor was applied to the distal one-third of the participant's forearm, and force toward the oor was applied by the examiner. GHER was measured in the side-lying position with the shoulder exed and internally rotated to 90°, and the elbow exed to 90°. Then the dynamometer was applied to the distal one-third of the participant's radial forearm, and force toward the oor was applied by the examiner.
For posture analyses, forward head posture, rounded shoulder angle, shoulder slope angle, and scapular downward rotation ratio were measured using kinematic analyses of photographs using ImageJ software (National Institutes of Health, Bethesda, MD, USA). Forward head posture was quanti ed by the craniovertebral angle (angle between the horizontal line passing through C7 and a line extending from the tragus of the ear to C7). The rounded shoulder angle was quanti ed using the angle (θ) between two lines (from a horizontal line in the medial roots of the scapula to the acromion, and from the root of the scapula to the acromion). The angle θ, composed of the two distances, is one apex of a right-angled triangle. Therefore, 90 -θ was de ned as the rounded shoulder angle. Shoulder slope angle was quanti ed by the angle between two lines (a horizontal line with the acromion and a line between the spinous process of the seventh cervical vertebrae and acromion). Scapular downward rotation ratio was quanti ed by the ratio between two lines (a vertical line from the center to the root of the scapula, and a vertical line from the center to the scapular inferior angle).

Procedures
The present study was performed from March 2016 to November 2016. Participants were assessed at the work conditioning center in a theme park. Variables were measured in the following order: psychological and physical domains (posture, ROM, and strength). Participants were asked to complete a questionnaire (age, sex, BDI, and BRPE scale) and then were photographed to measure posture. PPS, two ROMs (cervical lateral-bending and rotation), and four muscle strengths (SA, LT, biceps, and GHER) were measured in that order.

Statistical Analyses
The Kolmogorov-Smirnov Z-test was used to assess the assumption of distribution normality. Demographic characteristics are shown as means. Independent t-tests were used to compare psychological and physical domains between participants with PPH and participants without PPH, and to identify signi cant predictors for cut-off in each sex, because of sex differences in demographic and clinical features. Variables with signi cant differences between participants according to PPH status in men and women were selected. For each sex, variables associated with PPH were selected from univariate analyses. Finally, for each sex, forward multivariate logistic regression was used to test our a priori hypothesis among selected variables that a combination of psychosocial and physical factors contributed to the risk for PPH. The analyses were adjusted for previously established covariates of age and body mass index. Goodness-of-t was calculated using the Hosmer-Lemeshow test. Statistical analyses were conducted using SPSS Statistics (ver. 18.0; IBM Corp., Armonk, NY, USA) and the signi cance level was set at p < 0.05.

Results
Comparisons of psychological and physical domains in men according to PPH status Table 2 shows comparisons of psychological and physical domains between men with PPH and men without PPH. There were no signi cant differences in BDI, BRPE scale, forward head posture, rounded shoulder angle, shoulder slope angle, or scapular downward rotation ratio between men with PPH and men without PPH. For cervical ROM, the dominant painful contralateral cervical side-bending (p = 0.02) and dominant painful ipsilateral cervical rotation ROM (p = 0.01) were signi cantly greater in men without PPH than in men with PPH. For muscle strengths, SA (p = 0.005), LT (p = 0.002), biceps (p < 0.001), and GHER strength (p = 0.001) were signi cantly greater in men without PPH than in men with PPH. Comparisons of psychological and physical domains in women according to PPH status Table 3 shows comparisons of psychological and physical domains between women with PPH and women without PPH. Most variables were not signi cantly different, but dominant painful ipsilateral cervical rotation ROM (p = 0.033) was signi cantly greater in women without PPH than in women with PPH.  The results of adjusted multivariate analyses of predictors of PPH of UT in men and women are shown in Table 5   Discussion PPS can result from impairments at multiple levels throughout the neuromuscular system [32]. There is increasing evidence that changes in pain processing may enhance sensitivity to noxious stimuli among individuals with chronic pain, compared to pain-free controls [14,33]. The present study investigated whether physical and psychological domains were related to PPH of the UT in each sex among FWs with nonspeci c neck/shoulder MP, because biological differences have been suggested to cause sex differences in pain perception [34][35][36]. LT strength and dominant painful ipsilateral cervical rotation ROM were characterized as risk factors for PPH of the UT in male and female FWs with nonspeci c neck/shoulder MP in adjusted multivariate analyses. Although our interpretations are limited because of the cross-sectional study design, the LT strength and dominant painful ipsilateral cervical rotation ROM identi ed in the present study could be useful for establishing guidelines for the prevention and management of PPH in FWs with nonspeci c neck/shoulder MP.
Concerning LT strength as risk factor for PPH, scapulothoracic muscle imbalances could be cause of impaired biomechanics, postural adaptations, and neck/shoulder pain [37,38]. These imbalances may occur when the UT becomes tight and the LT becomes weak [39,40]. Conversely, LT weakness could result in UT overload because of poor scapular mechanics (e.g., increasing scapular elevation and decreasing scapular upward rotation and posterior tilting) [37,38] and weakly synergistic acceleration of UT overactivation (e.g., involving the SA and LT) [41]. LT strength was signi cantly different between the ipsilateral (mean ± standard deviation (SD): 21.8 ± 10.0 N) and contralateral sides (mean ± SD: 25.7 ± 11.5 N) in individuals with unilateral neck and shoulder pain [40]. In the current study, LT strength in male FWs with PPH was 31.61 ± 15.58% normalized by body weight. Before LT strength was divided by body weight, LT strength was 23.50 N, which was similar to the results of a previous study involving individuals with neck/shoulder pain [40]. However, Shahidi et al. investigated physical risk factors for chronic neck pain [32]. They found that LT strength was not a risk factor, using a multivariate prediction model that involved cervical active ROM, cervical muscle strength and endurance, and scapular muscle strength. Although this explanation is limited by the cross-sectional study design, LT could be linked to PPH of the UT and could potentially be weaker in terms of PPH of the UT. The process may function in an inverse manner.
Cervical mobility as a risk factor for neck/shoulder pain has been suggested in prospective studies of other populations, but the results have been con icting. Reduced cervical exion mobility was more likely to cause neck/shoulder pain in laundry workers (risk ratio: 3.1; 95% CI: 1.2-8.3) [42] and increased cervical exion-extension mobility was protective against neck/shoulder pain in o ce workers (hazard ratio: 0.97; 95% CI: 0.94-0.99) [43]. With respect to dominant painful ipsilateral cervical rotation ROM as a risk factor for PPH, cervical rotation ROM is related to pain intensity in patients with chronic neck/shoulder pain [44,45]. Moreover, patients with nonspeci c neck/shoulder pain show less cervical rotation ROM, compared to asymptomatic controls [46,47]. Reduced extensibility of upper quadrant neural structures evaluated by the median nerve tension test has been related to decreasing UT length [48]. Furthermore, the presence of PPH of the UT was associated with cervical intervertebral joint dysfunctions [49]. Although interpretations are restricted by the cross-sectional study design, dominant painful side ipsilateral cervical rotation ROM could be linked to PPH of the UT and could potentially cause shortness involving PPH of the UT, or the process could function in an inverse manner. UT length affects ipsilateral cervical rotation ROM and contralateral cervical side-bending ROM because of the muscle attachment locations [50]. Thus, tenderness or PPH of the UT could affect the restriction of cervical ipsilateral rotation ROM. Conversely, reduced UT length could affect PPH by scapular dyskinesis (e.g., scapular elevation during arm lifting) [50]. UT shortness and scapular dyskinesis could generate reduced activity of the SA and/or LT, as well as enhanced activity of the UT, resulting in UT overactivation [51,52]. The experimental protocol was established according to the ethical guidelines of the Helsinki Declaration.
The study protocol was approved by the Yonsei University Mirae Campus Institutional Review Board (certi cation number: #1041849-201603-BM-005-02). The participants were then asked to sign a written informed consent. The informed consent was made in two identical copies that the participants could retain one.

Consent for publication
Not applicable.

Availability of data and materials
The datasets analyzed during the current study are available from the corresponding author on reasonable request.

Competing interest
The authors declare that they have no potential con icts of interest with respect to the research, authorship, and publication of this article. The results are presented clearly, honestly, and without fabrication, falsi cation, or inappropriate data manipulation.

Funding
Yonsei University Research Fund (grant numbers: 2020-52-0016) provided funding for this study.
Author contributions UJH contributes to conceptualization, project administration, writing-original draft, data curation, investigation, methodology and formal analysis in the present study. OYK contributes to conceptualization, funding acquisition, project administration and writing-original draft in the current study. All authors reviewed the manuscript.