Involvement of Cervical Muscle Lesions and Autonomic Nervous System in Myalgic Encephalomyelitis / Chronic Fatigue Syndrome (ME/CFS)

Background: Patients with myalgic encephalomyelitis / chronic fatigue syndrome (ME/CFS) sometimes coincide with stiffness of cervical muscles. This study examined the effect of local modulation of the cervical muscles on ME/CFS and the underlying mechanism. Methods: In total, 1,226 inpatients with ME/CFS who were resistant to outpatient care were enrolled in this study for 11 years. All patients underwent daily physical therapies to the cervical muscles during hospitalization. Self-rated records documenting the presence and absence of ME/CFS and the representative eight disorders that frequently accompany it at admission and discharge were compared. The pupil diameter was also measured to examine involvement of autonomic nervous system function. Results: The recovery rate of ME/CFS after local therapy was 55.5%, and did not differ signicantly by sex, age strata, and hospitalization period. The recovery rates of the eight disorders were variable (36.6 (cid:0) 86.9%); however, those of ME/CFS in the disorder subpopulations were similar (52.3 (cid:0) 55.8%). The recovery rates of all disorders showed strong associations with that of ME/CFS (p<0.001). The pupil diameter was decreased at discharge, and the change was signicantly higher in the ME/CFS-recovered patients than ME/CFS-unrecovered patients in the total population and the subpopulations stratied by sex, age, and hospitalization period. Conclusions: Local therapy to the cervical muscles led to recovery in more than half of patients with ME/CFS, at least partly through amelioration of the autonomic nervous system. There may be a causal relationship between recoveries of ME/CFS and these related whole-body disorders.


Background
Myalgic encephalomyelitis / chronic fatigue syndrome (ME/CFS) is a serious, chronic, and complex disease that occasionally affects the lives of patients due to debilitating fatigue [1][2][3][4]. It is accompanied by various symptoms throughout the body, such as headache, cervical stiffness, vertigo, cardiovascular and gastrointestinal disorders, fever of unknown etiology, and psychological disorders. The high prevalence and low employment rates of patients with ME/CFS impose an enormous burden on society. The Institute of Medicine of the National Academy of Sciences reported that ME/CFS affects up to an estimated 2.5 million people in the United States and generates direct and indirect expenses of approximately $17-$24 billion annually [5].
Since ME/CFS is a heterogeneous condition with a complex and multifactorial etiology, reaching a conclusive diagnosis using the current methods is di cult.
Although several studies have suggested the involvement of abnormal widespread metabolites [6][7][8], infection and neurological disorders [9], calcium ion channels [10], or anaerobic thresholds [11,12], the pathogenic mechanism of ME/CFS remains unclear. As such, patients are diagnosed through the exclusion of other conditions that could be responsible for the subjective symptoms with abnormalities across many domains [13. 14]. Thus, treatment remains symptom-based, multidimensional, and tailored to the needs of the individual patient [15,16].
The symptoms of ME/CFS were recently reported to improve by the surgical treatment of cervical spine [17]. Patients diagnosed with ME/CFS occasionally have a history of trauma to the neck, such as whiplash-associated disorders (WAD) [18]. In our clinical experience, we have observed a potential trend of inde nite symptoms throughout the body including ME/CFS occasionally coinciding with stiffness or tenderness of the cervical muscles and proposed a new medical concept called "cervical neuro-muscular syndrome" [19]. For the treatment of the inde nite symptoms, we have tried local modulation of the cervical stiffness. Among physical therapies, low-frequency electrical stimulation [20,21] and far-infrared irradiation [22] are reportedly effective at recovering muscle stiffness of the neck and shoulders. In fact, our previous study of patients with WAD showed that the combined application of these two physical therapies to the cervical muscles recovered not only local symptoms in the neck and shoulder, but also inde nite symptoms in the whole body [23]. Furthermore, our recent study of 1,863 patients with inde nite whole-body symptoms showed that therapies applied to the cervical muscles signi cantly improved signs and symptoms such as headache, cervical pain or stiffness, vertigo or dizziness, palpitation, dazzling, nausea or stomachache, fever of unknown etiology, and depression [24].
We propose autonomic nervous system involvement as the underlying mechanism. This system, which regulates the unconscious actions of the body via the sympathetic and parasympathetic nerves, reportedly plays a role in myalgic disorders such as bromyalgia and low back pain [25,26]. In contrast to the sympathetic nervous system's excitatory role under a stressful situation, the parasympathetic nervous system oversees resting, recovery from stress, and maintenance of homeostasis. Several nuclei of the hypothalamus generate coordinated patterns of responses of the two systems to internal or social stressors [27]. To evaluate autonomic nervous system function, the pupil light re ex is known to be a representative indicator; the constrictor muscle of the pupil decreases its diameter under control of the ciliary ganglion, which is activated and innervated by a preganglionic autonomic nerve ber [28][29][30]. In fact, the preliminary pupil light re ex test performed in a subpopulation of our recent study suggested possible autonomic nervous system dysfunction in the cervical muscles of patients with whole-body disorders [24].
The present study initially examined the effect of the two physical therapies to the cervical muscles in 1,226 patients with ME/CFS during hospitalization and examined the relationships between ME/CFS recovery and representative eight whole-body disorders that are frequently accompanied with ME/CFS. To learn the possible involvement of autonomic nervous system as an underlying mechanism, we further compared the changes of the pupil diameters during hospitalization between patients who did versus those without ME/CFS recovery.

Methods
Among patients who visited our institutions between May 2006 and May 2017 and were diagnosed with ME/CFS according to Fukuda's de nition [13], we enrolled 1,363 patients who were resistant to outpatient care and required hospitalization. Outpatient care was variable and included pharmacological and behavioral strategies but did not include the application of physical therapies to the cervical muscles. In principle, hospitalization was decided by consent between patients and physicians independent of the severity of ME/CFS. The main reason for hospitalization was persistent symptoms that required more intensive treatments as well as detailed examinations of other organs. Discharge was also decided by consent between patients and physicians independent of the ME/CFS recovery by the de nition above [13], mainly due to considerable improvement of symptoms of ME/CFS or related whole-body disorders.
Patients who were hospitalized for 5-120 days were enrolled.

Intervention
All patients underwent low-frequency electrical stimulation and far-infrared irradiation applied to the cervical muscles for 15 minutes two or three times daily throughout the hospitalization period. No other treatments including medication, injection, external xation, or cervical traction were performed. A combination of silver spike point (SSP; Nihon Medix, Chiba, Japan) and pain topra (LCF-30; Celcom, Inc., Fukuoka, Japan) was used for the low-frequency electrical stimulation, while a CERAPIA 3300 (Nihon Medix, Chiba, Japan) was used for the far-infrared ray irradiation. For all participants, the self-rated records on the medical interview sheets documenting the presence or absence of the representative eight disorders that frequently accompany ME/CFS [1][2][3][4]; including headache, cervical pain or stiffness, vertigo or dizziness, palpitation, dazzling, nausea or stomachache, fever of unknown etiology, and depression, were collected at admission and discharge. Pupil diameter was also measured using a binocular infrared pupilometer (Iriscoder Dual C10641; Hamamatsu Photonics, Shizuoka, Japan) at admission and discharge, and each patient provided informed consent.

Statistical analysis
Statistical analyses were performed using SPSS 16.0J for Windows. P values less than 0.05 were considered statistically signi cant; all reported p values were two-sided. As the sample size (n = 1,226) was su cient, the central limit theorem could be applied to con rm that the data were normally distributed and that violation of the normality assumption would not cause major problems [31]. Hence, the paired Student's t-test was used to examine the difference in the number of symptoms at admission versus discharge. The difference in the number of patients with each symptom between admission and discharge was evaluated using the chi-square test. The unpaired t-test was used to compare means between the recovered and unrecovered ME/CFS groups. In the univariate and multivariate logistic regression analyses, all variables were force entered into the multivariate model. Forward stepwise multivariate logistic regression analyses were also performed. The best model was selected based on likelihood ratio tests.

Results
Flow and backgrounds of participants Figure 1 shows a owchart of the patient enrollment process of the present study. A total of 1,363 patients who were diagnosed with ME/CFS according to the de nition above [13] and hospitalized in our institutions were initially enrolled in this study. Of them, 137 were excluded after enrollment, including 42 who were discharged from the hospital after less than 5 days, 14 who were hospitalized for more than 120 days, 59 who were diagnosed with speci c diseases in other organs after admission (one of whom died during hospitalization), seven who were transferred to other hospitals for treatment of speci c diseases, ve who refused to undergo pupil diameter measurement, and 10 who discharged themselves from the hospital based on their own judgement with unknown reasons. After the removal of these patients from the study population, 1,226 completed the study protocol. Among these patients, 680 (55.5%) were diagnosed as having recovered from ME/CFS according to the de nition above [13] at discharge, while 546 (44.5%) remained unrecovered. The eight representative disorders accompanying ME/CFS and the pupil diameters were assessed and compared between the ME/CFS-recovered and -unrecovered groups. Table 1 shows the baseline characteristics of the 1,226 participants (448 men, 778 women) with a mean age of 46.4 ± 16.3 years (mean ± standard deviation) and a mean hospitalization period of 62.5 ± 26.4 days. The recovery rate was not signi cantly altered by sex (men versus women), age strata (10-49 versus 50-89 years), or hospitalization period (5-60 versus 61-120 days) (p > 0.05). Relationship between ME/CFS recovery and eight related disorders Among the representative eight disorders accompanying ME/CFS [1][2][3][4], more than 70% of ME/CFS patients reported headache, cervical pain or stiffness, palpitation, dazzling, fever of unknown etiology, and depression; while 43.1% for vertigo or dizziness and 56.7% for nausea or stomachache ( Table 2). The recovery rates of these disorders in the total population after physical therapies during hospitalization were variable: more than 70% in patients with vertigo or dizziness, nausea or stomachache, and depression; 50-70% in patients with cervical pain or stiffness, palpitation, dazzling, and fever of unknown etiology; and 36.6% in patients with headache. However, the recovery rates of ME/CFS among patients with the eight disorders were similar (52.3-55.8%) to that of the total population (55.5%). Furthermore, the chi-square test ( Table 2) and logistic regression analyses (Table 3) between the recovery (versus non-recovery) of these disorders clearly showed a strong association with ME/CFS recovery in all disorders (p < 0.001). Among the disorders, the recovery from depression was most strongly associated with ME/CFS recovery (odds ratio, 13.70; Table 3). We further examined the possible involvement of autonomic nervous system function by comparing pupil diameters of admission and discharge (D-A) as well as the change ratio adjusted by the diameter at admission ([D-A]/A) ( Table 4). In the total population, both change in pupil diameter (D-A=-0.046 ± 0.633 mm; mean ± standard deviation) and change ratio ([D-A]/A=-0.002 ± 0.123) decreased during hospitalization, suggesting improved autonomic nervous system function by the physical therapies. These decreases were strongly seen in ME/CFS recovered patients (D-A=-0.099 ± 0.700 mm, [D-A]/A= -0.011 ± 0.134); however, there were no decreases, but rather increases, in the unrecovered patients (D-A = 0.020 ± 0.532 mm, [D-A]/A = 0.009 ± 0.107). A statistical analysis of the total population revealed a signi cant difference in the change in pupil diameter between the recovered and unrecovered groups (p = 0.001 for D-A, and p = 0.007 for [D-A]/A), suggesting an association between autonomic nervous system function and ME/CFS recovery. In subgroup analyses strati ed by sex (men versus women), age strata (10-49 versus 50-89 years), and hospitalization period (5-60 versus 61-120 days), the decreases were reproducible in all subgroups except (D-A)/A of the women and young generation (10-49 years) groups. Just like the total population analysis, pupil diameter was signi cantly decreased in the recovered versus unrecovered patients independent of sex and age strata (p < 0.05), although the difference was not signi cant only in the longer hospitalization period (61-120 days) subgroup.

Discussion
This study is the rst to show that the local therapies to the cervical muscles led to recovery in more than half of the patients with ME/CFS, suggesting that local modulation of the cervical muscles is a possible treatment for ME/CFS. Interestingly, however, among the representative eight disorders accompanying ME/CFS, the recovery rate of cervical pain or stiffness was lower (56.6%) than those of other systemic disorders such as vertigo or dizziness, nausea or stomachache, and depression (> 70%) ( Table 2). This nding suggests that there might be other mechanisms independent of cervical muscle modulation. In fact, both electrical stimulation and far-infrared irradiation reportedly cause nerve regeneration and repair directly [32,33]. Further studies using objective and quantitative measurements of muscle stiffness like the ultrasound elastography technique [34] may clarify the local therapy target.
As an underlying mechanism, we propose amelioration of the autonomic nervous system that passes through the cervical muscles by measuring pupil diameter. The canonical pathway that regulates the pupil diameter is such that the ganglion cell axons project to the Edinger-Westphal nucleus in the midbrain, where the preganglionic parasympathetic neuron ber in oculomotor nerve which is activated and commands the constrictor muscle of the pupil [30]. Although the oculomotor nerve does not pass through the cervical muscles, another non-canonical pathway via the afferent parasympathetic neuron ber in the vagus nerve, arising from the brainstem and extending through cervical muscles down to the thoracic and abdominal viscera [35], may be involved in the regulation of pupil diameter. Also, the sympathetic nerve reportedly enters the orbit via the divisions of the trigeminal nerve and a plexus of nerves surrounding the ophthalmic artery, a part of which commands the constrictor muscle of the pupil as a long ciliary nerve [36,37]. In addition, here we measured pupil diameter without using a light stimulation as the indicator of autonomic nervous function. However, the pupil light re ex under light stimulation is known to be more sensitive [28,29]. In fact, the pupil light re ex has been used to test patients with clinical signs of autonomic nerve dysfunction such as those with Parkinson's disease, Alzheimer's disease, and diabetes mellitus [38][39][40]. Although this study initially aimed to measure the pupil light re ex parameters under light stimulation, such as constriction rate and velocity, the institutional review board (IRB) did not allow us to deliver external stimulation that was not approved for the diagnosis or treatment of ME/CFS. In our recent study on patients with inde nite symptoms throughout the body, however, a subpopulation analysis of patients with dazzling exhibited a proportional improvement in the constriction rate and velocity with the pupil diameter without stimulation by local therapies [24]. Hence, we assume that pupil diameter measured without light stimulation could represent the pupil light re ex parameters with stimulation as an indicator of autonomic nervous system function.
The recovery (versus non-recovery) of all representative disorders accompanying ME/CFS showed a strong association with the recovery of ME/CFS (p < 0.001) ( Table 3). Among them, the recovery from depression was most strongly associated with ME/CFS recovery (odds ratio, 13.70; Table 3). However, whether depression is causes or consequences of ME/CFS remains unclari ed as previously reported [1,4]. It is possible that local modulation of the cervical muscles by the physical therapy initially improves psychological disorders such as depression through the cerebral limbic system, which then leads to recovery of the hypothalamus coordination of responses to sympathetic and parasympathetic nervous systems [27]. It is also noteworthy that the odds ratio of the recovery from dazzling, which is caused by the pupil light re ex dysfunction, to that from ME/CFS was the lowest (odds ratio, 2.755). Taken together, there may be a causal, although remains unclear, relationship between ME/CFS recovery and those of the related disorders through both autonomic nervous system dysfunction and other common pathologies.
Regarding the effect of the local therapies on ME/CFS, a limitation of this study is its lack of a control group of patients who did not undergo physical therapy to the cervical muscles. Although the participants did not receive any other interventions during the hospitalization, ME/CFS and the related disorders might possibly improve through resting alone. This project was initially planned as a prospective randomized controlled trial with two arms (with and without local therapy). However, neither the authorities of the Japanese Ministry of Health, Labour and Welfare nor the IRB granted permission to form a control group due to ethical problems, i.e. resting alone without interventions for hospitalized patients was not allowed. Notably, the patients enrolled in this study had already completed conventional outpatient care including resting at home for 12-91 days (mean, 49.3 days) before the hospitalization and were resistant to that care.
Therefore, it seems improbable that more than half of the patients with ME/CFS would recover via resting alone, even during hospitalization.

Conclusions
Local therapy to the cervical muscles led to recovery in more than half of patients with ME/CFS, at least partly through amelioration of the autonomic nervous system. There may be a causal relationship between recoveries of ME/CFS and these related whole-body disorders. These ndings suggest that the local modulation of cervical muscles may be a novel treatment for ME/CFS. However, the inpatient physical therapies performed in this study; two or three times daily for a mean of 62.5 days, is too costly for both health care providers and individuals. The development of more simple and feasible treatments is the next task. For the modulation of cervical muscles, we previously performed a prospective trial of the effects of an oral muscle-relaxant on ME/CFS. While the systemic modulation of muscle stiffness by the drug was somewhat effective at relieving local symptoms in the neck or shoulder, it had a minimal effect on whole-body disorders including ME/CFS (unpublished observation). Since we believe that local modulation of the cervical muscles independent of physical or medical intervention would effectively treat ME/CFS, we are now planning a clinical trial that will examine the effects of a topical muscle-relaxant poultice or ointment in patients with this disease.
Abbreviations ME/CFS: myalgic encephalomyelitis / chronic fatigue syndrome; FIR: far-infrared irradiation; IRB: Institutional review board; SSP: silver spike point; whiplashassociated disorders; WAD: whiplash-associated disorders Declarations Ethics approval and consent to participate The study was conducted with the approval of the institutional review board (IRB) of Tokyo Neurological Center and Matsui Hospital. All participants signed written consent to participate in the study. Written informed consent was obtained from a parent or guardian for participants under 16 years old. All rights of the patients were protected against any kind of disadvantage and individual matters.

Consent for publication
Not applicable. We are permitted to reproduce copyrighted materials.

Funding
There was no funding associated with this study. The patients were hospitalized for intensive treatment and detailed examination, but not for this study. The cost of treatment was borne partly by the patients and partly by the Japanese national insurance.

Competing interests
The authors declare no competing interests.

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

Authors' contributions
TM and KH conceived of the study idea. TM, KH, MI, YE, NS, and HK started the study. TM and HK obtained the ethical approval. TM, MI, YE, NS, SH, HF, MM, and HK acquired the data. TM, KH, and HK interpreted the data. TM and HK performed the statistical analysis. TM and HK wrote the initial manuscript. All authors revised the manuscript and approved the submitted version for publication.