Problems with Automatic Body Regulation in Chronic Fatigue Syndrome: Results from a Large US Clinical Study

Symptoms of autonomic dysfunction are common in infection-associated chronic conditions and illnesses (IACCI), including myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS) [1]. Autonomic dysfunction is among the proposed pathologies that may contribute to core symptoms of ME/CFS, which is a debilitating, long-term illness that affects multiple body systems. However, dysautonomia is not unique to ME/CFS. Dysautonomia is an overarching term that describes multiple symptoms and disorders involving the autonomic nervous system, including orthostatic intolerance (OI), heart rate dysregulation, gastrointestinal dysmotility, fatigue, bladder dysfunction, abnormal sweating, and others [2]. OI is included in the 2015 Institute of Medicine diagnostic criteria for ME/CFS [3] and is one of the most common presenting clinical conditions seen in those with autonomic dysfunction. The heterogeneity of ME/CFS limits the ability to identify a consistent pattern of autonomic dysfunction across affected individuals. To address this variability, researchers have proposed ME/CFS subtypes based on the presence or absence of specific autonomic abnormalities, such as postural orthostatic tachycardia syndrome (POTS). Autonomic function is typically evaluated through a range of methods—including heart rate variability analysis, tilt table testing, and self-reported questionnaires—although these tools vary in their application and results are not always strongly correlated across studies [3].

The Composite Autonomic Symptom Scale 31 (COMPASS-31), originally derived from the Autonomic Symptom Profile, is a relatively short but broad self-assessment instrument for autonomic symptoms and function in six domains covering several organ systems (orthostatic intolerance, pupillomotor, vasomotor, secretomotor, gastrointestinal, and bladder) that can be used as a clinical screening tool for dysautonomia [4]. The COMPASS-31 has been used, with varying degrees of validation, to screen for autonomic symptoms experienced by people with a variety of conditions, such as multiple sclerosis, Parkinsonism, compressive myelopathy, diabetes, and small fiber polyneuropathy [5,6,7,8,9,10,11].

Studies of OI in people with ME/CFS, including postural orthostatic intolerance syndrome (POTS), neurally mediated hypotension (NMH), or orthostatic hypotension (OH), using varied methods report a wide range (30–70%) in the prevalence of dysautonomia [3]. Research on dysautonomia overall in ME/CFS has most recently focused on activity-related data like heart rate variability and patient-reported outcome measures, such as the COMPASS-31 [12,13,14,15,16]. These studies have indicated that dysautonomia of varying kinds can be identified in people with ME/CFS but have not directly examined the association of dysautonomia with symptom burden and functioning. The present study aims to evaluate dysautonomia among a cohort of patients with ME/CFS seen in seven different US specialty clinics using clinical history, the COMPASS-31, and the Lean Test [17], as well as to evaluate whether the presence of any dysautonomia affects illness severity among people with ME/CFS.

Socio-demographic and clinical characteristics are summarized in Table 1. The majority of participants in both groups were female (ME/CFS: 69.8% vs. HC: 70.9%, p = 0.8048). Participants with ME/CFS were older than the HCs (51.4 vs. 47.0 years, p = 0.0012) and more likely to identify as White and non-Hispanic (ME/CFS vs. HC: White: 90.4% vs. 56.0%, p < 0.0001; non-Hispanic: 84.1% vs. 61.7%, p < 0.0001). ME/CFS participants were also more likely to have health insurance (96.3% vs. 87.2%, p = 0.0001) and reported a higher number of clinic office visits (5.1 vs. 0.7 visits, p < 0.0001). Conversely, ME/CFS participants were less likely to be employed (19.9% vs. 75.8%, p < 0.0001) or to be obese (11.3% vs. 26.2%, p < 0.0001). No statistically significant differences between ME/CFS and HC individuals were observed in terms of marital status or educational attainment. Among ME/CFS participants, the mean duration of illness was 18.4 ± 10.98 years, with 63.1% reporting a sudden onset of illness.

Table 2 highlights the autonomic symptoms by group. COMPASS-31 total scores, as well as domain-specific scores, were significantly higher in ME/CFS participants than in HCs (p < 0.0001). Effect sizes for group mean differences in six domain scores ranged from very large in the pupillomotor domain (d = 1.75) to moderate in bladder and vasomotor domains (d = 0.67). Additionally, ME/CFS participants demonstrated significantly higher OGS scores than HCs (8.3 vs. 0.6, p < 0.0001, Cohen’s d = 1.86).

According to medical records, autonomic-related symptoms or abnormalities were significantly higher in ME/CFS participants than HCs (p < 0.01), with the exception of “Other Dysautonomia.” Large Cohen’s h effect sizes were observed for the proportion of participants having autonomic-related symptoms, including dizziness or vertigo (42.6% vs. 2.8%, h = 1.08), cold extremities (38.6% vs. 5.7%, h = 0.86), OI (33.9% vs. 0.7%, h = 1.07), irregular heartbeat or palpitations (29.9% vs. 2.8%, h = 0.82), and dry mouth (27.5% vs. 2.1%, h = 0.81). Moderate effect sizes were noted in other symptoms, such as constipation (28.2% vs. 2.8%, h = 0.78) and low blood pressure (22.5% vs. 3.5%, h = 0.61).

Among three assessment tools (COMPASS-31, medical history, and Lean Test), nearly all ME/CFS participants (97%) were identified as having one or more autonomic abnormalities. In contrast, only 38% of the healthy controls had autonomic abnormalities. Among 301 ME/CFS participants, 282 (93.7%) were identified by COMPASS-31 as having any of the six autonomic dysfunction risks: OI (66.8%), GI (66.2%), pupillomotor (57.3%), secretomotor (52.7%), bladder (29.8%), and vasomotor (27.8%) (Table 3). Notably, 181 (64.2%) participants identified by COMPASS-31 as having risk of dysfunction had no documented ongoing autonomic abnormalities in their medical history reviews. However, 22 of them were later confirmed through the Lean Test to have POTS or OH. Since OI was the only autonomic dysfunction assessed by all three tools, we examined their concordance in identification. Of 219 ME/CFS participants who had data from all three assessment tools, 174 had OI identified by at least one method; 60.3% (105) had OI identified by one tool only (80 by COMPASS-31, 17 by medical history, and 8 by the NASA Lean Test), and 39.7% (69) had OI identified by two or three tools (Figure 1).

Autonomic symptoms significantly contributed to both symptom burden and functional impairment in ME/CFS participants, as evidenced by associating these measures with the COMPASS-31 dysautonomia risk. These scores revealed notable differences between ME/CFS participants with and without a risk of dysautonomia (Table 3 and Table 4).

As expected, measures addressing similar concepts showed large effects (d >= 0.8): OGS scores demonstrated a substantial difference between those with OI risk and those without (d = 1.11, Table 3), while CDC-SI “sensitivity to light” scores differed significantly between those with pupillomotor risk and those without (d = 1.26, Table 4).

OI risk showed a significant effect on fatigue-related measures—[PROMIS Fatigue (∆ = 6.0, d = 0.81), SF-36 Vitality (∆ = 6.0, d = 0.66), CDC-SI “fatigue after exertion” (∆ = 3.5, d = 0.63)], sleep measures, [PROMIS Sleep Impairment (∆ = 5.4, d = 0.73)], and measures of function [SF-36 Physical Functioning (∆ = 6.6, d = 0.64) and Social Functioning (∆ = 6.2, d = 0.53)]. Pupillomotor risk showed a significant effect on pain-related measures [SF-36 Bodily Pain (∆ = 7.4, d = 0.73), PROMIS Pain Interference (∆ = 6.4, d = 0.71) and Pain Behavior (∆ = 4.3, d = 0.59), CDC-SI “muscle aches and pains” (∆ = 3.6, d = 0.70), and “joint pain” (∆ = 2.8, d = 0.56)], as well as measures of function [SF-36 general health (∆ = 5.6, d = 0.65) and SF-36 Physical Functioning (∆ = 6.3, d = 0.60)], cognitive measures [CDC-SI “concentration” (∆ = 3.3, d = 0.64), “memory problems” (∆ = 2.9, d = 0.59)], fatigue [PROMIS Fatigue (∆ = 4.7, d = 0.63)], and OGS (∆ = 2.5, d = 0.53). GI risk showed a significant effect on sleep-related measures [PROMIS Sleep Impairment (∆ = 4.4, d = 0.58), CDC-SI “sleep problems” (∆ = 2.7, d = 0.51) and “unrefreshing sleep” (∆ = 2.6, d = 0.50)]. After adjusting for confounding factors (age, sex, illness duration, and the number of ongoing comorbidities), the dysautonomia risks identified by COMPASS-31 still showed a statistically significant impact on ME/CFS-related symptoms and functioning.

We found that people with ME/CFS exhibited a higher burden of autonomic symptoms than healthy controls, as determined through both clinical evaluation (NASA Lean Test) and patient-reported measures (medical records and COMPASS-31). These findings align with previous studies documenting dysautonomia or autonomic dysfunction in people with ME/CFS [21,33,34,35]. However, most prior research has relied on a single assessment tool to evaluate autonomic dysfunction or its subtype, orthostatic intolerance (OI).

In contrast, our study used multiple assessment tools to reveal that nearly all ME/CFS participants (97%) had one or more autonomic abnormalities. Reviewing a patient’s medical history is a routine clinical practice that can serve as a valuable tool for monitoring both past and ongoing medical issues. However, we lack information regarding whether the dysautonomia recorded in medical records was confirmed through objective clinical tests, such as tilt table tests, cardiovascular autonomic reflex tests, or heart rate variability assessments. It is important to note that some objective tests can be costly and may require out-of-pocket payments, adding to the financial burden of ME/CFS patients, who already bear a substantial portion of their medical expenses personally [36]. To address this challenge, accessible screening tools like COMPASS-31, a shortened version of the 73-item COMPASS, could be highly beneficial. COMPASS has been proven to be a valid screening tool for autonomic dysfunction [33]. More affordable objective tests, such as the Lean Test, could also help reduce financial barriers. Our study demonstrated that the Lean Test is very useful in identifying additional ME/CFS patients with POTS or OH who were not previously documented in medical records for these conditions. Furthermore, additional presentations of OI can also be identified using the NASA Lean Test with capnography, as implemented at one of the MCAM study sites [21].

By leveraging available tools like the Lean Test, COMPASS-31, and patient medical history, clinicians can enhance the diagnosis of OI and other autonomic dysfunctions. The clinical heterogeneity of patients with ME/CFS and/or dysautonomia often leads to delays in diagnosis. Employing these tools could help reduce those delays, ensuring that patients receive timely support for symptom management.

People with ME/CFS who also had dysautonomia risks identified in COMPASS-31 exhibited an increased symptom burden and poorer functioning compared to those without risks, particularly in relation to OI, GI issues, and pupillomotor dysfunction. Specifically, the co-occurrence of the OI, pupillomotor, and GI risks had a substantial impact on orthostatic grading scale (OGS) scores, sensitivity to light, and diarrhea, respectively, as they measured similar concepts. Although research into the influence of autonomic dysfunction on ME/CFS severity remains limited, existing studies have reported its impact on worsening function, particularly when OI, secretomotor, and GI dysfunction co-occur with other conditions [19,37,38]. Our study further supports the impact of autonomic dysfunction on ME/CFS illness severity across multiple domains. We found that both OI and pupillomotor risks are significantly associated with increased fatigue and greater impairments in physical functioning among people with ME/CFS. Additionally, sleep-related impairment in people with ME/CFS appears to be further aggravated by both OI and GI risks, a relationship supported by prior research [39,40]. Circadian rhythm misalignment, recognized as a key factor in sleep disturbances [41,42], may contribute to these disruptions through mechanisms involving wakefulness-related neuropeptide orexin [42]. Furthermore, our findings indicate a significant impact of the pupillomotor risk on pain and cognitive impairment, while OI further intensified the post-extensional malaise (PEM). The association of autonomic dysfunction with pain, cognitive impairment, and PEM has not been extensively investigated in people with ME/CFS. These findings reinforce the notion that autonomic dysregulation may further contribute to the level of severity in fatigue, sleep disturbances, cognitive function, pain, PEM, and functional impairment in those with ME/CFS.

Utilizing available pharmacologic and non-pharmacologic treatment options for dysautonomia holds promise in alleviating the autonomic symptom burden in people with ME/CFS [43,44,45,46,47]. Additionally, our findings highlight the need for further research into clinically significant aspects of dysautonomia, particularly subgroup analyses, to refine treatment and management approaches [48].

Our study revealed consistent and robust differences in dysautonomia burden between patients with ME/CFS and healthy controls, as assessed using clinical history and multiple assessment tools. Dysautonomia abnormalities, as captured by the COMPASS-31 domains, were associated with greater illness burden and functional impairment. Our results highlight the potential value of recognizing and addressing dysautonomia symptoms in ME/CFS to reduce the illness severity. Future studies are necessary to confirm whether individualized treatment strategies can effectively improve the quality of life for this population.

The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention. The study group included the following: The Multi-Site Clinical Assessment of Myalgic Encephalomyelitis/Chronic Fatigue Syndrome (MCAM): Centers for Disease Control and Prevention/Division of High-Consequence Pathogens and Pathology/Chronic Viral Diseases Branch, Atlanta, Georgia: Elizabeth Unger (Principal Investigator (PI)), Jin-Mann Sally Lin (Co-PI), Monica Cornelius, Britany Helton, Yang Chen, Elizabeth Fall, Anindita Issa; Bateman Horne Center, Salt Lake City, Utah: Lucinda Bateman (PI), Jennifer Bland, Patricia Jeys, and Veronica Parkinson; Hunter-Hopkins Center, Charlotte, North Carolina: Charles Lapp (PI) and Wendy Springs; Institute for Neuro Immune Medicine, Miami, Florida: Nancy Klimas (PI), Elizabeth Balbin, Jeffry Cournoyer, Melissa Fernandez, Shuntae Parnell, and Precious Leaks-Gutierrez; Mount Sinai Beth Israel, New York, New York: Benjamin Natelson (PI), Michelle Blate, Sarah Khan, and Diana Vu; Open Medicine Clinic, Mountain View, California: Andreas Kogelnik (PI), Joan Danver, David Kaufman, Macy Pa, Catt Phan, and Sophia Taleghani; Richard N. Podell Medical, Summit, New Jersey: Richard N. Podell (PI), Trisha Fitzpatrick, and Beverly Licata; and Sierra Internal Medicine, Incline Village, Nevada: Daniel Peterson (PI), Elena Lascu, Gunnar Gottschalk, and Marco Maynard.

The following abbreviations are used in this manuscript:

The following supporting information can be downloaded at https://www.mdpi.com/article/10.3390/jcm14176269/s1↗: Table S1: Characteristic of participants with ME/CFS (n = 301)—overall functioning and symptom status at intake.

Conceptualization, A.I., J.-M.S.L., Y.C. and E.R.U.; data curation, J.-M.S.L., Y.C. and J.A.; formal analysis, J.-M.S.L. and Y.C.; funding acquisition, J.-M.S.L. and E.R.U.; investigation, A.I., J.-M.S.L., Y.C., B.H.N., C.W.L., R.N.P., A.M.K., N.G.K., D.L.P., L.B. and E.R.U.; project administration, J.-M.S.L.; resources, D.B., J.B. and E.R.U.; supervision, J.-M.S.L., B.H.N., C.W.L., R.N.P., A.M.K., N.G.K., D.L.P., L.B. and E.R.U.; visualization, J.-M.S.L. and Y.C.; writing—original draft, A.I., J.-M.S.L. and Y.C.; writing—review and editing, A.I., J.-M.S.L., Y.C., J.A., D.B., J.B., B.H.N., C.W.L., R.N.P., A.M.K., N.G.K., D.L.P., L.B. and E.R.U. All authors have read and agreed to the published version of the manuscript.

This study was conducted in accordance with the Declaration of Helsinki and approved by the Institutional Review Board (IRB) of OMI (Western IRB protocol number: 1130042, approved 11 March 2012); Nova Southeastern University (protocol number: 01301214, approved 13 April 2012; Mount Sinai School of Medicine (protocol number: 283-11, approved 24 January 2012); and the CDC (protocol number: #6286.0, approved 26 April 2012).

Informed consent was obtained from all participants.

Restrictions imposed by the data custodians prevent the datasets from being publicly available or provided by the authors. The program codes used in the current study are available from the corresponding author upon reasonable request. Researchers interested in accessing the datasets used in this study should email the CDC’s ME/CFS Program () and discuss the next steps for the data request. The ME/CFS program data review committee will grant access after the review and data use agreement are finalized. cfs@cdc.gov

Author Jacob Attell was employed by the company Booz Allen Hamilton, Inc. Author Andreas M. Kogelnik was employed by the company ProDx Health. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention.

This work was funded by the Centers for Disease Control and Prevention.