Abstract

Background

Chronic fatigue syndrome (CFS) is a debilitating disorder characterised by prolonged medically unexplained, disabling fatigue, with concomitant unrefreshing sleep, as well as other constitutional and neurocognitive symptoms [1]. Some of these core symptoms are shared by patients with major depressive disorder (MDD), notably fatigue, unrefreshing sleep, and neurocognitive difficulties [2, 3]. Both CFS and MDD are prevalent conditions [4, 5], yet despite international research efforts, their pathophysiological basis remains obscure, in part reflecting heterogeneity within the syndromal diagnoses [6, 7]. The disease burden and costs associated with these disorders are substantial both for the individual and society [8, 9], yet available management strategies for both conditions have only partial efficacy [10–12].

Autonomic nervous system (ANS) dysfunction is a common feature of a wide range of medical and neuropsychiatric conditions including CFS and MDD [13–15]. The ANS comprises sympathetic and parasympathetic divisions. Broadly, the sympathetic division is concerned with energy-demanding processes during times of activity and stress (e.g. the ‘fight-or-flight response’), whereas the parasympathetic division (notably via action of the vagus nerve) is associated with vegetative and restorative functions. These two divisions operate together to maintain homeostasis. The heart is dually innervated by both the excitatory sympathetic and inhibitory parasympathetic (vagal) divisions, resulting in constant flux of the interval between heartbeats. This variation in inter-beat interval is referred to as heart rate variability (HRV) and is widely regarded as a sensitive and reliable non-invasive method of assessing autonomic function [16–18]. Optimal autonomic function is characterised by flexible adaptation to environmental challenges and a marked increase in vagal activity during rest (reflected by high HRV), whereas low HRV and a shift towards sympathetic dominance reflect a vigilant, defensive physiological state lacking dynamic flexibility, suggestive of a system under stress [19].

There is a growing body of literature linking constitutionally reduced HRV with sleep difficulties in neuropsychiatric conditions including CFS and MDD [13, 20, 21]. Specifically, low HRV (which persists even during sleep) correlates strongly with the core symptoms of unrefreshing sleep, fatigue, and cognitive impairment in CFS [14, 15]. Similarly, somatic symptoms of depression such as fatigue and sleeping difficulties are more closely associated with lowered HRV than other symptoms of depression [22, 23]. In light of the evidence supporting a role for reduced HRV in the manifestation of debilitating symptoms, therapeutic interventions targeting autonomic imbalance may be beneficial to patients affected by these and potentially other fatigue-related conditions.

In the absence of curative treatments for CFS and MDD, interventions are aimed at symptom management and functional improvements [24–28]. These interventions include variants of physical activity programs such as graded exercise therapy (GET) and cognitive-behavioural therapy (CBT) [10, 29–31]. In addition to these therapies, behavioural interventions including relaxation and mindfulness-based approaches [32–34] are increasingly utilised, and more recently HRV biofeedback [35, 36] has been explored in CFS and MDD. The beneficial effects of such techniques, including improvements in autonomic functioning, are thought to be mediated via their impact on neural circuits involved in self-regulation and adaptation [19, 37, 38].

Despite their widespread use, these psycho-behavioural interventions typically show only small to moderate effect sizes, with average response rates rarely exceeding 30% [10, 12, 25, 31, 33, 39]. Drill-down analyses have revealed subgroups of patients who do not respond or who respond unfavourably to such interventions [40, 41]. Although it is recognised that individuals can differ considerably in their response to both pharmaceutical and psycho-behavioural therapies [11, 40, 42], individual variation in response to common therapies in fatiguing illnesses has not been examined in any depth.

We recently reported the first detailed study of individual differences in ANS responses to a variety of relaxation methods in healthy individuals [43]. We found that engagement with four different methods (soothing white noise, classical music, paced breathing, and guided relaxation) elicited significant differences in vagal activity (HRV, continuously monitored). Close inspection of the results revealed diversity in individual response patterns across the four relaxation methods. Indeed, the method that elicited the greatest HRV response overall varied significantly between participants. Moreover, there was no close association between participants’ subjective evaluation of the relaxation methods and the method that elicited their optimal HRV response. Thus, to maximise health outcomes, it could be beneficial to pre-test participants in order to determine their optimal method, rather than relying solely on subjective responses or, as is common practice, employing a ‘one-size-fits-all’ approach.

Methods/design

Study design

This study is a two-parallel group RCT comparing a personalised relaxation intervention to a sleep- and symptom-monitoring control condition. All assessments are conducted at the University of New South Wales Sydney (UNSW), Australia. The trial has been registered on the Australian and New Zealand Clinical Trials Registry (ANZCTR, ACTRN12616001671459). All dimensions of the study protocol have been described adhering to the Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) checklist (Additional file 1; Fig. 1).

Open in a separate window

Fig. 1

Schedule of enrolment, interventions, and assessments (Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) diagram)

Data collection

All consenting participants complete two face-to-face assessments, 4 weeks apart, and a further set of questionnaires completed online 8 weeks after the initial assessment (Fig ​(Fig2).2). At the first assessment (lasting approximately 90 min), participants’ demographic, health, and symptom information are documented. Autonomic activity at rest and in response to each of the relaxation methods (see below) is recorded. Results from this initial testing determine which method participants randomised to the intervention group will practise daily during the 4-week intervention (i.e. the method which induced the greatest increase in HRV for that participant). Before leaving, all participants are fitted with a non-invasive ambulatory bioharness (Equivital EQ-02, Hidalgo Ltd., Cambridgeshire, UK) for 24-h monitoring and issued a sleep/activity/symptom log to complete over the next 4 weeks (see subsequent paragraphs). The second assessment lasts approximately 45 min and consists of a briefer version of the initial assessments (i.e. assessment of health, symptoms, and autonomic activity). Those in the control arm will be offered the personalised intervention at this assessment. The third assessment consists of online self-report measures, including questionnaires about the participant’s sleep, symptoms, and current frequency of relaxation practice. This will allow us to assess possible enduring benefits of the intervention in the active group, and to determine whether subjective improvements are evident in the monitoring group who will have practised their assigned intervention for 4 weeks. Participants will be emailed a personalised link to the online questionnaire. An automated reminder email will be sent after 1 week and again after 2 weeks, if the participant has not completed the questionnaire. The online nature of this assessment will reduce inconvenience for the participants and hence minimise attrition.

Open in a separate window

Fig. 2

Flow diagram of participant enrolment, assessment, and analysis. *Control participants were offered their personalised relaxation intervention

Ethics and dissemination

Potential participants are informed that their decision of whether or not to participate will not affect their relationship with the UNSW Fatigue Clinic or the BDI. All participants provide written informed consent before commencing any assessments and are free to withdraw at any time without consequence (see Additional file 2). Participants are given a signed consent form to take home which also includes information about the study and the contact details of a trained clinician, and a telephone helpline beyondblue. Participants can call these numbers in the unlikely event that they become distressed from the study and require support from someone not involved in any direct assessments. Participants are encouraged to respond in an open and honest manner to all questionnaires and are reassured that their participation and the responses they provide will remain confidential and have no bearing on their involvement in the study or on their current or future treatment.

All response data are entered into a secured computer database. Only research staff directly involved with the project has access to the physical and electronic data. Upon enrolment, participants are assigned a unique study identification number which is used across all study materials (questionnaires and computer files), allowing data to be linked during analysis without revealing the identity of the individual participants. To maintain anonymity of participants, only de-identified data, presented as group means and differences, are used in all forms of dissemination. In line with National Health and Medical Research Council (NHMRC) recommendations, all data will be kept for a minimum of 15 years. At time of disposal, hard-copy data will be shredded, and electronic media will be erased and re-formatted.

The findings from this study will be disseminated through international peer-reviewed journals and presentations at relevant national and international conferences. Participants who indicated on their consent form an interest in the overall results of the study will be sent (via email or post) a lay summary of the findings. If the results of the trial point towards beneficial effects of the intervention, the online links to the relaxation methods will be distributed to relevant stakeholders such as the UNSW Fatigue Clinic and the BDI.

Discussion

Patients with CFS or MDD frequently experience unrefreshing sleep and debilitating fatigue, which have negative consequences for daytime functioning [1, 3]. Although the pathophysiology of these conditions is not well understood, ANS dysfunction is evident in both patient groups and may influence symptoms [13, 55]. Current management strategies, including CBT, exercise programs, and relaxation methods, are only moderately successful and do not always deliver beneficial outcomes [10, 12, 25, 31, 33, 39]. Abandoning a ‘one-size-fits-all’ approach and moving towards personalised treatment options may offer another way forward.

To the best of our knowledge, this is the first RCT to prescribe a personalised intervention based on the participant’s maximal autonomic relaxation response. We propose that using such an approach will lead to more substantial improvements in sleep quality, reduced daytime fatigue and other symptoms, and enhanced functional status. The inclusion of two patient groups with fatiguing illness due to CFS or MDD will provide insights into generalisability of the potential health benefits resulting from personalised relaxation intervention to other relevant patient populations, thus maximising translational utility.

Trial status

The trial was registered on 5 December 2016 (ACTRN12616001671459). Study recruitment commenced in April 2017 and will continue for 2 years or until the required sample size is reached.

Additional files

Acknowledgements

Abbreviations

Authors’ contributions

All authors (CM, EC, GP, AL, GL, JB, and UVC) contributed to the development and design of this study. CM, EC, and GL collected preliminary data. EC provided statistical advice. CM drafted the manuscript. All authors revised the manuscript critically at each stage and have read and approved the final manuscript.

Notes

Footnotes

Contributor Information

Claire L. Macnamara, Email: moc.liamg@aramancamlerialc.

Erin Cvejic, Email: ua.ude.yendys@cijevc.nire.

Gordon B. Parker, Email: ua.ude.wsnu@rekrap.g.

Andrew R. Lloyd, Email: ua.ude.wsnu@dyoll.a.

Gina Lee, Email: moc.kooltuo@eenijeah.

Jessica E. Beilharz, Email: ua.ude.wsnu@zrahlieb.j.

Ute Vollmer-Conna, Phone: +61 2 9385 2945, Email: ua.ude.wsnu@etu.

References