Exploring levels of TSH and FT4 in patients with chronic fatigue syndrome (CFS), fibromyalgia (FM) and healthy controls did not reveal any associations between fatigue score and level of thyroid hormones

Abstract

Aim

The diagnoses of Chronic Fatigue Syndrome (CFS) and Fibromyalgia (FM) are highly associated with fatigue and pain, respectively. Physiologically and clinically an effect of thyroid status on fatigue and pain is expected. There may be clinically relevant differences in thyroid hormone axes though within values of reference in both patients with normal thyroid hormones, or in patients with well-regulated thyroid disease. These potential differences are explored in this study.

Materials and methods

In the present study, female patients with CFS (n = 49) and FM (n = 58) as well as female healthy controls (n = 53) were included. We explored plasma levels of TSH and FT4 between the groups using Kruskall-Wallis, and the relation between fatigue score and levels of TSH and FT4 by means of Spearman’s rho.

Results

There were no group differences between CFS patients, FM patients, and healthy controls in levels of TSH and FT4.

Conclusion

As one might clinically and physiologically expect an association between thyroid function and fatigue, which may be associated with clinical disorders such as CFS and FM, we suggest future studies to examine the field further by exploring the influence of thyroid receptors and responses of the thyroid hormone cascade.

Keywords:

• CFS
• FM
• TSH
• FT4
• fatigue

1. Introduction

Chronic fatigue syndrome (CFS) and fibromyalgia (FM) are two syndromes with unknown etiology and pathogenesis. They are both characterized by prolonged fatigue and chronic pain. It has been hypothesized that both disorders share a similar pathophysiology regarding inflammation and central sensitivity [1,2], yet this has not been fully concluded. In addition to imposing suffering upon patients and their surroundings, they also incur a significant societal cost [3,4]. For example, a study from Stockholm reviewing sick leave reported fibromyalgia as the most common reason for long-term job impairment [5].

Further knowledge is important to enhance understanding of these conditions and thereby improve the process of sustainable prevention, diagnosis, treatment, and care for the patient population.

CFS is a clinical diagnosis of exclusion, where other somatic and psychiatric illnesses must be ruled out as the cause of symptoms. It is characterized by pronounced fatigue and reduced functioning. The condition must have persisted for six months or more. At least four additional symptoms must be present, such as problems concentrating, memory issues, sore throat, tender lymph nodes, muscle pain, joint pain, newly onset headaches, poor sleep quality, and post-activity exhaustion lasting more than 24 hours [4,6]. The prevalence of CFS has been estimated to vary from 0.03% to 2.52% according to the diagnostic criteria employed [7] and is between 1.46% and 2.52% when applying the Fukuda criteria [7].

Potential mechanisms discussed for CFS include inflammation [8], mitochondrial dysfunction [9], and autoimmunity [10]. As fatigue, diffuse muscle pain, and mood disturbances [6,11] are all shared common symptoms of hypothyroidism (HT) and CFS, exploring thyroid function in CFS is interesting. The potential effects of altered activity in thyroid hormone axes have been explored in several studies. From our group, we have previously reported altered thyroid levels in the acute phase of mental illness [12]. Regarding CFS, different studies have compared thyroxine stimulating hormone (TSH) and free thyroxine (FT4) between CFS patients and healthy controls with varying results. A study of adolescents aged 12 to 18 years observed significantly higher FT4 values and no difference concerning TSH in CFS patients compared to controls [13]. Another study [14] found significantly higher TSH values in patients with CFS, but no differences in FT4. Furthermore, two additional studies reported the absence of a statistically significant difference in TSH and FT4 between CFS and controls [15,16].

FM is a syndrome consisting of chronic widespread pain and multi-organ symptoms like fatigue, headaches, sleep disturbances, muscle stiffness, anxiety, and depression. The prevalence has been described to vary from 2% to 8% in the general population, with females being more frequently affected [17]. Diagnostic criteria consist of a high widespread pain score, pain in at least four out of five body regions, and duration over 3 months [18]. The update from 2016 excludes previous criteria that made diagnosing multimorbid patients with FM challenging.

While the pathophysiology behind the condition remains to be established, several theories have been published during the last few years. Central problems with pain processing, hormone resistance, chronic infections, and inflammation are some of the mechanisms discussed [8,17,19]. HT may present with fatigue and gastrointestinal (GI) discomfort as well as pain (e.g. muscle aches, tenderness, stiffness, and joint pain particularly in hands and knees) [20–22]. Thus, as discussed for CFS, altered thyroid function must be explored as a potential cause of symptoms in FM.

One study concluded with a lower resting metabolic rate (RMR) in patients with FM, but found no correlation between RMR and TSH or FT4 [23]. A metastudy from 2022 found that thyroid autoantibodies were more commonly present in the FM group than in controls [24]. In a trial from Germany in 1998, 15 FM patients were injected with thyroid-releasing hormone (TRH), and patients with FM showed a blunted response to TRH injections compared to healthy controls. Tests done before the trial also showed a difference in triiodothyronine (T3), where FM patients showed significantly lower levels. Their basal levels of TSH and FT4 were, however, similar between the groups before injections [25].

There are physiological indications that thyroid hormone axis function could be related to CFS and FM. However, findings are so far inconclusive, and no clinically relevant role is established. Thus, the field needs further exploration.

In the present study, we aim to compare levels of TSH and FT4 between FM, CFS and controls in women aged 18 to 60. Furthermore, we aim to explore potential associations between levels of TSH and FT4 with fatigue score.

Hypotheses

TSH and FT4 will differ between the groups CFS, FM and healthy controls.

TSH and FT4 will vary with fatigue scores independent of the diagnostic group.

TSH and FT4 will vary with subgroups of fatigue scores.

2. Materials and methods

2.1. Patient group

As previously reported (2020, 2021) [8], this study includes non-pregnant women aged 18 to 60 years. Participants were recruited from three groups: 49 patients with CFS, 58 with FM, and 53 healthy controls. Patients admitted to the Multidisciplinary Pain Centre at St. Olav’s Hospital who fulfilled the CFS or FM criteria were asked to participate in the study. All patients were assessed using the 1990 ACR-criteria [26] for patients with FM and the Fukuda criteria [5] for patients with CFS. The Fukuda criteria state that any other somatic or psychiatric disorder that may explain the symptoms will exclude CFS diagnosis. CFS diagnosis can be given if any hyper- or hypothyroidism is medically regulated and stabilised. Although the 1990 ACR-criteria does allow for some comorbidity alongside FM, we chose to exclude any comorbidity similar to those given for CFS and Fukuda criteria [5]. All patients were diagnosed by an expert team of medical doctors, psychologists, and physiotherapists.

2.2. Healthy controls

The control group consisted of 53 female participants aged 18 to 60 years, recruited via advertisements on websites among staff at St. Olav’s Hospital and the Norwegian University of Science and Technology (NTNU). Participants in this group underwent a comprehensive evaluation, including a questionnaire assessing symptoms related to CFS and FM and a structured medical history assessment to document their somatic health status.

2.3. Procedure

Patients diagnosed with CFS and FM received information letters from the hospital regarding the study. For patients with CFS, the letter was handed out either during, or after, their clinical appointment. For FM, the informational letter was delivered during their outpatient appointment. Subsequently, both groups were contacted regarding participation in the study, and if they expressed interest, appointments were scheduled via telephone.

The assessment included blood sample collection, interviews, and the completion of questionnaires. Questionnaires included were the Hospital Anxiety and Depression Scale (HADS) [27,28], Chalder Fatigue Scale [29,30], FM 2011 and 2016 criteria [18,31], and Brief Pain Inventory (BPI) [32,33]. Data collection for this study was conducted by the last author Nina Groven (NG) between March 2015 and December 2016, with the assessments being performed in a random order. The study received approval from the Regional Committee for Medical and Health Research Ethics (REK 2014/711) and written informed consent was obtained from all participants.

2.4. Chalder fatigue questionnaire

Chalder fatigue questionnaire consists of 13 items, each scored on a scale of 0–3, and is used to assess fatigue in patients and healthy controls in this study. Questions 12 and 13, which are added in the Norwegian translation [30] regarding the severity and duration of the symptoms, were excluded from the study due to them not being a part of the fatigue score of the original questionnaire [29].

Consequently, a fatigue score spanning from 0 to 33 is established, where a higher score implies more fatigue. These inquiries are often divided into two types of fatigue, leading us to categorize them into these subgroups: physical fatigue (questions 1–7), and mental fatigue (questions 8–11) [29,30]. This differentiation facilitated a more targeted examination of the relationship between TSH and FT4 levels and fatigue compared to the total fatigue score.

2.5. Interview

The interview was conducted by NG and was compromised of a structured, clinical interview in addition to measurements of height and weight. Medical history, both current and previous mental and somatic illnesses (including infections, immune disease and comorbidity), and duration of CFS or FM were recorded. Current medications, menstrual cycle/menopause status, and potential use of contraceptives were also registered. Physical activity over the past two weeks was assessed on a scale ranging from 1 to 4; 1 being bedridden and 4 training twice a week or more. As mentioned in sections 2.1. Patient group any current comorbidity would lead to exclusion from this study.

2.6. Blood samples and analysis

The blood samples were collected at St. Olav’s Hospital and sent directly to the accredited, clinical laboratory for analysis according to laboratory procedures. The patients were not given any restrictions (e.g. regarding fasting, medication, caffeine) before blood sampling. Blood was screened for inflammation and participants were excluded from the study if inflammation was evident as measured by CRP >10mg/L. Blood for TSH and FT4 was collected from plasma using lithium heparin tubes with gel and was processed in accordance to laboratory protocols. The reference levels from St. Olav’s Hospital laboratory were used for both TSH (0.5–4.0 mIU/L) and FT4 (12.2–19.6 pmol/L) for excluding participants with thyroid disorder.

Patients who used levothyroxine were excluded only when comparing thyroid hormone levels between the groups as seen in Table 1 and were otherwise included in the analysis.

Table 1. Descriptives of age, body mass index (BMI), fatigue and thyroid hormones FT4 and TSH for CFS, FM and controls.

	CFS				FM				Control
	(n = 43)				(n = 54)				(n = 53)
	Missing	M (SD)	Mdn	Range	Missing	M (SD)	Mdn	Range	Missing	M (SD)	Mdn	Range	p
	(n)			(min– max)	(n)			(min– max)	(n)			(min– max)
Age	0	33.6 (11.3)	35	18–60	0	41.6 (9.0)	42	22–57	0	39.4 (10.4)	39	23–59	.001
BMI	2	23.4 (3.1)	22.9	18.1–31.2	1	27.0 (5.6)	25.9	16.3–40.4	0	24.7 (4.0)	23.8	16.3–41.7	.002
Total fatigue score	0	25.2 (5.3)	26	12–33	1	22.7 (5.3)	23	13–33	1	10.3 (3.2)	10	3–22	<.001
Physical fatigue score	2	16.9 (3.4)	18	7–21	3	15.2 (3.4)	16	8–21	2	6.3 (2.2)	6	3–14	<.001
Mental fatigue score	0	8.3 (2.4)	8	4–12	1	7.4 (2.5)	7	3–12	2	4.1 (1.2)	4	0–8	<.001
TSH	0	1.66 (0.83)	1.52	0.42–4.97	1	1.52 (0.67)	1.5	0.01–3.34	0	1.66 (0.73)	1.54	0.35–4.00	.619
FT4	0	15.2 (1.8)	14.9	12.0–19.2	1	15.2 (2.4)	15.0	10.0–21.4	0	15.0 (1.9)	14.6	11.7–19.5	.809

Patients using levothyroxine (n = 10) were excluded from these analyses. Statistical analysis: Kruskal–Wallis test. Total fatigue score: Total score from the Chalder fatigue questionnare. Physical fatigue score: Based on question items 1–7 from the Chalder fatigue questionnare. Mental fatigue score: Based on question items 8–11 from the Chalder fatigue questionnare. BMI = body mass index. TSH = thyroid stimulating hormone. FT4 = thyroxine.

2.7. Statistical analysis

To perform the analysis Statistical Software Package (SPSS) version 29 was used. The data were not normally distributed, and the Kruskal–Wallis test was applied to compare levels of TSH and FT4 between groups. Mann–Withney U test was conducted for post-hoc analysis of pair-wise comparison when differences were found. To assess a possible correlation between the thyroid hormones and the degree of fatigue we used Spearman’s rho (ρ). These analyses included both the total fatigue score and the subgroups of fatigue scores. Significance levels were set to p < .05.

3. Results

In this study, there were 160 participants in total, consisting of 49 patients with CFS, 58 patients with FM, and 53 healthy controls. Descriptive data are presented in Table 1.

3.1. Level of thyroid hormones between the groups

As reported in Table 1, no significant differences in plasma concentration of TSH and FT4 are evident between the three groups. In this analysis, a subset of participants (CFS n = 6, FM n = 4, control n = 0; total n = 10) using levothyroxine has been excluded. When including the medicated patients, there were still no differences between the groups (data not shown).

3.2. Fatigue score, age, and BMI between the groups

Regarding fatigue, patients with CFS scored higher than FM (z = −2.345, p = .019) and controls (z = −8.173, p < .001), and FM scored higher than controls (z = −8.461, p < .001).

Regarding age, FM patients were older than controls (z = −1.104, p = .270) and CFS (z = −3.703, p <.001). The patients with CFS were significantly younger than the control group (z = −2.540, p = .011).

Regarding BMI the FM group has a higher BMI than both controls (z = −2.265, p = .023) and CFS (z = −3.410, p < .001). CFS have lower BMI than controls (z = −1.664, p = .096).

3.3. Thyroid hormones and total fatigue score

As shown in Table 2, there was no correlation between either TSH or FT4 levels and Total Fatigue scores for the total study population (N = 156). Analyzing each diagnostic group separately gave similar results.

Table 2. Correlation of FT4 and TSH wth total fatigue score for CFS, FM, controls, and the total population.

		CFS		FM		Control		Total population
		n = 48^a		n = 57^b		n = 52^a		n = 156^c
		ρ	p	ρ	p	ρ	p	ρ	p
Fatigue score	TSH	−.16	.29	.01	.96	−.05	.74	−.06	.48
	FT4	.04	.78	−.14	.31	.08	.59	.04	.67

Correlations coefficient: Spearman’s rho (ρ). Fatigue score: Total score from the Chalder fatigue questionnare. FT4 = thyroxine. TSH = thyroid stimulating hormone. ^aMissing: 1. ^bMissing: 2. ^cMissing: 4.

3.4. Thyroid hormones and subgroups of fatigue

A correlation between TSH and the Chalder fatigue score subgroup mental fatigue was found in the control group (p=.023).

Excluding participants using levothyroxine (n = 10) showed a correlation, although not significant, between FT4 and mental fatigue in the FM group (p=.052).

Table 3. Correlation of FT4 and TSH to subgroups of fatigue score, and their p-value using spearman’s rho (ρ).

	CFS^a				FM^b				Control^c				Total population^d
	TSH		FT4		TSH		FT4		TSH		FT4		TSH		FT4
Parameter	ρ	p	ρ	p	ρ	p	ρ	p	ρ	p	ρ	p	ρ	p	ρ	p
Physical fatigue score	−.137	.365	−.088	.561	.156	.259	−.065	.640	.026	.855	.025	.863	−.018	.826	.027	.746
Mental fatigue score	−.185	.208	.179	.223	−.117	.392	−0.184	.173	−.317	.023	.213	.134	−.143	.077	.060	.460

^a 3 missing from original n = 49. ^b4 missing from the original n = 58. ^c2 missing from the original n = 53. ^d9 missing from the original n = 160. Physical fatigue score: Based on question items 1–7 from the Chalder fatigue questionnare. Mental fatigue score: Based on question items 8–11 from the Chalder fatigue questionnare. FT4 = thyroxine. TSH = thyroid stimulating hormone.

No other correlations between TSH, FT4, and the different fatigue scores were seen (Table 3).

4. Discussion

Our main findings are that there were no differences in the level of TSH and FT4 between the groups CFS, FM, and controls. Furthermore, there was no correlation between total fatigue or the subgroups of the fatigue score and TSH and FT4 in the total population though slight correlations were seen in subgroups.

We did not find any differences in TSH and FT4 between CFS and the other groups in this study (FM and healthy controls). This is in alignment with previous reports [15,16]. Still, these studies reported slightly lower values of T3 in the CFS patients compared to healthy controls. We did not measure T3 in our material. In other studies on CFS, elevated FT4 [13], and elevated TSH [14] are reported. The studies however represent different populations regarding age and diagnostic criteria; where [13] investigated a population of adolescents diagnosed with CFS aged 12–18 years and [14] investigated both women and men. This is in contrast to our study where we only included women aged 18 to 60 years and may partly explain the observed differences.

No differences for TSH and FT4 plasma levels between FM and the other groups in this study were seen. This is in line with other studies [25,34]. However, lower levels of T3 have been reported in FM compared to controls [25], a finding we have not repeated as we did not measure T3 in our material.

Interestingly, in patients with HT, no differences of TSH and FT4 were seen between HT patients without FM and the HT patients with comorbid FM [35]. This study did however report significantly higher levels of antithyroid peroxidase antibody (anti-TPO) in the HT patients with FM compared to the HT group without FM. In our study, we did not measure anti-TPO.

We postulated that independent of diagnostic groups there might be associations between symptom levels and levels of TSH and FT4. However, we did not find any such associations. This is in line with the study by Tomic et al. [16] using Fibro fatigue score (FFS). In this study, T3 was also assessed, and a moderate, negative correlation between T3 and the FFS was seen. In contrast to our findings, Ruíz-Pacheco et al. [36] found a positive correlation between TSH and fatigue, and a negative correlation between FT4 and fatigue in an HT population. Also, Haliloglu [35] found a correlation between the Fibromyalgia Impact Questionnaire (FIQ) score and TSH. As these were populations with FM comorbid with HT, these findings may not be directly comparable to our study.

We also explored associations between subgroups of fatigue and the thyroid axis hormones. A weak, negative correlation between TSH and mental fatigue among the healthy controls was seen. A similar tendency was seen in the FM group when excluding participants using levothyroxine from the analysis. No other associations were found between TSH, FT4, and the subgroups of the fatigue score. Since we did not correct for multiple testing, we suggest these correlations represent a type 1 error.

As CFS and FM show symptoms of a clinically hypothyroid state [37] an association with thyroid axis hormones was expected. We did not find any associations for TSH or FT4. However, there may be type 2 errors masking associations. Also, we have not measured T3, anti-TPO, or other markers of thyroid function. Furthermore, we have not gone into any mechanisms of the thyroid axis receptors regarding the prevalence of receptors, binding, transcription (e.g. fibronectin transcription [19]) and other intracellular responses. Other theories include a deficit in transforming T4 to T3 [15]. These mechanisms remain to be explored.

Limitations and strengths

As discussed above, we have only measured TSH and FT4. Several other biomarkers could have been studied, e.g. T3 and anti-TPO. A larger participant sample always would be interesting, however, compared to other studies on CFS and FM our sample size is satisfying. Still, no power calculations were done for these particular analyses as the main aim of the study was measuring immune markers and sample size was calculated in accordance with this. Also, fatigue is only measured using the Chalder Fatigue Questionnaire which might be limited when measuring fatigue in relation to TSH and FT4. Furthermore, all our participants were females aged 18 to 60 years. We cannot conclude anything regarding children, patients older than 60 years or men. Finally, since comparisons between groups were done by means of Kruskal–Wallis tests, this study did not adjust for age or BMI. Additional analysis revealed a weak positive association between TSH and BMI in the FM group only but no other associations between age and BMI and TSH and FT4 (data not shown).

Strengths in our study include the thorough assessment and strict inclusion criteria resulting in strictly defined groups of FM and CSF, and as such we could compare these two groups. All participants lived in a relatively homogenous society and were diagnosed at the same clinic by the same interdisciplinary team of highly experienced clinicians.

Conclusion

In this study, we did not find any group differences between CFS patients, FM patients, and healthy controls in levels of TSH and FT4.

We suggest that future studies examine the field further by exploring the influence of thyroid receptors and responses of the thyroid hormone cascade in clinical disorders such as CFS and FM. Also, the role of thyroid hormones in clinical disorders where fatigue is a symptom should be explored.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. The last author has received a PhD grant from the Central Norway Region Health Authority for carrying out the study (Project number: 46056826).