Sage Journals: Discover world-class research

Abstract

Background

The role of thyroid regulation in migraine is poorly understood, and data is contradictory.

Objective

To study the possible association of clinical features of migraine with patients’ thyroid function.

Patients and methods

One hundred and thirty migraine patients of a tertiary headache center took part in an open-label, cross-sectional comparative study. The Migraine Disability Assessment questionnaire, Spielberger State-Trait Anxiety Inventory, Beck Depression Inventory, Vanderbilt’s Questionnaire of Pain Management, Gothenburg Quality of Life Questionnaire and Migraine-Specific Quality of Life questionnaire were used. The effectiveness of the attacks’ therapy was assessed according to the Migraine Assessment of Current Therapy questionnaire. Levels of thyrotropine (thyroid stimulating hormone), thyroxine, and triiodothyronine were studied by standard immune chemiluminescent method using the Immulite-2000 set.

Results

An inverse correlation between levels of thyroid stimulating hormone in serum and duration of headache attacks was revealed. The effectiveness of abortive therapy for attacks showed a statistically significant positive correlation with thyroid stimulating hormone level. Quality of life measured by a general quality of life questionnaire, as well as the functional and social indices of a migraine-specific questionnaire, showed direct correlation with serum thyroid stimulating hormone.

Conclusion

These results show an association of a more severe clinical course of migraine with lower thyroid stimulating hormone levels.

Keywords

Headache thyroid quality of life efficacy of treatment

Introduction

The role of hormonal regulation in the pathogenesis of migraine attacks is confirmed by the existence of its menstrual-dependent variants. The role of thyroid hormones in migraine is less studied, and the data on this subject are scarce and contradictory. On the one hand, migraine patients do not show changes in the TSH levels (1,2,3). Studying levels of thyroid hormones in patients with various headaches, Larner (4) did not reveal cephalgias associated with hyper-or hypothyroidism. In the study of Ekici and Cebeci (5), subclinical hypothyroidism was found in only five of 98 children with migraine (5.1%). On the other hand, headaches are described among the clinical symptoms of thyroid disease (6,7), and in the International Classification of Headache Disorders section 10.4, headache associated with hypothyroidism is present (8). In the Norwegian population-based study, high levels of thyroid stimulating hormone (TSH) were associated with lower frequency of occurrence of headaches in all age groups (9). The trend was most pronounced in women without thyroid dysfunction in their medical history. In all age groups, TSH was lower in patients suffering from headache, especially in patients with migraine, than in those without headache (9). A correlation of chronic migraines (without medication abuse) with hypothyroidism was found by Bigal et al. (10). Martin et al. (11), as the result of a longitudinal retrospective cohort study, concluded that headache disorders themselves may be a risk factor for new-onset hypothyroidism. Besides, analysis of the risk factors of migraine-related brain white matter hyperintensities (WMH) showed significantly higher frequency of subclinical thyroid dysfunction in the WMH group of patients compared with non-WMH (12). Decreased production of TSH after administration of thyrotropin-releasing factor was revealed in patients that had chronic migraine with medication abuse, which correlated significantly with the duration of the disease (3). Previously, the lack of physiological response of TSH after administering thyrotropin-releasing factor was found by Murialdo et al. (13), who interpreted these results in the light of dopaminergic and noradrenergic hypersensitivity. It was noted additionally that the TSH response to the introduction of thyrotropin-releasing factor in migraine is different from that in depression (13). During a “triple test”, including the introduction of thyrotropin-releasing factor, the increase in TSH levels in patients with migraine was lower than in the control group, but the differences did not reach statistical significance (14), which could reflect the sensitivity of the relevant central receptors (15). So, the influence of thyroid axis state on clinical features of migraine is not clear yet.

Objective

To study the thyroid function of migraine patients and the possible association of severity of migraine symptoms with thyroid hormone levels.

Patients and methods

One hundred and thirty out-patients, referred in 2011–2016 to a tertiary headache center located in Perm, Russia (aged 16–57 years, average 38.11 ± 10.16) (M ± SD) took part in an open-label, cross-sectional comparative study. Diagnostic criteria of the International Classification of Headache Disorders (ICHD-2 and ICHD-3 beta) were used. One hundred and seventeen patients met the criteria for migraine without aura, and in 13 migraine with aura (typical aura with headache) was diagnosed. The study was approved by the Ethical Committee of Perm State Medical University. All patients gave written informed consent. All patients were investigated during the interictal period (a minimum of 72 hours since the end of the last attack).

Inclusion criteria were: Headache matching ICHD criteria for migraine with or without aura; age 16–60; written informed consent; absence of somatic diseases; absence of peripheral nervous system pathology. Exclusion criteria were: Pathological changes on neurological examination; combination of migraine with other types of headaches (except medication overuse headache); trigeminal neuralgia or neuropathy; known thyroid pathology; head trauma in the past; intake of medications influencing thyroid function.

Patients were studied before starting preventive medications. The control group consisted of 15 healthy age-matched volunteers without headache through their lifetime who gave informed consent to participate in the study.

Methods

Methods included a neurological assessment and survey of clinical features of the disease, with registration of data in the patients’ standardized charts. The duration of the disease, the age of onset, the frequency of attacks during the last 3 months preceding the survey, the maximum duration of pain attacks, the average number of days with headache per month, the presence or absence of aura in the structure of the attacks, and the predominant localization of the pain during the attacks were estimated. The intensity of pain was evaluated using a 100-point visual analogue scale (VAS). Information about the occurrence of similar cephalgias in families of patients was obtained. A list of medications used for abortive treatment of attacks was drawn up, and the amount of tablets taken per attack was clarified. Computed tomography/magnetic resonance imaging (CT/MRI) of the brain and consultation with an endocrinologist were used if needed. On entry to the study, patients completed the Migraine Disability Assessment (MIDAS) questionnaire, Spielberger State-Trait Anxiety Inventory, Beck Depression Inventory, and the Vanderbilt Questionnaire of Pain management to assess the patient's habitual coping strategies (all adapted for a Russian-language population (16)). The quality of life of patients was assessed by the modified Gothenburg Quality of Life Questionnaire (GQI) and migraine-specific quality of life questionnaire (MSQOL) (17), adapted for use in the Russian-language environment (18). The effectiveness of abortive therapy for painful episodes was assessed according to the Migraine-ACT (Assessment of Current Therapy) (19) questionnaire. The study of thyrotropin – thyroid stimulating hormone (TSH), thyroxin (T4), and triiodothyronine (T3) was carried out by the standard immune chemiluminescent method using the Immulite 2000 set. Patients with hormonal levels exceeding normal were not excluded from the study. In order to avoid bias, measurements were performed twice for each patient, and the mean value for the patient was taken. Only patients who presented laboratory results were included in the study.

Statistical processing of the results was carried out using the Statistica 6.0 software. Depending on the distribution of the analyzed parameters, we used parametric (mean values, standard deviation, t-tests) and non-parametric (median (Me) and 95% confidence interval (CI)) methods, with preference given to non-parametric methods. Comparative analysis of independent groups was performed using the Kruskal-Wallis test, Mann-Whitney U-test, two-sample Kolmogorov-Smirnov test and analysis of variance (ANOVA). Correlation analysis was performed using the non-parametric Spearman coefficient (R). Results were considered to be statistically significant at p < 0.05.

Results

According to the diagnostic criteria of the International Headache Society, the surveyed patients experienced attacks of severe headache lasting from 4 to 72 hours with nausea, in some cases vomiting, with photo- and phonophobia. Patients suffering from migraine with aura complained of photopsias and/or narrowing of the field of vision for up to 60 minutes before the headache (typical aura with headache). Clinical characteristics of migraine in studied patients are shown in Table 1.

Table 1.

Clinical characteristics of patients (Me; 95% CI).

Length of disease (years)	15.00 (14.32–17.84)
Age at beginning of disease (years)	20.00 (20.35–23.75)
Frequency of attacks (per month, during the last 3 months)	3.00 (4.08–5.71)
Length of attacks (hours)	48.00 (42.71–54.22)
Days with headache per month (days)	6.00 (7.24–9.47)
Intensity of headache (VAS score)	89.00 (83.93–88.72)
Quantity of analgesic medications per attack (tablets)	2.00 (3.10–4.61)
State anxiety (points)	42.00 (41.12–44.38)
Trait anxiety (points)	47.00 (45.16–48.51)
Depression (points)	9.00 (8.14–10.58)
Disability (MIDAS score)	25.00 (25.20–34.79)

Cephalgic attacks in the studied patients were severe – Me 89.00 points VAS (from 32 to 100 points), and the frequency of attacks during the previous 3 months was high (up to 25 attacks per month, Me three attacks). Eighteen patients (13.8%) matched the IHS criteria for chronic migraine (15 or more days per month). The median number of headache days per month in the group of migraine sufferers was 6 days. The average duration of attacks was found to be 48 hours. Forty-one (31%) patients had a history of attacks lasting more than 72 hours (status migrainosus). More than half of the patients (73 patients, 56%) had nocturnal headache attacks. The maximum duration of disease was 42 years, and the average was 15 years. Fifty-nine patients (45%) reported a family history of headache attacks. Pain attacks significantly limited the daily activities of patients: The average MIDAS score was 25 points, although this varied widely from 0 to 158 points. Patients used various medications for pain relief; usually these were combined analgesics, the number of tablets per attack ranged from 1 to 30 (Me 2; 95% CI 2.50–5.08), indicating low efficiency of treatment. Triptans were used by 37 patients (28%). Migraine-ACT scores also showed low satisfaction with the action of medicines: Me 1 point; 95% CI 1.18–1.99 (full satisfaction with the treatment corresponds with 4 points).

Thyrotropin plasma concentrations in migraine patients ranged from 0.1 to 8.7 μIU/ml (Me 1.83; 95% CI 1.76–2.68), while the normal range was 0.4–4.0 μIU/ml. Abnormal levels of thyrotropin were found in seven patients (5%). The levels of triiodothyronine were normal in all patients; in one patient, thyroxin concentration exceeded normal levels. In one control, the level of triiodothyronine also exceeded normal concentrations; thyrotropin was normal in all controls. In general, median levels of TSH and thyroxin did not differ between patients and controls; triiodothyronine in the patients’ group (Me 3.01; 95% CI 2.71–3.24 pg/ml) was lower compared with controls (Me 3.56; 95% CI 3.37–4.05 pg/ml; p = 0.01). The thyroid status of patients and healthy subjects is shown in Table 2. Thyrotropin and thyroxin levels correlated significantly: R = −0.318, p = 0.028.

Table 2.

Thyrotropin, thyroxin and triiodothyronine levels in migraine patients and controls (Me; 95% CI).

Hormone	Migraine patients n = 130	Controls n = 15
Thyrotropin (μIU/ml)	1.83 (1.76–2.68)	1.92 (1.63–2.83)
Thyroxin (ng/dl)	1.35 (1.26–2.39)	1.41 (1.31–1.55)
Triiodothyronine (pg/ml)*	3.01 (2.71–3.24)	3.56 (3.37–4.05)

statistically significant difference, p = 0.01.

Thyrotropin levels did not correlate with the age of patients nor with the duration of the disease. No correlations were found between thyrotropin levels and frequency of attacks (p = 0.857); patients with chronic migraine showed TSH Me = 1.14 μIU/ml (95% CI 0.28–2.39) vs Me = 1.87 mkIU/ml (95% CI 1.82–2.81) in patients with episodic migraine. This difference was not statistically significant. Thyrotropin concentrations inversely correlated with migraine attack duration (R = −0.344, p = 0.014) and positively with Migraine-ACT questionnaire score (R = 0.365, p = 0.015). Correlations of thyrotropin levels and clinical features of migraine are shown in Table 3.

Table 3.

Correlations of thyrotropin levels with clinical features of migraine, anxiety and depression of patients.

Clinical features of migraine	Results of analysis
R*	p
Frequency of attacks	0.026	0.857
“Headache days” per month	−0.139	0.335
Duration of the disease	−0.184	0.198
Age of onset of migraine	0.032	0.820
Number of analgesic tablets per attack	−0.164	0.252
Length of attacks	−0.344	0.014
Pain intensity (VAS)	0.076	0.597
Trait anxiety	−0.101	0.503
State anxiety	−0.211	0.159
Depression	−0.250	0.093
MIDAS score	−0.126	0.408
Migraine-ACT score	0.365	0.014
Active coping strategies	−0.051	0.732
Passive coping strategies	−0.193	0.198

Spearman’s coefficient.

Statistically significant correlations between quality of life (QoL) scores and thyrotropin levels were also found. Plasma thyrotropin positively correlated with QoL of patients measured by the Goetheborg Questionnaire (R = 0.355; p = 0.033). Functional and social indexes of the migraine-specific MSQOL questionnaire also showed correlations with plasma thyrotropin (p = 0.046 and p = 0.017 respectively; Table 4).

Table 4.

Correlations of quality of life indexes and thyrotropin levels in migraine patients.

QoL indexes	Correlations and statistical significance
R*	p
GQI scores**	0.355	0.033
MSQOL – global index	0.206	0.167
MSQOL – functional index**	0.294	0.046
MSQOL – psychological index	0.175	0.242
MSQOL – social index**	0.348	0.017
MSQOL – medical index	0.027	0.856

Spearman’s coefficient.

Statistically significant correlations.

No correlations were found between clinical features of migraine or QoL scores with thyroxin and triiodothyronine levels. A small group of seven migraine patients with abnormal thyroid function did not show differences in clinical features of migraine compared with the rest of the study group.

Discussion

Understanding migraine as a multifactorial disease, the effect of hormonal systems on its clinical course should be taken into account. In contrast to the relationship of migraine with the level of ovarian hormones, which is widely discussed in the literature, the dependence of the clinical features of migraine on the state of the thyroid system is not yet clear. The only information is about lower levels of TSH in patients with migraine compared to those who do not suffer from headaches (9), which was not confirmed in our study, but the significance of this finding is limited by the low sample size of the control group. Nevertheless, the absence of changes in the levels of thyrotropin and thyroxin in migraine patients compared with the control group is consistent with data of Rainero et al. (3). The median level of triiodothyronine (which did not correlate with clinical features of migraine or QoL scores) was lower in the patients’ group.

In migraine patients, we found an association of clinical features of the disease and the level of thyroid-stimulating hormone in the blood serum. A moderately strong, statistically significant inverse correlation between the levels of TSH in serum and the duration of headache attacks was revealed. The effectiveness of abortive therapy for attacks showed a statistically significant moderately positive correlation with TSH level. Quality of life measured by a general QoL questionnaire, as well as the functional and social indices of the migraine-specific questionnaire, showed a moderately strong, statistically significant direct correlation with serum TSH. The mechanisms of these associations remain unclear and demand further study.

Conclusion

Our study demonstrates that a relatively low level of serum thyrotropin, even in terms of its normal values, may be associated with a more prolonged and refractory clinical course of migraine, with greater impact on quality of life. The pathogenesis of this association demands further investigation. Another finding of the study is the absence of a difference in median values of TSH between migraine patients and controls. Besides, we can mention that in our study only about 5% of 130 consecutively enrolled migraine patients showed abnormal thyroid function.

Limitations

Firstly, only patients referred to a specialized headache center (i.e. patients with active and rather severe migraine) took part in the study. This makes it difficult to expand the results to all categories of migraine patients. Besides, as patients were studied on reference to the headache center before starting preventive medications, most had not previously kept headache diaries. For such patients, the data on frequency of attacks was based on their subjective remembrance. Secondly, a possible limitation of the study is that patients with abnormal hormonal levels were not excluded (though there were only a few of those). Thirdly, there was the small size of the control group, which consisted of only 15 volunteers, and fourth, this was an exploratory study and therefore no Bonferroni corrections were made for multiple comparisons.

Clinical implications

Despite the fact that only 5% of migraine patients showed abnormal thyrotropin, thyrotropin levels should be checked in patients with severe migraine, with appropriate correction if needed.

The mechanisms of the association of migraine with thyrotropin levels need further investigation.

Footnotes

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

References

Amy

. Tests of thyroid function in chronic headache patients. Headache 1987; 27: 351–353.

Parashar

Bhalla

Rai

et al.

Migraine: Is it related to hormonal disturbances or stress?

Int J Womens Health 2014; 6: 921–925.

Rainero

Ferrero

Rubino

et al.

Endocrine function is altered in chronic migraine patients with medication-overuse. Headache 2006; 46: 597–603.

Larner

. Thyroid dysfunction and headache. J Headache Pain 2006; 7: 51–52.

Ekici

Cebeci

. The debate on the link between subclinical hypothyroidism and childhood migraine: Is initial endocrinological evaluation necessary for children with migraine? Acta Neurol Belg 2015; 115: 123–127.

Carvalho

MFL

de Medeiros

Valença

. Headache in recent onset hypothyroidism: Prevalence, characteristics and outcome after treatment with levothyroxine. Cephalalgia 2017; 37: 938–946.

Stone

Foulkes

Adamson

et al.

Thyrotoxicosis presenting with headache. Cephalalgia 2007; 27: 561–562.

Headache Classification Committee of the International Headache Society. The International Classification of Headache Disorders, 3rd edition (beta version). Cephalalgia 2013; 33: 629–808.

Hagen

Bjøro

Zwart

et al.

Low headache prevalence amongst women with high TSH values. Eur J Neurol 2001; 8: 693–699.

10.

Bigal

Sheftell

Rapoport

et al.

Chronic daily headache: Identification of factors associated with induction and transformation. Headache 2002; 42: 575–581.

11.

Martin

Pinney

Xie

et al.

Headache disorders may be a risk factor for the development of new onset hypothyroidism. Headache 2017; 57: 21–30.

12.

Trauninger

Leél-Őssy

Kamson

et al.

Risk factors of migraine-related brain white matter hyperintensities: An investigation of 186 patients. J Headache Pain 2011; 12: 97–103.

13.

Murialdo

Masturzo

Ferrari

et al.

Thyroid-stimulating hormone and prolactin responses to thyrotropin-releasing hormone in common migraine. Cephalalgia 1987; 7: 267–272.

14.

Awaki

Takeshima

Takahashi

. A neuroendocrinological study in female migraineurs: Prolactin and thyroid stimulating hormone responses. Cephalalgia 1989; 9: 187–193.

15.

Gordon

Lipton

Brown

S-L

et al.

The neuroendocrine challenge paradigm in headache research. Cephalalgia 1995; 15: 292–295.

16.

Belova A. Scales, tests and questionnaires in neurology and neurosurgery. Guide for physicians and research fellows. Moscow: 2004, 432 pp. (in Russian).

17.

El Hasnaoui

Vray

Blin

et al.

Assessment of migraine severity using the MIGSEV scale: Relationship to migraine features and quality of life. Cephalalgia 2004; 24: 262–270.

18.

Shestakov

Starikova

. Adaptation of Russian version of QVM questionnaire for quality of life study in migraine. Pain 2009; 22: 18–22 (in Russian).

19.

Kilminster

Dowson

Tepper

et al.

Reliability, validity and clinical utility of the Migraine-ACT questionnaire. Headache 2006; 46: 553–562.

Thyrotropin levels and severity of symptoms in migraine patients of tertiary headache center

Abstract

Background

Objective

Patients and methods

Results

Conclusion

Keywords

Introduction

Objective

Patients and methods

Methods

Results

Discussion

Conclusion

Limitations

Clinical implications

Footnotes

Declaration of conflicting interests

Funding

References