Sage Journals: Discover world-class research

Abstract

Background

Migraine with cranial autonomic symptoms is well described in the literature, but its prevalence in previous studies varies enormously. A precise estimate of the prevalence in a population-based material is important because migraine with cranial autonomic symptoms might represent an endophenotype, in which genetic and pathophysiological features differ from those without cranial autonomic features. The aim of the present study, therefore, was to estimate the prevalence in a big population-based sample using both questionnaire-based diagnosis (N = 12,620) and interview-based diagnosis (N = 302). We validate questionnaire-based diagnosis of migraine with cranial autonomic symptoms and develop the first diagnostic criteria for future research of this possible endophenotype.

Methods

The Danish Blood Donor Study included 127,802 persons who all received a migraine diagnostic questionnaire. Participants who had answered the diagnostic questionnaire constituted the Danish Migraine Population Cohort (N = 62,677) of whom 12,620 had migraine. The diagnostic migraine questionnaire included questions about the following cranial autonomic symptoms: Facial/forehead sweating, lacrimation, ptosis, conjunctival injection, rhinorrhea, nasal congestion, and miosis. Validation was performed by a follow-up semi-structured, purpose-built interview of 302 participants with migraine, where detailed questions were asked to ascertain the validity of the symptoms.

Results

The questionnaire-based prevalences of one, respectively two cranial autonomic symptoms were 57% and 31%. The semi-structured interview-based prevalences of one, respectively two symptoms were 44% and 22%. The most common symptoms were facial/forehead sweating (39%) and lacrimation (24%). The specificity of the questionnaire was 80% and the sensitivity was 68%. Correlation analysis showed a weak correlation between symptoms ranging from 0.07 – 0.41, and no clear clustering of symptoms was detected. We suggest the first diagnostic appendix criteria for genetic and epidemiological studies and tighter criteria for clinical and pathophysiological studies. We encourage further studies of severity and consistency of symptoms.

Conclusion

Migraine with cranial autonomic symptoms is prevalent in the general population. Suggested diagnostic appendix criteria are important for future studies of this possible migraine endophenotype.

Keywords

Migraine cranial autonomic symptoms diagnostic criteria

Introduction

Cranial autonomic symptoms (CAS), such as conjunctival injection, lacrimation, nasal congestion, miosis, ptosis, rhinorrhea, forehead and/or facial sweating, are typically observed in cluster headache and other trigeminal autonomic cephalalgias. The symptoms are included in the diagnostic criteria of these disorders in the International Classification of Headache Disorders (1). CAS have also been observed during migraine attacks for many years (2 –5), although they are rarely part of the primary complaints of the patients according to our clinical experience. The literature on migraine with CAS shows a wide range of prevalences from 3.1% to 82% (6,7). Further epidemiological and clinical data, preferably population-based, are therefore needed for a reliable estimate of the prevalence.

In the recent years, potential endophenotypes of migraine have been of high interest (8). Endophenotypes are biological traits that represent an association between a genetic predisposition for a disease and the clinical phenotype. They are used to strengthen genetic analyses of complex genetic disorders (9). It remains uncertain whether migraine with CAS represents such an endophenotype of migraine (10), because necessary genetic and epidemiological studies that can distinguish between migraine with and without CAS, require large samples to provide sufficient statistical power for identification of specific associations. Most previous studies have not been large enough for such distinction. Face-to-face semi-structured interviews are preferable but time consuming, and it is unlikely that enough data can be collected with this methodology in the foreseeable future. Thus, a validated questionnaire-based, diagnostic study is the only method that can provide sufficient power. We used the Danish Migraine Population Cohort which is largely representative of the general Danish migraine population, and includes 12,620 migraine subjects diagnosed by a validated, diagnostic questionnaire (11). The questionnaire contained specific questions about CAS. The aim of the present study was to estimate the prevalence of migraine with CAS in a large population-based sample using both questionnaire-based diagnosis and interview-based diagnosis. A further aim was to validate questionnaire-based diagnosis of migraine with cranial autonomic symptoms and to develop the first diagnostic criteria to date for future research of this possible endophenotype.

Materials and methods

The Danish Migraine Population Cohort

This study is based on the Danish Migraine Population Cohort (DaMP) (11), a subgroup of the Danish Blood Donor Study (DBDS). DBDS is a large ongoing prospective cohort and biobank. Blood donors between 18 and 67 years from all regions of Denmark have been invited to the study from 2010 and the study is still ongoing. Less than 5% decline the study invitation (12). We recontacted all participants from the DBDS, who were connected to the Danish public electronic mailing system (e-Boks) between May 2020 and August 2020 (n = 127,802). All participants were asked to fill out an extensive diagnostic migraine questionnaire. Participants who had answered the diagnostic questionnaire constituted the DaMP cohort (N = 62,677), resulting in 12,620 participants with either migraine without aura (MO), migraine with typical aura (MA), or both migraine with and without typical aura (MAMO) according to the third edition of the International Classification of Headache Disorders (1). The diagnostic questionnaire was previously validated by a specially trained neurology resident and had a positive predictive value of 97% (11). Further, DaMP is representative of the general Danish migraine population in terms of age, gender-ratio, migraine prevalence and triptan use, but with fewer chronic migraine cases. The cohort has been described in detail elsewhere (11).

Diagnostic questionnaire: Diagnostic questions regarding migraine with CAS

The online diagnostic questionnaire contained 105 questions about migraine – including questions about the presence of CAS during migraine attacks.

The presence of migraine with CAS was determined by asking the following questions, to which the participants could answer yes or no:

During migraine attacks, it is possible to experience different autonomic symptoms. These symptoms are caused by an activation of the autonomic nervous system. Have you experienced any of the following symptoms, during migraine attacks?

(1) Your eye starts to tear?

(2) The white part of your eye (sclera) turns red?

(3) One of your pupils (the black part of the eye) becomes very small?

(4) One of your eyelids is hanging?

(5) Your nose starts to run?

(6) Your nose congests?

(7) You sweat extensively on your forehead?

(8) You sweat extensively on your face?

Validation of questionnaire-based assessment of migraine with CAS

To validate the questions about migraine with CAS, we performed a blinded, semi-structured follow-up interview by telephone, described elsewhere (11). The interview was carried out by a specially trained neurology resident, who was blinded to the answers from the diagnostic questionnaire, and 302 participants who fulfilled the diagnostic criteria for migraine. In the follow-up interview, the questions regarding migraine with CAS were repeated and detailed questions were asked to ascertain the validity of the symptoms. For all symptoms, we asked if the symptoms occurred exclusively during attacks. Regarding conjunctival injection, miosis, and ptosis, we asked if the participants had confirmed these symptoms themselves by looking in the mirror, or if anyone else had confirmed the symptoms. Regarding forehead and facial sweating, we asked if the sweating was extensive, and the participants were asked if they could distinguish between facial and forehead sweat.

Statistics

Statistical analyses were performed using R version 4.0.0. Descriptive analyses were made to assess the prevalence of the individual CAS in the DaMP cohort. Variables were described as number of participants (n) and frequency (%). Participants with missing data on all questions regarding migraine with CAS were excluded. Confusion matrices were created using the R packages tidyverse, rstatix, and caret. We constructed a confusion matrix to calculate specificity (SP) and sensitivity (SE) using caret. SP was the ability of the diagnostic questionnaire to correctly not classify migraine without CAS as migraine with CAS and was defined as:

\frac{Number of true negatives}{Number o f true negatives + number o f false positives}

and SE was the ability of the diagnostic questionnaire to correctly classify participants with migraine with CAS and was defined as:

\frac{Number of true positives}{Number o f true positives + number o f false negatives}

The correlation between the individual CAS was calculated. Only significant Pearson’s correlation coefficients were presented. A correlation of one suggested the strongest correlation, and a correlation of zero suggested the weakest correlation. To identify clustering patterns in the autonomic symptoms, we applied K-modes unsupervised clustering analysis using the R-package klaR version 0.6–15 (13,14). K-modes clustering is an algorithm for clustering of large-scale categorical data sets. We performed the clustering analysis using a maximum of 10 iterations, a weighted distance between categories in the data, and the number of modes set to three. To identify dominant patterns of clusters, we performed the clustering analysis 50 times. We identified 15 studies on migraine with CAS, which included a population prevalence of migraine with CAS (3 –7,10,15 –23). Data on the population prevalence of having one or more CAS and the population prevalence of individual CAS items were extracted from the articles and reported in this study for comparison with the population prevalences in this study. If the population prevalence was presented in the respective article, it was read directly from the article, if it was not presented, the population prevalence was calculated by the given data.

Standard protocol approval, registrations, and patient consent

Written informed consent was obtained from all participants. The study was approved by the Danish Ethical Standards Committee (1-10-72-95-13, SJ-740, 1-90-09-88 and 1-70-04-07) and the Danish Data Protection Agency (P-2019-99).

Results

Questionnaire-based prevalence of CAS

Among all participants in DaMP (N = 62,677), 33,444 were female and 29,233 were male. The mean age was 48 years (SD = 14), and n = 12,620 fulfilled the diagnostic criteria for MO, MA, or MAMO. The prevalence of one or more CAS was 57% (n = 7,179) and two or more CAS was 31% (n = 3,968). Facial and/or forehead sweat was the most common symptom with a prevalence of 39% (n = 4,909). Twenty-four percent (n = 3,031) had lacrimation, 14% (n = 1,747) had ptosis, 13% (n = 1,615) had conjunctival injection, 11% (n = 1,439) had rhinorrhea, and 11% (n = 1,362) had nasal congestion. Miosis was the least common symptom with a prevalence of 6% (n = 752) (Table 1).

Table 1.

Prevalence of cranial autonomic symptoms during migraine and cluster headache.

	Prevalence of CAS during migraine attacks in the DaMP cohort		Prevalence of CAS during cluster headache attacks in DHC²⁴
Cranial autonomic symptoms	Diagnostic questionnaire (N = 12,620)	Semi-structured validation interview (N = 302)	Questionnaire (N = 57)
	n (%)	n (%)	n (%)
Facial/forehead sweat^a	4,909 (39)	59 (20)	94 (19)
Lacrimation	3,031 (24)	40 (13)	272 (54)
Ptosis	1,747 (14)	31 (10)	224 (45)
Conjunctival injection	1,615 (13)	50 (17)	241 (48)
Rhinorrhea	1,439 (11)	16 (5)	183 (37)
Nasal congestion	1,362 (11)	23 (8)	200 (40)
Miosis	752 (6)	19 (6)	51 (10)
Number of symptoms
≥1 CAS	7,179 (57)	132 (44)	408 (82)
≥2 CAS	3,968 (31)	66 (22)	–
≥3 CAS	2,153 (17)	30 (10)
≥4 CAS	1,018 (8)	6 (2)
≥5 CAS	396 (3)	3 (1)
≥6 CAS	118 (1)	1 (0.3)
= 7 CAS	23 (0.2)	0 (0)

CAS: cranial autonomic symptoms; DaMP: Danish Migraine Population cohort; DHC: Danish Headache Center.

^a In the DaMP cohort the participants answered yes to having either or both facial and forehead sweat. In the DHC cluster headache cohort the prevalence refers to participants who answered yes to having both facial and forehead sweat.

Semi-structured interview-based prevalence of CAS

We performed a blinded, semi-structured validation interview of 302 participants, who fulfilled the criteria for migraine, randomly selected from the DaMP. In the semi-structured interview, detailed questions were applied to ascertain the validity of the symptoms as described in the methods section. The semi-structured interview-based prevalence of two or more CAS during migraine attack was 22% (n = 66). Facial and/or forehead sweat was the most common symptom with a prevalence of 20% (n = 59). Seventeen percent (n = 50) had conjunctival injection, 13% had lacrimation (n = 40), 10% had ptosis (n = 31), 8% had nasal congestion (n = 23), 6% had miosis (n = 19), and 5% had rhinorrhea (n = 16) (Table 1).

Validation of questionnaire-based diagnostic questions regarding CAS

The specificity (SP) and sensitivity (SE) of the questionnaire-based questions are listed in Table 2. SP of having one or more CAS was $\frac{109}{109 + 56} = 0.66$ and SE was $\frac{102}{102 + 24} = 0.81$ . The SP of having two or more symptoms was 0.80 and SE was 0.68. Having additional symptoms increased the SP but generally weakened the SE. Two confusion matrices used for the calculations are presented in Table 3. For the individual symptoms, mean SP was 0.88 (ranging from 0.73–0.96) and mean SE was 0.54 (ranging from 0.30–0.75). Based on the possible parasympathetic and sympathetic contributions to peripheral and central sensitization during migraine, we grouped the CASs in three groups:

Table 2.

The specificity and sensitivity of the questionnaire-based questions regarding cranial autonomic symptoms.

Cranial autonomic symptoms	SP	[95% CI]	SE	[95% CI]
Cumulative cranial autonomic symptoms
≥1 CAS	0.66	[0.59–0.73]	0.81	[0.74–0.88]
≥2 CAS	0.80	[0.75–0.85]	0.68	[0.57–0.80]
≥3 CAS	0.88	[0.84–0.92]	0.61	[0.43–0.79]
≥4 CAS	0.91	[0.88–0.95]	0.83	[0.54–0.13]
≥5 CAS	0.97	[0.95–0.99]	0.67	[0.13–1.20]
Individual cranial autonomic symptoms
Forehead and/or facial sweating	0.73	[0.68–0.79]	0.71	[0.60–0.83]
Lacrimation	0.83	[0.78–0.88]	0.70	[0.56–0.84]
Ptosis	0.93	[0.90–0.96]	0.58	[0.41–0.75]
Conjunctival injection	0.89	[0.85–0.93]	0.41	[0.27–0.55]
Rhinorrhea	0.90	[0.87–0.94]	0.75	[0.54–0.96]
Nasal congestion	0.91	[0.88–0.95]	0.30	[0.12–0.49]
Miosis	0.96	[0.94–0.99]	0.32	[0.11–0.52]
Mean	0.88		0.54
Grouping of symptoms
1. Secretory symptoms: lacrimation and/or rhinorrhea	0.80	[0.75–0.85]	0.70	[0.57–0.83]
2. Vasodilatory symptoms: Conjunctival redness and/or nasal congestion	0.82	[0.78–0.87]	0.45	[0.33–0.58]
3. Sympathetic symptoms: Ptosis, miosis and/or facial/forehead sweat	0.69	[0.63–0.76]	0.70	[0.61–0.79]
Mean	0.77		0.62

SP: Specificity; SE: Sensitivity; CAS: Cranial autonomic symptoms. For confidence interval (CI) calculations 1-α was set to 0.95.

Table 3.

Semi-structured interview vs diagnostic questionnaire.

Confusion matrix based on the presence of n ≥ 1 CAS
		Diagnostic questionnaire
		Yes	No
Semi-structured interview	Yes	102	24
Semi-structured interview	No	56	109
Confusion matrix based on the presence of n ≥ 2 CAS
		Diagnostic questionnaire
		Yes	No
Semi-structured interview	Yes	43	20
Semi-structured interview	No	46	182

Confusion matrices based on data from the diagnostic questionnaire and the semi-structured interview. Confusion matrices for the presence of n ≥ 1 or n ≥ 2 cranial autonomic symptoms (CAS) are presented.

1)Secretory symptoms: Lacrimation and/or rhinorrhea.

2) Vasodilatory symptoms: Conjunctival redness and/or nasal congestion.

3) Sympathetic symptoms: Ptosis, miosis, facial and/or forehead sweating.

Groups one and three had the highest SE of 0.70 (Table 2). We also calculated SP and SE on additional groups based on alternative parasympathetic and sympathetic attributes, which are presented in Supplementary Table 1.

Correlation of CAS

We calculated the correlation between the individual CAS. In general, there was a weak correlation between symptoms, ranging from 0.07–0.41, where one is the strongest correlation and zero is the weakest correlation. Among all the symptoms, nasal congestion and rhinorrhea were the most correlated symptoms with a correlation coefficient of 0.41. The second most correlated symptoms were rhinorrhea and lacrimation with a correlation coefficient of 0.37 (Figure 1).

Figure 1:

Correlation of cranial autonomic symptoms. Only significant Pearson’s correlation coefficients are presented. A correlation of one and the color red represents the strongest correlation between two symptoms and a correlation of zero and the color white represents the weakest correlation.

K-modes unsupervised clustering analysis

To identify whether a cluster pattern exists among the cranial autonomic symptoms, we performed K-modes unsupervised clustering analysis 50 times and identified two dominant patterns of clusters. In 10 of 50 analyses, we identified the following three clusters: 1) facial and forehead sweating, 2) lacrimation, 3) conjunctival redness, ptosis, miosis, rhinorrhea, and nasal congestion. In nine of 50 analyses, we identified the following three clusters: 1) facial and forehead sweating, 2) ptosis, 3) conjunctival redness, lacrimation, miosis, rhinorrhea, and nasal congestion. In the remaining 31 of 50 analyses, we did not find any patterns of clusters that occurred in more than five of 50 analyses. Results of the cluster analysis is presented in Supplementary Table 2.

Proposed diagnostic appendix criteria for Migraine with CAS

We do not suggest to diagnostically divide migraine according to the presence of CAS. For future specific studies, however, we have developed proposed diagnostic appendix criteria for migraine with CAS. Two sets of criteria were developed. The first set was developed for use in genetic and epidemiological studies, where large numbers, usually collected by self-reported questionnaire, are necessary (25) (Table 4). The second set was developed for use in clinical and pathophysiological studies, where a smaller sample size is accepted but specificity is of prime importance (Table 5).

Table 4:

Proposed diagnostic appendix criteria for migraine with cranial autonomic symptoms for use in genetic and epidemiological studies.

A. Attacks fulfilling the diagnostic criteria for:

1.1 Migraine without aura and/or

1.2 Migraine with aura

B. Data obtained by questionnaire

C. At least two of the following fully reversible autonomic symptoms have been present during migraine attack:

1. Conjunctival injection

2. Lacrimation

3. Nasal congestion

4. Rhinorrhea

5. Forehead or facial sweating

6. Miosis

7. Ptosis

8. Eyelid oedema

D. Not better accounted for by another ICHD-3 diagnosis

Table 5:

Proposed diagnostic appendix criteria for migraine with cranial autonomic symptoms for use in clinical and pathophysiological studies.

A. Attacks fulfilling the diagnostic criteria for:

1.1 Migraine without aura and/or

1.2 Migraine with aura

B. During at least one third of the attacks, symptoms according to criterion C have been present. Data must be obtained by semi-structured interview

C. At least two of the following fully reversible autonomic symptoms:

1. Conjunctival injection

2. Lacrimation

3. Nasal congestion

4. Rhinorrhea

5. Forehead or facial sweating

6. Miosis

7. Ptosis

8. Eyelid oedema

D. Not better accounted for by another ICHD-3 diagnosis

Discussion

In this study we present the largest population-based migraine cohort that has been investigated with respect to cranial autonomic symptoms (CAS). Our analysis includes the first validation of CAS by a subsequent semi-structured, purpose-built interview. We report a very high prevalence of migraine with respectively one CAS of 57% and two CAS of 31% in the general population. We further propose two sets of diagnostic appendix criteria, one for genetic and epidemiological studies and another stricter set of criteria, for clinical and pathophysiological studies. The criteria are ready for further field testing.

The prevalence of migraine with CAS varies highly in previous studies

The prevalence of migraine with CAS varies greatly in the literature from 3.1–82% (6,7). An overview of the previous studies is presented in Figure 2 and Table 6. We identified 15 studies, that examined the prevalence of migraine with CAS. Most previous studies report only the prevalence of having one or more CAS during a migraine attack. In general, the studies have been low-powered and have lacked validation of the symptoms. Nine of the 15 studies were performed by face-to-face interviews (5 –7,10,15 –17,20,23) One of these nine studies reported a remarkably low prevalence of migraine with CAS of 3.1% (6) while the other eight studies reported a wide range of prevalences from 41.4–82% with a mean of 55.4% (5,7,10,15 –17,20,23) This corresponds to our questionnaire data as we report a prevalence of having one or more CAS during migraine attack of 57%. The variation in the prevalence of migraine with CAS in the literature is probably explained by different study methodologies and study populations. The highest prevalence of 82% was reported in a cohort of participants with chronic migraine (CM) recruited from a tertiary headache center (7), while the lower prevalence of 26.9% was reported in a cohort reflecting the general population (4). The study with the lowest prevalence of 3.1% did not state if it was examined in a clinical or a general study population (6).

Figure 2.

Overview of the literature on migraine with cranial autonomic symptoms. Study population and prevalence of migraine with cranial autonomic symptoms (CAS) in the available literature and the current study are presented. CAS: cranial autonomic symptoms, a = the number refers to the prevalence of CAS ipsilateral to the headache, b = the number refers to the prevalence of unilateral CAS.

Table 6.

Prevalence of migraine with ≥ 1 CAS from the literature and the current study.

Study	Year	N	Method of CAS data acquisition	Intention	Prevalence of ≥1 CAS (%)
Population: CM recruited from headache center.
Riesco et al. (7)	2016	100	Face-to-face interview by neurologist	To analyze the prevalence and profile of cranial autonomic parasympathetic symptoms in CM patients.	82
Population: EM and CM recruited from a hospital department
Recruited from a neurology department
Nasir et al. (22)	2016	105	Not specified	To determine the frequency of CAS in migraine patients	70.5
Fatima et al. (21)	2021	132	Not specified	To evaluate the presence of CAS in migraine patients	55.3
Recruited from Section of Radiology, Regional Education and Research Hospital
Ceylan et al. (15)	2019	1,080	Face-to-face interview by neurologists using a questionnaire form	To investigate the coexistence CAS, vertigo, dizziness and allodynia, and their effects on quality of life in patients with CM and EM.	69.4
Population: EM and CM recruited from a headache center
Recruited from a headache clinic of a department of psychiatry
Gupta et al. (16)	2007	78	Face-to-face interview, does not specify by who	To determine the prevalence of CAS in migraineurs, to characterize migraineurs with CAS and determine association between CAS and laterality of headache.	73.1
Recruited from a headache center
Karsan et al. (18)	2022	340	Review of first consultation clinic letter/journal	To study the prevalence and phenotype of CAS among migraineurs as well as the reporting of throat swelling and/or voice change as possible CAS among all primary headache disorders.	74
Togha et al. (19)	2021	904	Self-reported questionnaire	To evaluate the frequency and characteristics of CAS in patients with episodic and chronic migraine and their relation to disease course	61
Lai et al. (3)	2009	786	Self-reported questionnaire	To characterize CAS in patients with migraine compared with CAS in cluster headache patients	56
Rozen et al. (20)	2011	117	Face-to-face interview by a neurologist	To determine whether a history of cigarette smoking is associated with the development of CAS with migraine.	52
Barbanti et al. (5)	2002	177	Face-to-face interview using a structured questionnaire	To assess the frequency of U-CAS and make a clinical characterization of migraineurs with U-CAS.	45.8^a
Danno et al. (23)	2020	373	Self-reported questionnaires followed up by face-to-face interviews with headache specialists	To make a clinical characterization of CAS in Japanese migraine patients and to get insight into the pathophysiological implications.	42.4
Barbanti et al. (10)	2016	757	Face-to-face interview by neurologist using a semi-structured questionnaire	To investigate the clinical phenotype of migraine with U-CAS	37.4^a
Population: EM recruited from headache center
Guven et al. (17)	2013	186	Face-to-face interview by neurologists	To investigate the frequency of U-CAS during migraine attacks, and to compare clinical characteristics of migraineurs with and without CAS.	41.4^b
Population: General migraine population with EM and CM
Danish blood donors
DaMP, questionnaire	2022	12,620	Self-reported questionnaire	To clinically characterize migraine in the general population	56.9
DaMP, follow-up	2022	302	Semi structured telephone interview by a specially trained neurology resident	To validate the DaMP diagnostic questionnaire	43.7
General population in Germany
Obermann et al. (4)	2007	841	∼60%: self-reported questionnaire and ∼40%: telephone interview by specially trained medical student.	To provide a population-based prevalence of U-CAS in migraine patients, to assess whether the expression of unilaterality of CAS as well as the unilaterality of head pain is more pronounced in patients with U-CAS compared to migraine patients without, and to assess whether pain intensity is stronger in migraine patients with U-CAS than migraine patients without.	26.9^a
Turkish headache database ^c
Uluduz et al. (6)	2016	2,862	Face-to-face interview by a headache specialist using a structured questionnaire.	To evaluate the prevalence of U-CAS in participants with migraine and to assess the expression of these symptoms. In addition to compare the pain characteristics of migraine patients with and without U-CAS.	3.1^b

CAS: cranial autonomic symptoms; CM: Chronic migraine; EM: Episodic migraine; U-CAS: Unilateral cranial autonomic symptoms.

^a the number refers to the prevalence of CAS ipsilateral to the headache.

^b the number refers to the prevalence of unilateral CAS.

^c the Turkish headache database was created to investigate headache in the Turkish population, but it was not possible to identify if the participants were recruited from the general population or from, for example, a specialized headache center.

The correlation between chronicity of migraine and high prevalence of CAS is not clear from the current literature. On the one hand, Togha et al. (19) found a significantly higher proportion of CM among participants with CAS compared with participants without CAS. On the other hand, Ceylan et al. (15) examined CAS prevalence separately in patients with episodic migraine (EM) and CM and found no significant difference. Other studies also included both EM and CM and found no significant correlation between CAS and number of headache days (3,5,10,16). In general this literature contrasts with expert clinical experience that few patients spontaneously complain of CAS, and that such symptoms seem to be mild when specifically asked for.

Comparison of migraine with CAS and cluster headache

In this study we report that migraine with CAS is prevalent in the general population, however, we did not assess the severity of CAS and the proportion of migraine attacks with CAS. Lai et al. (3) assessed and characterized CAS in migraine patients compared with CAS in patients with cluster headache (CH). The authors reported that 437/786 (56%) of migraine patients experienced CAS, while 95/98 (97%) of CH patients experienced CAS during attacks. The migraine patients had a significantly lower number of CAS items, with a mean of 1.8 CAS item compared with CH patients, who had a mean of 3.5 CAS items (3). Seven other studies reported that between 27–53% of migraine patients only had one CAS item per migraine attack (4,5,7,10,17,21,23). These findings support our proposal that the diagnostic criteria for migraine with CAS should require two or more CAS items to prevent an overestimation of the prevalence.

CAS in migraine patients is more often bilateral than CAS in CH patients. Thus, 67-95% of migraine patients reported bilateral CAS versus 20-28% of CH patients (3). CH patients further reported that the CAS are ipsilateral to the site of headache pain, while 66–84% of migraine patients reported no association between headache site and CAS site (3). Danno et al. (23) also examined the laterality of CAS in participants with migraine and found that 56.9% had bilateral symptoms.

Lai et al. (3) further asked the participants to quantify the intensity of CAS on a scale from one (mild) to three (severe). Severe CAS was reported by 8–22% of migraine patients and by 22–55% of CH participants. Migraine participants experienced CAS less consistently than CH participants. During every migraine attack, 14–27% of migraine participants had CAS, while the same was true for 44–52% of CH participants (3).

Snoer et al. (24) reported the frequency of each CAS item in CH attacks (Table 1). Facial/forehead sweat was the most common CAS in the DaMP cohort but the second least common symptom during CH attacks. In contrast, nasal congestion is much more common during CH attacks than during migraine attacks. In CH, a high proportion of CAS occur either pre- or postictally. Preictal CAS have been reported in 175/500 attacks, ictal CAS have been reported in 408/500 attacks, and postictal CAS have been reported in 204/500 attacks (24). In migraine, CAS are in general thought to occur at the peak of pain intensity and not before the pain (16,26), and the presence of CAS is associated with more severe headache pain (3 –5,10,19). Nevertheless, CAS have also been reported as premonitory symptoms in provoked attacks of migraine and CH (27 –30), suggesting that pain may not be necessary for their presence. Persistent Horner-like syndrome has been noted in CH patients (31,32), but there are no reports of chronic Horner-like syndrome in migraine patients in the literature.

Proposed diagnostic appendix criteria for migraine with CAS

We propose two sets of diagnostic appendix criteria for migraine with CAS. One set for genetic and epidemiological studies, to be used in self-reported questionnaires for example, (Table 4) and another set for clinical and pathophysiological studies (Tables 5). So far, no diagnostic appendix criteria for migraine with CAS have been proposed. While we do not recommend to generally subdivide migraine according to CAS, diagnostic appendix criteria are needed for clinical, epidemiological, and genetic studies. In genetic and epidemiological studies high numbers are essential. These are only achieved by self-reported questionnaires, where a lower diagnostic specificity and sensitivity must be accepted. Our validation study showed that questionnaire diagnostic of CAS is possible, and that specificity increases by demanding two or more CAS items (80%) as opposed to one CAS item (66%). Therefore, we require a minimum of two CAS items in questionnaire-based studies.

The situation is different when it comes to studies exploring the possible significance of CAS in pathophysiological or therapeutic studies. It is still uncertain if CAS are just a reflection of pain or whether migraine with CAS signifies an endophenotype worthy of separate study. For the further study of the significance of CAS it is important to first study a core group with marked CAS. Therefore, we also present stricter diagnostic appendix criteria for such studies. Thus, we propose that data for clinical studies should be collected by semi-structured interview and not by self-reported questionnaire. Here, we go beyond the data of our own study by requesting that CAS should be present in at least one out of three attacks. This should be further tested, and another frequency may perhaps prove better. Another possibility to improve the proposed criteria is to include degrees of severity of the individual CAS items. Again, this is a question for future investigation. We do not recommend grouping the autonomic symptoms according to sympathetic and parasympathetic mechanisms, since the SP and SE were slightly lowered with such grouping (Table 2). Furthermore, none of the symptoms were strongly correlated (Figure 1) and we did not detect a strong cluster pattern between symptoms (Supplementary Table 2).

We have included the symptom eyelid oedema in the proposed diagnostic criteria based on the literature, since eyelid oedema is a well described autonomic symptom among patients with migraine (Supplementary Table 3) (3 –5,7,10,15 –23), although, we did not have data on this symptom in our cohort. We believe that the diagnostic criteria suggested here are valid and useful. We suggest, however, that they may be further improved by prospective field testing.

Diagnostic questionnaire vs semi-structured interview

The prevalence of migraine with CAS was high using a diagnostic questionnaire (31%) compared with a semi-structured interview (22%). Comparing retrospective headache studies with prospective headache diaries there is also an overestimation of the severity of headache and number of CAS items (33,34). This is probably because of a tendency to recall the worst headache attacks. Our study suggests that this recall bias diminishes using a semi-structured interview, probably because of the confirmatory follow-up questions asked by the interviewer.

The confusion matrices (Table 3) showed that the discrepancy between the diagnostic questionnaire and the semi-structured interview was bidirectional. Fifty-six participants reported to have at least one CAS item in the questionnaire, but did not have any when asked by telephone, and 24 participants who reported not having any CAS item, turned out to have one or more at interview. We conclude that a validated diagnostic questionnaire can be used when the purpose is to identify a large number of cases. For clinical or pathophysiological studies semi-structured interviews should be used to ensure high diagnostic validity.

Limitations of our study

Our study is based on retrospective data which constitutes a risk of recall bias. Our cohort was large, but the validation cohort was of moderate size. Therefore, our confidence intervals widened with increased number of CAS items (Table 2). We did not include eyelid oedema in our diagnostic questionnaire, however, results from other studies (Supplementary Table 3) imply that eyelid oedema is common with a prevalence up to 40% (15). Thus, including this symptom in the diagnostic questionnaire could have resulted in a higher overall prevalence of CAS. We did not estimate consistency or severity of CAS which could have improved the proposed diagnostic criteria. This is an item for future study.

Conclusion

Cranial autonomic symptoms are prevalent among individuals with migraine. It is possible to apply a diagnostic questionnaire to diagnose migraine with CAS with the purpose of genetic and epidemiological studies, where big sample sizes are crucial. In clinical and pathophysiological studies, even higher specificity is needed, and a semi-structured interview should be preferred.

Clinical implications

A precise estimate of the prevalence migraine with cranial autonomic symptoms in a population-based material is important, because migraine with cranial autonomic symptoms might represent an endophenotype, in which genetic and pathophysiological features differ from those without cranial autonomic features.

It is possible to apply a diagnostic questionnaire to diagnose migraine with cranial autonomic symptoms with the purpose of genetic and epidemiological studies, where large sample sizes are crucial.

In clinical and pathophysiological studies, even higher specificity is needed, and a semi-structured interview should be preferred.

Supplemental Material

sj-pdf-1-cep-10.1177_03331024221094548 - Supplemental material for Population-based prevalence of cranial autonomic symptoms in migraine and proposed diagnostic appendix criteria

Supplemental material, sj-pdf-1-cep-10.1177_03331024221094548 for Population-based prevalence of cranial autonomic symptoms in migraine and proposed diagnostic appendix criteria by Charlotte Grønvold Christensen, Tanya Ramdal Techlo, Lisette JA Kogelman, Lise Wegner Thørner, Janna Nissen, Erik Sørensen, Jes Olesen, Thomas Folkmann Hansen, Mona Ameri Chalmer and the DBDS Genomic Consortium in Cephalalgia

Footnotes

Data availability

Individual data are available from the corresponding author upon reasonable request and require both a material transfer agreement and memorandum of understanding in order to obtain ethical and data protection agency approval.

Declaration of conflicting interest

The authors declare that there is no conflict of interest with respect to the research, authorship, and/or publication of this study.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The study was financed by a grant from Candys Foundation, the CEHEAD program (JO) and The Research Funding Pool at Rigshospitalet (MAC and CC). The funding sources had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Acknowledgements

We would like to thank the Danish blood donors for their valuable participation in the Danish Blood Donor Study as well as the staff at the blood centers for making this study possible. We thank Rigmor Jensen and Agneta Snoer for their valuable inputs.

ORCID iDs

Lisette JA Kogelman

Mona Ameri Chalmer

Supplemental material

Supplemental material for this article is available online.

References

Headache Classification Committee of the International Headache Society (IHS) The International Classification of Headache Disorders, 3rd edition. Cephalalgia 2018; 38: 1–211.

Blau

Davis

Small blood-vessels in migraine. Lancet 1970; 2: 740–742.

Lai

Fuh

Wang

SJ.

Cranial autonomic symptoms in migraine: Characteristics and comparison with cluster headache. J Neurol Neurosurg Psych 2009; 80: 1116–1119.

Obermann

Yoon

M-S

Dommes

, et al. Prevalence of trigeminal autonomic symptoms in migraine: a population-based study. Cephalalgia 2007; 27: 504–509.

Barbanti

Fabbrini

Pesare

, et al. Unilateral cranial autonomic symptoms in migraine. Cephalalgia 2002; 22: 256–259.

Uluduz

Ayta

Özge

, et al. Cranial autonomic features in migraine and migrainous features in cluster headache. Noro Psikiyatr Ars 2016; 55: 220–224.

Riesco

Pérez-Alvarez

Verano

, et al. Prevalence of cranial autonomic parasympathetic symptoms in chronic migraine: Usefulness of a new scale. Cephalalgia 2016; 36: 346–350.

Lipton

Fanning

Buse

, et al. Identifying natural subgroups of migraine based on comorbidity and concomitant condition profiles: Results of the Chronic Migraine Epidemiology and Outcomes (CaMEO) Study. Headache 2018; 58: 933–947.

Dick

Jones

Saccone

, et al. Endophenotypes successfully lead to gene identification: Results from the collaborative study on the genetics of alcoholism. Behav Genet 2006; 36: 112–126.

10.

Barbanti

Aurilia

Dall’Armi

, et al. The phenotype of migraine with unilateral cranial autonomic symptoms documents increased peripheral and central trigeminal sensitization. A case series of 757 patients. Cephalalgia 2016; 36: 1334–1340.

11.

Chalmer

Callesen

Kogelman

LJA

, et al . Reappraisal of sex differences in migraine. SSRN Electron J. Epub ahead to print 10 Feb 2022. DOI:10.2139/ssrn.4031580.

12.

Pedersen

Erikstrup

Kotzé

, et al. The Danish Blood Donor Study: A large, prospective cohort and biobank for medical research. Vox Sang 2012; 102: 271.

13.

Huang

Extensions to the k-means algorithm for clustering large data sets with categorical values. Data Min Knowl Discov 1998; 2: 283–304.

14.

Weihs C, Ligges U, Luebke K, Raabe N. klaR Analyzing German Business Cycles. In: Baier D, Decker R, Schmidt-Thieme L (eds). Data Analysis and Decision Support. Studies in Classification, Data Analysis, and Knowledge Organization. Berlin, Heidelberg: Springer, 2005, pp.335–343. https://doi.org/10.1007/3-540-28397-8_36.

15.

Ceylan

Yalcin

Coexistence of symptoms associated with trigeminal pathways in chronic and episodic migraine and the effects on quality of Life. Pain Med 2019; 20: 172–179.

16.

Gupta

Bhatia

A report of cranial autonomic symptoms in migraineurs. Cephalalgia 2007; 27: 22–28.

17.

Guven

Çilliler

Çomoǧlu

SS.

Unilateral cranial autonomic symptoms in patients with migraine. Acta Neurol Belg 2013; 113: 237–242.

18.

Karsan

Nagaraj

Goadsby

PJ.

Cranial autonomic symptoms: prevalence, phenotype and laterality in migraine and two potentially new symptoms. J Headache Pain 2022; 23: 18.

19.

Togha

Jafari

Moosavian

, et al. Cranial autonomic symptoms in episodic and chronic migraine: a cross sectional study in Iran. BMC Neurol 2021; 21: 493.

20.

Rozen

TD.

A history of cigarette smoking is associated with the development of cranial autonomic symptoms with migraine headaches. Headache 2011; 51: 85–91.

21.

Fatima

Asghar

Ali

, et al. Frequency of cranio-autonomic symptoms in the patients of migraine presenting in an outpatient department of Dow university hospital: A cross-sectional study. Rom J Neurol 2021; 20: 81–87.

22.

Nasir

Abrar

Sher

, et al. Frequency of cranial autonomic symptoms (cas) in migraine patients. Pakistan J Neurol Sci 2016; 11: 21.

23.

Danno

Wolf

Ishizaki

, et al. Cranial autonomic symptoms of migraine in Japan: Prospective study of 373 migraine patients at a tertiary headache center. Headache 2020; 60: 1592–1600.

24.

Snoer

Lund

Beske

, et al. Cluster headache beyond the pain phase: A prospective study of 500 attacks. Neurology 2018; 91: e822–e831.

25.

Zheng

Tian

The impact of diagnostic error on testing genetic association in case-control studies. Stat Med 2005; 24: 869–882.

26.

Al-Din

Mir

Davey

, et al. Trigeminal cephalgias and facial pain syndromes associated with autonomic dysfunction. Cephalalgia 2005; 25: 605–611.

27.

Karsan

Bose

Thompson

, et al. Headache and non-headache symptoms provoked by nitroglycerin in migraineurs: A human pharmacological triggering study. Cephalalgia 2020; 40: 828–841.

28.

Snoer

Lund

Beske

, et al. Pre-attack signs and symptoms in cluster headache: Characteristics and time profile. Cephalalgia 2018; 38: 1128–1137.

29.

Pedersen

Snoer

Barloese

, et al. Prevalence of pre-cluster symptoms in episodic cluster headache: Is it possible to predict an upcoming bout? Cephalalgia 2021; 41: 799–809.

30.

Blau

Engel

HO.

Premonitory and prodromal symptoms in cluster headache. Cephalalgia 1998; 18: 91–93.

31.

Nieman

Hurwitz

LJ.

Ocular sympathetic palsy in periodic migrainous neuralgia. J Neurol Neurosurg Psychiatry 1961; 24: 369–373.

32.

Horton

BT.

Histaminic cephalgia: differential diagnosis and treatment. Bull Tufts N Engl Med Cent 1955; 1: 143–156.

33.

Snoer

Lund

Jensen

, et al. More precise phenotyping of cluster headache using prospective attack reports. Eur J Neurol 2019; 26: 1303–e85.

34.

McKenzie

Cutrer

FM.

How well do headache patients remember? A comparison of self-report measures of headache frequency and severity in patients with migraine. Headache 2009; 49: 669–672.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.18 MB