Sleep apnea in patients with cluster headache: A case-control study

Abstract

Objective

Polysomnographic investigations have shown an unspecific association between cluster headache and obstructive sleep apnea syndrome. The aim of this study was to investigate this association in a cluster episode compared with a symptom free interval, and to further characterize this association.

Methods

We investigated 42 patients with episodic (n = 26) or chronic (n = 16) cluster headache by means of polygraphic screening for sleep apnea and compared the data to 28 healthy control subjects matched according to age, sex, and BMI. The patients with episodic cluster headache were screened twice, once in a cluster episode and once in a symptom free interval.

Results

Patients with active cluster headache showed a significantly higher respiratory distress index (8.6 ± 16.0) compared with healthy control subjects (3.4 ± 2.1; p = 0.002). More patients fulfilled the criteria for an obstructive sleep apnea syndrome (29%) than control subjects (7%; p = 0.018). Patients only, but not the control subjects, had central apneas. These differences were only significant when measured during an active cluster episode but not during a symptom free interval.

Conclusion

Cluster headache is associated with a sleep apnea syndrome only in the active cluster episode. The increased rate of central apneas might be a result of involvement of the hypothalamus in the pathophysiology of cluster headache. Out of five anecdotal cases treated with nasal continuous positive airway pressure, only one patient showed benefit with respect to cluster headache attack frequency.

Keywords

Cluster headache obstructive sleep apnea syndrome central sleep apnea continuous positive airway pressure treatment

Introduction

The hypothalamus plays a pivotal role in the pathophysiology of cluster headache (1). There is strong evidence that cluster headache is a biorhythmic disorder and, in particular, is linked to sleep. Sleep disorders such as insomnia (2) seem to be common among patients with cluster headache. Previous polysomnographic investigations showed an association between both the occurrence of cluster headache attacks during REM sleep and an association with obstructive sleep apnea syndrome (OSAS) (3 –16). The association of cluster headache attacks with REM sleep, however, is controversial; recent polysomnographic studies could not show such an association (17,18). Most of these studies did not include healthy control subjects and did not differentiate between different types of sleep apnea and the different episodes of cluster headache.

The aim of this study was to investigate the association between cluster headache and sleep apnea syndrome in a cluster episode compared with a symptom free interval and healthy control subjects. Furthermore, we aimed to characterize the sleep apneas in cluster headache to conclude on the pathophysiological mechanisms. Therefore, we matched the patients and the healthy control subjects for body mass index (BMI) to control for this variable, which is an important factor in the pathophysiology of OSAS. Finally, we offered patients treatment by nasal continuous positive airway pressure (nCPAP) if they showed OSAS to evaluate the efficacy of this treatment on cluster headache. However, this offer was not part of the study protocol but a routine clinical procedure.

Patients and methods

Patients

We enrolled consecutive patients with active cluster headache by means of polygraphic screening for sleep apnea (MESAM). As a control group, we included healthy subjects, matched according to age, sex, and BMI. Attempts were made to screen patients with episodic cluster headache twice, once in a cluster episode and once in a symptom free interval. All patients were recruited from the headache outpatient clinic of the Department of Neurology, University of Münster, Germany, and from the Department of Neurology, Krankenhaus Lindenbrunn, Germany. They gave informed consent before inclusion into the study. The diagnosis of cluster headache was made according to the International Headache Society (IHS) criteria, 2nd edition (19).

Data registration

Before onset of sleep recording, patients filled in the Epworth sleepiness scale (20). Then, the MESAM device was provided and explained to the patients. We registered the following biosignals: flow at mouth and nose; oxygen saturation; thoracic and abdominal expansion (breathing excursion); heart rate; electrocardiogram; movements/body position; pressure of CPAP (if applicable). Ten patients used the MESAM device in the hospital as inpatients. The remaining patients and all control subjects used the device at home.

We evaluated hypopneas and apneas. A hypopnea was defined as an incomplete decrease of nasal and/or oral flow for more than 10 seconds in combination with a reduction of tidal volume by more than 50%. An apnea was defined as a complete stop of nasal and oral flow with a reduction of tidal volume by at least 20% for more than 10 seconds combined with a reduction of oxygen saturation by at least 4%. A sleep apnea syndrome was diagnosed when the number of hypopneas and apneas was more than five per hour (so called respiratory distress index, RDI) during sleep. Furthermore, we differentiated between obstructive and central apneas. An obstructive apnea was defined as a stop of nasal and oral flow combined with maintained thoracic and/or diaphragmal breathing activity. A central apnea was defined as a stop of nasal and oral flow with missing activity in the thoracic and diaphragmal muscles. Also, the desaturation index was evaluated (i.e. decrease of oxygen saturation by 4% or more for at least 10 seconds) (21).

The data curves registered by the device were visually re-analyzed. Then the following data were included in the analysis: RDI (sum of apnea and hypopnea index); apnea index (separately for obstructive and central apneas); hypopnea index; desaturation index; mean and minimal oxygen saturation; heart rate; time of oxygen saturation under 90% (Ct_90); percentage of patients with obstructive sleep apnea syndrome (i.e. RDI for obstructive apneas >5/h); percentage of central sleep apnea syndrome (i.e. RDI for central apneas >5/h).

Statistics

For statistical analysis, the program SPSS 15.0 was used. Data are presented as arithmetic mean with standard deviation or as percentage. We used non-parametric testing. The Mann-Whitney U test and the Wilcoxon test were carried out to analysis differences between quantitative data. Chi-square test and Fisher’s exact test (if applicable) were used for the analysis of qualitative data. Significance level was set at p = 0.05.

Results

We enrolled 42 patients with active cluster headache and 28 control subjects. The demographic and clinical characteristics of the patients and of the healthy control subjects are shown in Table 1. Matching was done by age, sex, and BMI; therefore, these parameters were not significantly different between the patient and control groups. Among the patients with cluster headache, the percentage of smokers was significantly higher, as was the number of pack years (smokers only). However, smoking status and pack years were not related to any of the polygraphic measures (data not shown).

Table 1.

Clinical and demographic data of patients with cluster headache and healthy control subjects.

	Patients (n = 42)	Control subjects (n = 28)	p
Age (years)	45 ± 9	47 ± 8	0.7000
Sex (male/female)	40/2	25/3	0.299
Body mass index (kg/m²)	26 ± 4	25 ± 6	0.566
Smoker	86%	21%	0.014
Pack years (smokers only)	34.3 ± 24.4	11.5 ± 15.0	0.019
Attacks per day	2.7 ± 1.6	na
Age at onset (years)	33.0 ± 11.4	na
Chronic cluster headache	38%	na

Values are given as mean ± standard deviation unless otherwise specified. p Values were calculated according to Mann-Whitney U test, and Chi-square test or Fisher’s exact test for qualitative parameters.

In Table 2, sleep data are presented for all recordings of active cluster headache patients (i.e. during the episode and chronic) and for the healthy control subjects. Patients with cluster headache showed a significantly higher RDI (caused by all types of apnea) compared with healthy controls, as well as a higher desaturation index. Significantly more patients than control subjects fulfilled the criteria for OSAS. Five patients but no control subject had central apneas.

Table 2.

Sleep parameters for all patients with active cluster headache (i.e. cluster episode and chronic cluster) and for healthy control subjects.

	Active cluster (n = 42)	Control subjects (n = 28)	p
Respiratory distress index	8.6 ± 16.0	3.4 ± 2.1	0.002
Obstructive apnea index	2.3 ± 7.1	0.6 ± 0.9	0.088
Central apnea index	1.1 ± 2.4	–	0.009
Hypopnea index	5.3 ± 9.4	3.0 ± 2.1	0.002
Desaturation index	7.3 ± 13.5	1.2 ± 1.3	0.012
Mean O₂-saturation (%)	94.2 ± 1.3	95.2 ± 1.0	0.357
Minimal oxygen saturation (%)	86.1 ± 3.8	86.0 ± 7.2	0.987
Ct_90 (%)	9.1 ± 13.7	1.3 ± 2.5	0.014
Obstructive sleep apnea syndrome	29%	7%	0.018
Central sleep apnea syndrome	12%	–	na
Epworth sleepiness scale	5.6 ± 2.3	4.9 ± 3.5	0.573

p Values were calculated according to Mann-Whitney U test, and Chi-square test or Fisher’s exact test for qualitative parameters.

Table 3 presents the comparison of episodic cluster headache patients during and outside the episode only for those patients from whom both measures were available. There were significantly higher RDI, obstructive apneas, central apneas, and desaturation index during the episode. Also, the diagnosis of OSAS was significantly more frequent during the episode than in the interval. Central apneas exclusively occurred during the episode. The data of the cluster headache patients during the interval were very similar to the data of the healthy control subjects.

Table 3.

Sleep parameters for all patients with episodic cluster headache who could be examined during and outside the episode (n = 14).

	During episode	Outside episode	p
Respiratory distress index	6.7 ± 16.0	2.1 ± 5.2	0.045
Obstructive apnea index	2.5 ± 6.6	–	0.016
Central apnea index	1.0 ± 3.6	–	0.038
Hypopnea index	3.2 ± 6.0	2.1 ± 5.2	0.366
Desaturation index	6.6 ± 13.9	2.8 ± 2.3	0.018
Mean O₂-saturation (%)	95.0 ± 1.3	95.4 ± 1.3	0.179
Minimal oxygen saturation (%)	87.2 ± 2.8	89.4 ± 2.1	0.047
Ct_90 (%)	0.7 ± 1.7	0.5 ± 1.2	0.824
Obstructive sleep apnea syndrome	29%	7%	0.016
Central apnea syndrome	14%	–	na
Epworth sleepiness scale	5.0 ± 2.4	8.1 ± 4.2	0.188

p Values were calculated according to Wilcoxon test, and Chi-square test or Fisher’s exact test for qualitative parameters.

Table 4 presents the comparison between episodic cluster headache patients in the episode and chronic cluster headache patients. Except for Ct_90 percentage and hypopnea index, no significant differences could be detected. For some parameters, however, trends towards more pathological values in chronic cluster headache could be observed suggesting that apnea syndromes are more severe in chronic cluster headache than in episodic.

Table 4.

Sleep parameters for episodic cluster headache patients during the episode versus chronic cluster headache patients.

	Episodic cluster (n = 26)	Chronic cluster (n = 16)	p
Respiratory distress index	7.6 ± 15.1	10.3 ± 22.0	0.080
Obstructive apnea index	2.5 ± 7.1	2.0 ± 5.3	0.188
Central apnea index	1.1 ± 2.4	1.0 ± 3.7	0.909
Hypopnea index	4.0 ± 9.4	7.4 ± 17.3	0.048
Desaturation index	6.5 ± 11.5	8.0 ± 12.9	0.745
Mean O₂-saturation (%)	94.7 ± 1.3	93.4 ± 2.6	0.357
Minimal oxygen saturation (%)	86.8 ± 3.8	84.9 ± 4.5	0.487
Ct_90 (%)	3.4 ± 6.7	18.3 ± 29.2	0.014
Obstructive sleep apnea syndrome	27%	31%	0.098
Central sleep apnea syndrome	8%	19%	0.084
Epworth sleepiness scale	5.2 ± 1.7	6.2 ± 2.1	0.078

p Values were calculated according to Wilcoxon test, and Chi-square test or Fisher’s exact test for qualitative parameters.

We also evaluated the data of MESAM recordings during a night with cluster headache attacks versus no cluster headache attacks. There were no significant differences in all parameters (data not shown).

Five patients (two episodic, three chronic) were willing to undergo nCPAP treatment during the night for 1 week. They were asked to record the number of cluster headache attacks during this week, which was compared with the average number of cluster headache attacks per week during the baseline. There was only a mild decrease from 18.4 ± 2.8 to 14.0 ± 2.7. Only one patient reported a meaningful reduction from 21 to 9 attacks per week, the other four patients did not report any subjective benefit. As the number of patients was small and this treatment was not part of the study protocol, we did not perform any further statistical analysis.

Discussion

Our results confirm an association of active cluster headache with sleep apnea syndrome. This association can be observed in the cluster headache episode but not in the cluster headache interval. There were some trends suggesting that the apnea syndrome is more severe in chronic cluster than in active episodic cluster headache. Furthermore, we could show that cluster headache patients not only have obstructive apneas, but also central apneas.

We matched our patients according to BMI, because we wanted to exclude this factor in the analysis. A high BMI is one of the major reasons for OSAS, and is often observed in cluster headache patients. It has also been shown to be one of the major risk factors for OSAS even in cluster headache patients (11). Our finding of a significantly increased RDI and of other pathological sleep parameters only during the cluster episode but not in the interval, suggests a specific role of the cluster headache pathophysiology in the occurrence of sleep apneas. Therefore, we do not believe that apneas in cluster headache are the consequence only of a high BMI or of increased rate of smoking.

The finding of central apneas in cluster headache is also an argument for a specific role of cluster headache pathophysiology in the occurrence of sleep apneas in this disorder (16). To our knowledge, we could demonstrate for the first time that cluster headache patients not only have obstructive sleep apneas, but also central apneas which are the result of a central dysregulation of ventilation. It is likely that the hypothalamic functions and, in particular, orexin metabolism, which have been shown to be involved in cluster headache as the major regulation mechanisms (22), are also dysfunctional in regulation of sleep in cluster headache patients. In fact, observational studies have recently shown that deep brain stimulation of the posterior hypothalamus not only decreases the severity of cluster headache, but also regulates sleep pattern in refractory cluster headache patients (23,24).

In our study, only 29% of the patients fulfilled the criteria for OSAS, very similar to two other studies (9,14), whereas different studies reported up to 80% of the cluster patients having an RDI higher than five (7,10). It is important to consider whether somnographic measures were performed during the cluster episode or during the cluster interval, and how the evaluation of OSAS was defined, before comparing data from different studies on sleep apnea in cluster headache. Furthermore, if we add the central apnea syndrome patients to the OSAS patients, this results in about 40% of all patients having a sleep apnea syndrome during active cluster headache.

Some previous studies suggested that nocturnal nCPAP treatment can lead to an improvement of cluster headache (6 –8,25); other studies did not find a relevant improvement of chronic headache types, including cluster headache with OSAS by nCPAP treatment (14). Prospective controlled studies, however, are still missing.

Interestingly, the Epworth sleepiness scale score was not different between cluster headache patients and healthy control subjects, and between the different cluster headache phases. This means that cluster headache does not go along with increased daytime sleepiness, although cluster headache is associated with insomnia (2). This also points to a specific sleep regulation in cluster headache.

The strength of our study is the comparison of active versus non-active cluster headache, which has been shown to be the most relevant factor for the association of cluster headache and sleep apneas. Another strength is the careful evaluation of apnea phases, making it possible to differentiate between obstructive and central apneas. Our study has also some limitations. First, we controlled for BMI to exclude this anatomic factor as a confounder on the occurrence of sleep apnea, but not for other confounders such as alcohol consumption. Second, we only measured one night and not two nights as is often recommended in sleep studies. It might be that we did not diagnose all patients with sleep apneas or that we diagnosed some patients with sleep apnea who normally do not have sleep related breathing disorders. Third, we did not control for smoking, which was found to be significantly different between cluster headache patients and control subjects in our study, not only with respect to frequency but also to amount of pack years. Smoking could be an additional risk factor for sleep apneas and explain at least a part of the increased frequency of sleep apneas in cluster headache. Finally, we did not perform complete polysomnography in our patients. Therefore, we are not able to analyze the sleep stages and the association with REM sleep or other events during sleep.

Clinical implications

Cluster headache is associated with both obstructive sleep apnea syndrome and central sleep apnea syndrome.

The association is only detectable during the cluster episode or in chronic cluster headache but not in the cluster interval.

There are merely differences between active episodic cluster headache and chronic cluster headache with respect to sleep parameters.

We found no evidence that nCPAP treatment is efficacious in cluster headache associated with OSAS.

Footnotes

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Conflict of interest

None declared.

References

Goadsby

. Pathophysiology of cluster headache: A trigeminal autonomic cephalgia. Lancet Neurol 2002; 1: 251–257.

Ofte

Berg

Bekkelund

Alstadhaug

. Insomnia and periodicity of headache in an arctic cluster headache population. Headache 2013; 53: 1602–1612.

Manzoni

Terzano

Moretti

Cocchi

. Clinical observations on 76 cluster headache cases. Eur Neurol 1981; 20: 88–94.

Kudrow

McGinty

Phillips

Stevenson

. Sleep apnea in cluster headache. Cephalalgia 1984; 4: 33–38.

Pfaffenrath

Pöllmann

Rüther

. Onset of nocturnal attacks of chronic cluster headache in relation to sleep stages. Acta Neurol Scand 1986; 73: 403–407.

Buckle

Kerr

Kryger

. Nocturnal cluster headache associated with sleep apnea. A case report. Sleep 1993; 16: 487–489.

Chervin

Nath Zallek

Lin

. Sleep disordered breathing in patients with cluster headache. Neurology 2000; 54: 2302–2306.

Lüdemann

Frese

Happe

Evers

. Sleep disordered breathing in patients with cluster headache. Neurology 2001; 56: 984–984.

Nobre

Filho

Dominici

. Cluster headache associated with sleep apnoea. Cephalalgia 2003; 23: 276–279.

10.

Graff-Radford

Newman

. Obstructive sleep apnea and cluster headache. Headache 2004; 44: 607–610.

11.

Nobre

Leal

Filho

. Investigation into sleep disturbance of patients suffering from cluster headache. Cephalalgia 2005; 25: 488–492.

12.

Pelin

Bozluolcay

. Cluster headache with obstructive sleep apnea and periodic limb movements during sleep: A case report. Headache 2005; 45: 81–83.

13.

Della Marca

Vollono

Rubino

. A sleep study in cluster headache. Cephalalgia 2006; 26: 290–294.

14.

Mitsikostas

Vikelis

Viskos

. Refractory chronic headache associated with obstructive sleep apnoea syndrome. Cephalalgia 2008; 28: 139–143.

15.

Ranieri

Tufik

de Siqueira

. Refractory cluster headache in a patient with bruxism and obstructive sleep apnea: a case report. Sleep Breath 2009; 13: 429–433.

16.

Graff-Radford

Teruel

. Cluster headache and obstructive sleep apnea: Are they related disorders? Curr Pain Headache Rep 2009; 13: 160–163.

17.

Terzaghi

Ghiotto

Sances

. Episodic cluster headache: NREM prevalence of nocturnal attacks. Time to look beyond macrostructural analysis? Headache 2010; 50: 1050–1054.

18.

Zaremba

Holle

Wessendorf

. Cluster headache shows no association with rapid eye movement sleep. Cephalalgia 2012; 32: 289–296.

19.

Headache Classification Committee of the International Headache Society. The International classification of headache disorders: 2nd edition. Cephalalgia 2004; 24(Suppl 1): 9–160.

20.

Johns

. A new method for measuring daytime sleepiness: The Epworth sleepiness scale. Sleep 1991; 14: 540–545.

21.

Rechtschaffen

Kales

. A manual of standardized terminology, techniques and scoring systems for sleep stages of human subjects, Washington, DC: Public Health Service, US, Government Printing Office, 1968.

22.

Holland

Goadsby

. Cluster headache, hypothalamus, and orexin. Curr Pain Headache Rep 2009; 13: 147–154.

23.

Vetrugno

Pierangeli

Leone

. Effect on sleep of posterior hypothalamus stimulation in cluster headache. Headache 2007; 47: 1085–1090.

24.

Kovac

Wright

Eriksson

. The effect of posterior hypothalamus region deep brain stimulation on sleep. Cephalalgia 2014; 34: 219–223.

25.

Nath Zallek

Chervin

. Improvement in cluster headache after treatment for obstructive sleep apnea. Sleep Med 2000; 1: 135–138.