Acta Neurol Scand 2014: 129 (Suppl. 198): 47–54 DOI: 10.1111/ane.
12237 © 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Sleep quality and arousal in migraine and
tension-type headache: the headache-sleep
study
Engstrøm M, Hagen K, Bjørk MH, Stovner LJ, Sand T. Sleep quality
and arousal in migraine and tension-type headache: the headache-sleep
study.
Acta Neurol Scand: 2014: 129 (Suppl. 198): 47–54.
© 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd.
Objectives – The present paper summarizes and compares data from
our studies on subjective and objective sleep quality and pain
thresholds in tension-type headache (TTH), migraine, and controls.
Material and methods – In a blinded controlled explorative study, we
recorded polysomnography (PSG) and pressure, heat, and cold pain
thresholds in 34 controls, 20 TTH, and 53 migraine patients. Sleep
quality was assessed by questionnaires, sleep diaries, and PSG.
Migraineurs who had their recordings more than 2 days from an
attack were classified as interictal while the rest were classified as
either preictal or postictal. Interictal migraineurs (n = 33) were also
divided into two groups if their headache onsets mainly were during
sleep and awakening (sleep migraine, SM), or during daytime and no
regular onset pattern (non-sleep migraine, NSM). TTH patients were
divided into a chronic or episodic group according to headache days
per month. Results – Compared to controls, all headache groups
reported more anxiety and sleep-related symptoms. TTH and NSM
patients reported more daytime tiredness and tended to have lower
pain thresholds. Despite normal sleep times in diary, TTH and NSM
had increased slow-wave sleep as seen after sleep deprivation.
Migraineurs in the preictal phase had shorter latency to sleep onset
than controls. Except for a slight but significantly increased
awakening index SM, patients differed little from controls in objective
measurements. Conclusions – We hypothesize that TTH and NSM
patients on the average need more sleep than healthy controls. SM
patients seem more susceptible to sleep disturbances. Inadequate rest
might be an attack-precipitating- and hyperalgesia-inducing factor.
Introduction
Sleep and headache are closely related (1–3).
Many people have experienced that pain/head-
ache can disturb sleep and conversely. Inadequate
sleep can increase the risk of headache in general
(2), can release headache (4), and can also reduce
pain thresholds (5). Reduced pain threshold is
found among headache patients compared to
controls (6, 7). In migraine, several studies sug-
gest that CNS excitability is changed in the preic-
tal phase compared to the interictal phase, for
example, for thermal PT (8). Detected similarities
ACTA NEUROLOGICA
SCANDINAVICA
M. Engstrøm1,2, K. Hagen1,3,
M. H. Bjørk4,5, L. J. Stovner1,3,
T. Sand1,2
1
Department of Neuroscience, Norwegian University of
Science and Technology, Trondheim, Norway;
2
Department of Neurology and Clinical
Neurophysiology, St. Olavs Hospital, Trondheim,
Norway; 3Norwegian National Headache Centre, St.
Olavs Hospital, Trondheim, Norway; 4Department of
Neurology, Haukeland University Hospital, Bergen,
Norway; 5Department of Clinical Medicine, University
of Bergen, Bergen, Norway
Key words: migraine; tension-type headache; sleep;
polysomnography; arousal; pain thresholds
M. Engstrøm, Department of Neuroscience, PB 8905,
MTFS, Norwegian University of Science and
Technology, N-7489 Trondheim, Norway
Tel.: +47 72 57 50 90
Fax: +47 73 59 87 95
e-mail: morten.engstrom@ntnu.no
Accepted for publication November 10, 2013
and differences between controls, tension-type
headache (TTH) patients, and migraineurs could
be useful to clarify the relation between sleep,
headache, and pain thresholds. As sleep distur-
bance is reported as a trigger for migraine and
TTH, it may be of interest to study if sleep
macro- and microstructure are disturbed in the
headache-free phase, also compared to the preic-
tal phase in migraine. If disturbed sleep is associ-
ated with PT reductions, presumably reflecting a
central sensitization, this may clarify whether a
sleep disturbance could be investigated further as
a potential modifiable risk factor to achieve
47
Engstrøm et al.
better headache prevention in future studies.
Another specific hypothesis was that patients with
nightly attacks have more disturbed sleep than
those without nightly predilection. The pres-
ent paper summarizes results on sleep and pain
in a comprehensive controlled and blinded cross-
sectional study. Data have been reported as three
different papers (9–11) on sleep and pain thresh-
olds among healthy controls, TTH, and migraine
patients. In addition, we also compare sleep in
TTH and migraine patients for the first time in
the present paper. As far as we know, no others
have performed a blinded study and evaluated
both sleep and pain thresholds in headache
patients.
Methods
Participants
The method is described in detail elsewhere (9–11).
One hundred and twenty-six persons, 85 women
and 41 men, (age range 18–64, mean age
38.9 years), including 41 healthy controls, 24
TTH, and 61 migraine patients participated in this
study (Fig. 1). They were mainly recruited by
advertising in local newspapers for people between
18 and 65 years with and without headache.
Potential participants with headache were diag-
nosed according to the ICHD-II criteria (12).
Patients were selected if they had either (frequent)
episodic TTH (1–14 headache days per month),
chronic TTH (≥15 headache days per month), or
migraine with two to six episodes per month with
or without aura. Based on headache diaries, the
date of the sleep recordings and pain threshold
measurements migraine were divided into interictal
(more than 2 days from an attack) and the rest
were classified as either preictal or postictal
Figure 1. Numbers of participants in each group in the pres-
ent study.
48
(Fig. 2). Participants with and without aura were
combined because few preictal and postictal
patients were available. Patients were also catego-
rized as having sleep-related migraine (SM) if usual
attack onset was ‘upon awakening’ or during the
night (‘waking me up’). Migraineurs who did not
have SM were classified as having non-sleep-
related migraine (NSM). SM (n = 15) and NSM
(n = 18) were analyzed in the interictal phase.
Three migraine patients with midictal PSG, ful-
filling both preictal and postictal criteria, were
excluded from this analysis. Of 24 TTH patients,
20 were included while four were excluded owing
to technical problems (battery error or lost elec-
trodes, n = 2) or moderate or severe sleep apnea
[apnea–hypopnea index (AHI) > 15 (n = 2)]. To
preserve statistical power, as many persons as
possible were kept in the control group, but to
have comparable age/sex distributions, some
female controls were randomly excluded from the
TTH study.
Pregnancy, major health problems, co-existing
migraine, or frequent tension-type headache
(TTH) were exclusion criteria. Painkillers or trip-
tans for acute migraine were allowed.
The study was approved by the regional ethics
committee, and participants signed an informed
consent before inclusion.
Procedure
In a blinded controlled exploratory study, we
recorded polysomnography. Recording details are
presented elsewhere (9–11). All participants filled
in sleep diaries 2 weeks before and 2 weeks after
measurement of pain thresholds and subsequent
polysomnography. Headache patients also filled
in headache diaries for the same period. Every
subject answered several questionnaires includ-
ing: Epworth Sleepiness Scale (13), questions
adapted from Karolinska Sleep Questionnaire
(14), Pittsburgh Sleep Quality Index (PSQI) (15),
Hospital Anxiety and Depression Subscales (16),
and 10 questions from the Autonomic Symptom
Profile (17, 18). As a supplement to the Epworth
Figure 2. The relationship between the nearest migraine
attack (lightening Z symbol) and the classification of the
recording as interictal, preictal, or postictal.
 The headache-sleep study

Sleepiness Scale, they also answered the question:
‘Do you have bothersome tiredness during day-
time?’(No(0)-Daily(4)). Patients and controls
underwent a full-night ambulatory sleep study
unattended in our patient hotel. The participants
were instructed to go to bed, sleep as normal,
and register lights-off, lights-on, and sleep times
in sleep diaries.
Automatic analysis was applied for leg move-
ments. Respiration was scored according to
modified ‘Chicago criteria’ (19). Sleep staging was
performed according to ‘The AASM Manual for
the scoring of sleep and associated events’ from
2007 (20) with a few exceptions, as described
below. First, fast arousal was defined according
to the AASM manual (20) as an abrupt shift of
EEG frequency (alpha, theta, and/or faster than
16 Hz activity) lasting 3–30 s, separated with at
least 10 s of sleep. We also scored D- and K-
bursts (21). Thermal pain thresholds (difference
from 32°C baseline) and pressure pain thresholds
(algometry) were recorded before the participants
had their PSG equipment mounted. Heat and
cold pain thresholds were measured on thenar
and the medial forehead on both sides. Pressure
pain thresholds were measured at four sites bilat-
erally in a fixed order. We used regional mean
values for heat, cold, and pressure PT.
Technicians and scorers of PSG and pain
threshold measurements were blinded for
diagnoses.
Statistics
Statistical analyses were performed with PASW
statistics v.18 and SYSTAT version 11. Univari-
ate two-group comparisons were made by non-
parametric Mann–Whitney U-tests. Categorical
data were analyzed with Pearson chi-square test
or Fisher’s exact test. The primary comparisons
reported in this paper will be: (i) controls vs
TTH, interictal NSM, and interictal SM, (ii)
between total migraine interictal and preictal
groups, and (iii) TTH vs the whole migraine
group and interictal migraineurs, both sleep- and
non-sleep-related migraine subgroups. Post hoc
we also compared PSG data on sleep TTH and
non-sleep TTH patients, classified in the same
way as migraineurs.
Results
The most important numerical results have been
reported in Tables 1 and 2. TTH as well as inte-
rictal migraineurs (both SM and NSM patients)
reported more anxiety symptoms in HADS (≥5.3
vs ≤3.0, P < 0.01) and higher autonomic index
score (≥5.2 vs ≤4.3, P < 0.001), more symptoms
of insomnia (Karolinska Sleep Questionnaire
insomnia score ≥5.8 vs 3.4, P < 0.01), more
symptoms in PSQIgs (5.9 vs 3.8, P < 0.001), and
more pain-related sleep trouble in PSQI than
controls (Tables 1 and 3). TTH patients had
more symptoms of insomnia and more subjective
tiredness than migraineurs (Table 1).
Both TTH and NSM patients had more slow-
wave sleep than controls (≥104 vs ≤86 min,
P < 0.05), less fast arousals (≥18.3 vs ≤15.5 per
hour, P < 0.05), and more frequent daytime
tiredness (0–4) (≥1.3 vs 0.7 P < 0.01). NSM also
had lower thermal pain thresholds than controls
(heat pain thresholds 11.2°C vs 16.6°C, cold pain
thresholds 16.1°C vs 20.7°C, P < 0.05). Among
TTH, only chronic TTH had lower pressure pain

Table 1 Population, sleep diary, questionnaire, and headache-related data for all patients : counts or mean (SD)

Controls for migraine (n = 34) Controls for TTH (n = 29) Migraine (n = 53) Tension-type headache (n = 20)

Age (years) 39.6 (13.7) 41.2 (13.6) 38.2 (12.0) 40.9 (13.5)
Sex: F/M 20/14 15/14 41/12 11/9
Headache frequency (1–4) NA NA 2.2 (0.7) 3.5 (0.7)5
Headache intensity (1–4) NA NA 2.5 (0.5) 1.5 (0.5)
Headache history duration (years) NA NA 21.6 (13.8) 15.8 (12.2)
Average diary sleep time (hour) 7.3 (0.8) 7.2 (0.8) 7.2 (1.0) 6.8 (1.1)
Long awakenings in diary (no) 0.1 (0.2) 0.1 (0.2) 0.4 (0.8) 0.4 (0.4)
Daytime tiredness frequency (0–4) 0.7 (0.8) 0.7 (0.9) 1.1 (0.9) 1.7 (1.0)6
Insomnia KSQ score (0–16)1 3.4 (2.3) 3.4 (2.4) 6.1 (2.9) 7.9 (2.4)7
Pain-related sleep trouble (1–4)2 3.8 (2.6) 3.9 (2.7) 6.3 (3.2) 7.2 (3.2)
Autonomic index (0–30)4 1.3 (0.7) 1.3 (0.6) 1.9 (1.0) 2.0 (1.0)
PSQIgs (0–21)2 1.5 (1.4) 1.4 (1.4) 6.6 (4.4) 5.3 (3.5)
HADS anxiety score (0–21)3 1.6 (2.1) 1.5 (2.2) 2.3 (2.3) 2.8 (3.0)
Insomnia KSQ score (0–16)1 2.9 (2.6) 3.0 (2.8) 5.4 (3.1) 5.9 (3.0)

1
Sum of four insomnia questions in Karolinska Sleep Questionnaire (KSQ); 2Pittsburgh Sleep Quality Index Global Score; 3Sum of seven questions about anxiety symptoms
during the last week from the Hospital Anxiety and Depression Scale Questionnaire; 4Sum of ten questions about autonomic instability during the last year. Tension-type
headache was compared statistically to migraine: Mann–Whitney U-test: 5P < 0.0005, 6P < 0.05, 7P = 0.006. Significant differences are shown in bold.

49
Engstrøm et al.

Table 2 Polysomnography and pain threshold mean values (SD) for controls, interictal migraineurs, with sleep migraine and non-sleep migraine subgroups, and tension-
type headache

Controls6 (n = 34) Migraine (n = 33) Sleep migraine (n = 15) Non-sleep migraine (n = 18) Tension-type headache (n = 20)

Total sleep time (min) 409 (68) 435 (61) 417 (67) 451 (52) 432 (46)
Sleep efficiency (%) 90.0 (8.1) 91.0 (6.1) 89.4 (7.6) 92.4 (4.2) 91.2 (5.6)
Awakening index (no/h) 0.99 (0.59) 1.27 (0.74) 1.45 (0.84) 1.12 (0.63) 1.1 (0.6)
Stage 1 (min) 27 (19) 32 (15) 35 (17) 29 (13) 29 (15)
Stage 2 (min) 197 (47) 201 (44) 194 (44) 206 (45) 185 (34)
Stage 3 (min) 86 (31) 97 (28) 88 (25)5 104 (28) 107 (21)5
REM (min) 99 (26) 106 (35) 99 (38) 112 (32) 111 (30)
AASM arousal index (per hour)1 18.3 (5.7) 16.3 (9.1) 17.4 (8.6) 15.5 (9.7) 14.5 (4.3)
D-burst index (per hour)2 11.8 (8.0) 9.5 (7.5) 7.3 (5.7) 11.3 (8.3) 7.9 (6.3)
K-burst index (per hour)2 3.0 (3.8) 3.3 (2.5) 2.4 (2.2) 4.0 (2.5) 3.7 (4.5)
Pressure pain threshold (kPa)4 661 (249) 549 (135) 586 (141) 519 (125) 543 (191)
Heat pain threshold (°C)3 13.4 (3.1) 11.7 (3.6) 12.4 (3.3) 11.0 (3.8) 12.7 (3.7)
Cold pain threshold (°C)3 20.8 (6.3) 17.2 (6.9) 18.4 (6.3) 16.2 (7.3) 19.4 (7.4)

1
Fast EEG arousal; 2Slow EEG arousal; 3Expressed as the difference above/below the 32°C baseline; 4Mean from four bilateral sites. Tension-type headache was compared
statistically to migraine, sleep migraine, and non-sleep migraine; 5Mann–Whitney U-test: P = 0.026; 6Control group for migraine. Significant differences are shown in bold.

thresholds than controls (506 vs 678 kPa) Preictal migraineurs had shorter latency to
(P < 0.05) (Table 3). sleep onset than interictal migraineurs (2.0 vs
Non-sleep migraine patients spent more time in 10.3 min, P < 0.01). Thermal pain thresholds
bed (488 vs 453 min, P < 0.05) and had more were lower in interictal migraine compared to
K-bursts (4.0 vs 3.0 per hour, P < 0.05) than con- controls (P < 0.04). Only five sleep TTH patients
trols. SM patients also had less D-bursts (7.3 vs were available. However, also non-sleep TTH
11.8 per hour, P < 0.05), more awakenings (1.45 vs (n = 15) had less fast arousals than sleep TTH
0.99 per hour, P < 0.05), and tended to have more patients (P < 0.05).
N1 sleep than controls (35 vs 27 min, P = 0.05).
Sleep migraine patients had less slow-wave
Discussion
sleep (88 vs 104 min, P < 0.05) and less K-bursts
(2.4 vs 4.0 per hour, P < 0.05) than NSM The major findings of the present study are (i)
patients and less slow-wave sleep than TTH Patients with TTH report more insomnia than
patients (P = 0.026) (Table 2). migraine patients; (ii) Patients with TTH and
migraine report lower sleep quality and more
Table 3 Summed main results for tension-type headache (TTH) patients, insomnia, tiredness, and autonomic symptoms
migraineurs in interictal (MI), and preictal (MP) phases or sleep- and non-sleep than controls; and (iii) In PSG, NSM and TTH
(SM and NSM)1 migraineurs compared to controls (C) patients had more slow-wave (N3) sleep than
Report data
controls, that is, better sleep quality, which seem-
Anxiety symptoms2 C < TTH**/NSM**/SM** ingly stood in contrast to the fact that they also
Insomnia symptoms3 C < TTH***/NSM**/SM** reported more frequent daytime tiredness and
Subjective sleep disturbances4 C < TTH***/NSM*/SM** tended to have reduced pain thresholds. We
Total sleep time (diary) C = TTH/NSM/SM
Subjective daytime tiredness5 C < TTH**/NSM**
hypothesize that the latter findings are explained
Examination data by a relative sleep deficit caused by a higher need
Awake index (>8 Hz, >30 s, per hour) C < SM* of sleep in TTH and NSM patients.
N3, slow-wave sleep, minutes C/SM < TTH**/NSM*
Fast arousals (lasting 3–30 s) C > TTH**/NSM*
Pressure pain thresholds (kPa)6 C > TTH(*)/NSM(*) Subjective and objective sleep quality in controls and headache
Thermal pain thresholds (°C)7 C < NSM* patients
Latency to sleep onset MP < MI**
Increased symptoms of anxiety, insomnia, and
1
Both NSM and SM patients were in the interictal phase; 2Sum of seven ques- subjective tiredness among migraineurs and TTH
tions about anxiety symptoms during the last week from the Hospital Anxiety
and Depression Scale Questionnaire; 3Sum of insomnia questions in Karolinska
patients have also been found by others (1, 22).
Sleep Questionnaire; 4Evaluated by Pittsburgh sleep quality and questions Compared to controls, the reduced subjective
adapted from the Karolinska Sleep Questionnaire; 5Question: Do you have bother- sleep quality in headache patients was most
some tiredness during daytime? (0(no)-4(daily)); 6Mean value of m. temporalis, m. striking. PSG sleep parameters differed, and
splenius, m. trapezius, distal dorsal middle finger, three tests on each place, both
left and right side; 7Mean value heat and cold pain thresholds for frontal and
although this difference was less striking, it
thenar region, three tests on each side; Mann–Whitney U-test: (*)P = 0.05, seemed paradoxical compared to controls. Mean
*P < 0.05, **P < 0.01, ***P < 0.001. total sleep time in diaries did not differ between

50

Table 4 Comparison and interpretation of important symptoms and objective findings for the TTH, NSM, and SM patients compared to controls
Group Summary of findings
Sleep migraineurs Number of awakenings ↑, amount of superficial sleep (↑), but not increased
daytime tiredness or sleepiness
Non-sleep migraineurs Slow-wave sleep ↑, fast arousals ↓, increased frequency of daytime
tiredness, and reduced pain thresholds
Tension-type headache Slow-wave sleep ↑, increased frequency of daytime tiredness and tendency
to reduced pressure pain thresholds
the groups (Table 2). It has been shown that a
history of chronic insomnia does not predict poor
EEG sleep (23). However, increased slow-wave
sleep, as found in the present study among TTH
and NSM patients, is one factor that usually indi-
cates better sleep quality (24).
Reduced amounts of fast arousals and
increased slow-wave activity have also been found
the night after experimental sleep deprivation (25)
and are in line with what we found in TTH and
NSM patients. After foregoing sleep deprivation,
a ‘hypoarousal state’ with reduced fast arousals,
increased slow-wave sleep, and increased daytime
tiredness can be regarded as normal. However, in
the present study, sleep diaries revealed normal
average total sleep times before the PSG. Hence,
the NSM and TTH patients seem to have a
hypo-arousal state possibly related to a relative
sleep deprivation caused by increased need for
sleep. This hypothesis is consistent with the
notion that EEG among migraineurs is found to
be either normal or have subtle findings that may
reflect drowsiness (26). Furthermore, among mi-
graineurs, the sleepiness was even more increased
in the preictal phase as mean sleep onset latency
was reduced compared to interictal phase.
Increased objective sleepiness in the preictal phase
is consistent with subjective symptoms of drowsi-
ness (27) and with objective EEG findings of slow
activity preictally (28). However, our PSG find-
ings differ from Karthik et al. (29) who demon-
strated longer latency to sleep onset and less
NREM sleep among migraineurs. Methodological
differences such as patient selection might explain
divergent results.
SM patients differed from NSM and TTH
patients by having reduced sleep quality in PSG
(more awakenings than controls and less slow-
wave sleep than TTH and NSM). Hence, SM was
the only group with an ‘objective parallel’ to their
reported sleep disturbance. Lower index of slow
EEG bursts in NREM and high-frequency EEG
arousals during REM sleep have previously been
found among interictal SM patients (30). These
findings have been interpreted as hypofunction of
the arousal system (30). In the present study,
The headache-sleep study
Interpretation
Reduced sleep quality, preserved arousal system, or
hyperarousal?
Increased sleep need, recovery sleep, increased sleep quality,
hypo-arousal?
Increased sleep need, recovery sleep, increased sleep quality,
hypo-arousal?
signs of hypo-arousability were prominent among
the NSM, but not among the SM patients. The
relatively few sleep TTH patients did probably
not have a significant effect on the main results,
but a post hoc comparison showed that also non-
sleep TTH patients had less fast arousals than
sleep TTH patients. Accordingly, we hypothesize
that the hypo-arousability observed among the
NSM and TTH patients (mainly non-sleep TTH)
may be related to a relative sleep deprivation and
that these patients also may need more sleep than
healthy controls (Table 4).
Pain and sleep
We found decreased thermal pain thresholds in
the NSM and reduced pressure pain thresholds in
chronic TTH patients in concert with other stud-
ies (7, 31). Reduced pain thresholds have also
been found among healthy volunteers after sleep
deprivation (5). Migraineurs with head allodynia
during attack also report more sleep disturbances
than non-allodynic migraineurs when compared
to controls (32). In the present study, PSG and
pain threshold findings among NSM and TTH
(mostly non-sleep TTH) patients are in principle
similar. These findings, together with normal
average sleep times, are also consistent with a rel-
ative sleep deprivation (5). Preictal reduction in
sleep onset latency in the present study and preic-
tal heat pain threshold reduction in a previous
study (8) are also compatible with sleep depriva-
tion as a headache trigger.
However, among SM patients in the present
study, disturbed sleep in PSG, probably disposing
for sleep deprivation, was not accompanied by
reduced daytime vigilance or reduction in pain
thresholds. Hence, we hypothesize that arousabil-
ity is constitutionally different in our groups:
decreased in NSM (and TTH) and conserved in
SM patients. Arousability can in fact be associ-
ated with autonomic activity and possibly
explains pain threshold changes. Indeed, arous-
ability has been linked to hypertension (33) while
high blood pressure also seems to be associated
with hypo-algesia (34).
51
Engstrøm et al.
Headache onset time
The NSM and TTH patients had quite corre-
sponding results, possibly because the majority of
our TTH patients were non-sleep TTH. This
interesting finding suggests that sleep and attack-
precipitating mechanisms are similar across dif-
ferent headache diagnoses. Hence, the increased
combined load of affective and sleep-related
symptoms might be the main etiological factor
for headache. If so, the individual headache onset
time may only depend on which ‘drop’ that made
the flood. Increased susceptibility to daytime load
and subsequently increased need for sleep are
probably characteristic for headaches that tend to
start during daytime, while increased susceptibil-
ity to sleep disturbances probably is characteristic
when headache begins during sleep. However, in
the present study compared to migraineurs the
TTH group had a higher headache frequency
and, not surprisingly, headache was accompanied
by increased insomnia symptoms.
The SM group had no signs indicating sleep
deprivation per se according to the sleep diary,
PSG, daytime tiredness score, and pain thresh-
olds. However, disturbed sleep could increase the
risk for sleep deprivation (35), and symptoms
related to sleep, anxiety, and autonomic activity
among SM did not differ from other headache
groups. The latter signs might be part of what
make TTH and NSM patients wearier and
increase the sleep need (Fig. 3). Hence, we
hypothesize that SM patients have a more robust
arousal system than NSM and TTH. SM patients
may even be in a hyperarousal state.
We could not detect differences in sleep dis-
turbing parameters (AHI and PLM indexes)
between SM patients and NSM or controls. A
susceptibility to clinically nonsignificant sleep dis-
turbances might be a disadvantage of a robust
arousal system. This notion is in concert with the
observed increase of SM by age as sleep gets
lighter with increasing age (36). One might specu-
late that the difference in arousability between
the groups with attack onset during awake hours
and during sleep is related to the periaqueductal
gray (PAG) matter as ‘hyperarousal’ and ‘hypoa-
rousal’ have been related to ventrolateral and lat-
eral/dorsal PAG regions, respectively (37).
Strengths and limitations
The blinded and controlled cross-sectional study
design increases the probability for valid results.
Furthermore, the participants were mainly
recruited by advertising in local newspapers, in
52
Figure 3. Hypothesized different nervous system responses to
increased daily life strain.
contrast to a hospital-based migraine population
which may include more severe and long-standing
cases. On the other hand, the selection of head-
ache patients responding to a newspaper adver-
tisement may not be fully representative for the
whole headache population. It is evidently impos-
sible to recruit a completely unbiased patient
population. However, as most migraine patients
had been prescribed triptans for their attacks and
because headache history and headache intensity
were comparable to a previous hospital-based
population (38), our study groups were probably
fairly representative for a patient group encoun-
tered in a clinical setting.
A study design with repeated PSGs may be
more powerful for the detection of phase-related
differences. Individualized matched controls for
age and sex would have been preferable rather
than a comparable control group. A slight under-
estimation of sleep problems is expected in our
study because patients with known and diagnosed
previous sleep disorders were excluded; however,
this exclusion is a strength regarding the major
aims of the study. This study was exploratory,
and we did not adjust for multiple comparisons
because we did not want to increase type II fail-
ures on the cost of reducing type I failures (39).
The possibility of both type I and type II errors
is acknowledged. Independent replication of our
results is accordingly needed.
Conclusions
We have unified and compared data reported in
three papers (9–11), we conclude that all our
headache groups reported more anxiety and
sleep-related symptoms than controls. Even
though average total sleep times in TTH (mainly
non-sleep TTH) and interictal NSM patients were
normal, they had objective and subjective signs
consistent with a hypo-arousal state consistent

with foregoing sleep deprivation. SM patients
had findings in PSG consistent with disturbed
and lighter sleep, but they paradoxically did not
report more frequent daytime tiredness. The SM
patients seemed to have a more robust arousabili-
ty, possibly making them more susceptible for
sleep disturbing factors like hypopneas or PLMs
within the normal range. The combined results
from these studies suggest that there are sub-
groups within both the tension-type headache
and the migraine group in which sleep patterns
and attack-precipitating mechanisms are different.
However, the TTH and NSM subgroups had very
similar sleep and pain thresholds suggesting the
presence of common attack-initiating mechanisms
across headache disorder boundaries.
Acknowledgments
We thank Grethe Helde, Marit Stjern, and Gøril Bruvik
Gravdahl for practical assistance. This study was supported
by grant from Department of Neuroscience; Norwegian Uni-
versity of Science and Technology, and Liaison Committee
between the Central Norway Regional Health Authority
(RHA) and the Norwegian University of Science and Tech-
nology (NTNU). Trondheim, Norway.
Conflict of interest
None.
References
1. ODEGARD SS, ENGSTROM M, SAND T, STOVNER LJ, ZWART
JA, HAGEN K. Associations between sleep disturbance
and primary headaches: the third Nord-Trondelag Health
Study. J Headache Pain 2010;11:197–206.
2. ODEGARD SS, SAND T, ENGSTROM M, STOVNER LJ, ZWART
JA, HAGEN K. The long-term effect of insomnia on pri-
mary headaches: a prospective population-based cohort
study (HUNT-2 and HUNT-3). Headache 2011;51:570–80.
3. ODEGARD SS, SAND T, ENGSTROM M, ZWART JA, HAGEN
K. The impact of headache and chronic musculoskeletal
complaints on the risk of insomnia: longitudinal data
from the Nord-Trondelag health study. J Headache Pain
2013;14:24.
4. KELMAN L. The triggers or precipitants of the acute
migraine attack. Cephalalgia 2007;27:394–402.
5. OKIFUJI A, HARE BD. Do sleep disorders contribute to
pain sensitivity? Curr Rheumatol Rep 2011;13:528–34.
6. FERNANDEZ-DE-LAS-PENAS C, ARENDT-NIELSEN L, CUADRA-
DO ML, PAREJA JA. Generalized mechanical pain sensitiv-
ity over nerve tissues in patients with strictly unilateral
migraine. Clin J Pain 2009;25:401–6.
7. BEZOV D, ASHINA S, JENSEN R, BENDTSEN L. Pain percep-
tion studies in tension-type headache. Headache
2011;51:262–71.
8. SAND T, ZHITNIY N, NILSEN KB, HELDE G, HAGEN K,
STOVNER LJ. Thermal pain thresholds are decreased in
the migraine preattack phase. Eur J Neurol
2008;15:1199–205.
The headache-sleep study
9. ENGSTROM M, HAGEN K, BJORK MH et al. Sleep quality,
arousal and pain thresholds in migraineurs: a blinded
controlled polysomnographic study. J Headache Pain
2013;14:12.
10. ENGSTROM M, HAGEN K, BJORK M, GRAVDAHL GB, SAND
T. Sleep-related and non-sleep-related migraine: interictal
sleep quality, arousals and pain thresholds. J Headache
Pain 2013;14:68.
11. ENGSTRØM M, HAGEN K, BJØRK M, STOVNER LJ, STJERN
M, SAND T. Sleep quality, arousal and pain thresholds in
tension-type headache. Cephalalgia 2013, doi: 10.1177/
0333102413515339.
12. OLESEN JE ed. The international classification of head-
ache disorders, 2nd edn. Cephalalgia 2004;24(Suppl 1).
13. JOHNS MW. A new method for measuring daytime sleepi-
ness: the Epworth sleepiness scale. Sleep 1991;14:540–5.
14. ENGSTRØM M, ØDEGARD  SS, SAND T, STOVNER LJ, ZWART
JA, HAGEN K. The reliability of a new sleep screening
questionnaire for large population-based studies: The third
Nord-Trøndelag Health Study. Open Sleep J 2011;4:14–9.
15. BUYSSE DJ, REYNOLDS CF 3rd, MONK TH, BERMAN SR,
KUPFER DJ. The Pittsburgh sleep quality index: a new
instrument for psychiatric practice and research. Psychi-
atr Res 1989;28:193–213.
16. ZIGMOND AS, SNAITH RP. The hospital anxiety and
depression scale. Acta Psychiatr Scand 1983;67:361–70.
17. SUAREZ GA, OPFER-GEHRKING TL, OFFORD KP, ATKIN-
SON EJ, O’BRIEN PC, LOW PA. The autonomic symptom
profile: a new instrument to assess autonomic symptoms.
Neurology 1999;52:523–8.
18. NILSEN KB, SAND T, STOVNER LJ, LEISTAD RB, WESTG-
AARD RH. Autonomic and muscular responses and recov-
ery to one-hour laboratory mental stress in healthy
subjects. BMC Musculoskelet Disord 2007;8:81.
19. FORCE AAOSMT. Sleep-related breathing disorders in
adults: recommendations for syndrome definition and
measurement techniques in clinical research. The Report
of an American Academy of Sleep Medicine Task Force.
Sleep 1999;22:667–89.
20. American Academy of Sleep Medicine W, IL 60154,
U.S.A. The AASM manual for the scoring of sleep and
associated events. Rules, terminology and technical speci-
fications. Westchester, IL: American Academy of Sleep
Medicine, 2007.
21. SFORZA E, JOUNY C, IBANEZ V. Cardiac activation during
arousal in humans: further evidence for hierarchy in the
arousal response. Clin Neurophysiol 2000;111:1611–9.
22. LUCCHETTI G, PERES MF, LUCCHETTI AL, MERCANTE JP,
GUENDLER VZ, ZUKERMAN E. Generalized anxiety disor-
der, subthreshold anxiety and anxiety symptoms in pri-
mary headache. Psychiatry Clin Neurosci 2013;67:41–9.
23. ROSA RR, BONNET MH. Reported chronic insomnia is
independent of poor sleep as measured by electroenceph-
alography. Psychosom Med 2000;62:474–82.
24. KEKLUND G, AKERSTEDT T. Objective components of
individual differences in subjective sleep quality. J Sleep
Res 1997;6:217–20.
25. DE GENNARO L, FERRARA M, BERTINI M. EEG arousals
in normal sleep: variations induced by total and selective
slow-wave sleep deprivation. Sleep 2001;24:673–9.
26. SAND T. EEG in migraine: a review of the literature.
Funct Neurol 1991;6:7–22.
27. GIFFIN NJ, RUGGIERO L, LIPTON RB et al. Premonitory
symptoms in migraine: an electronic diary study. Neurol-
ogy 2003;60:935–40.
28. BJORK M, STOVNER LJ, HAGEN K, SAND T. What initiates
a migraine attack? Conclusions from four longitudinal
53
Engstrøm et al.
studies of quantitative EEG and steady-state visual-
evoked potentials in migraineurs. Acta Neurol Scand
Suppl 2011;124(Suppl 191):56–63.
29. KARTHIK N, SINHA S, TALY AB, KULKARNI GB, RAMACH-
ANDRAIAH CT, RAO S. Alteration in polysomnographic
profile in ‘migraine without aura’ compared to healthy
controls. Sleep Med 2013;14:211–4.
30. DELLA MARCA G, VOLLONO C, RUBINO M, DI TRAPANI G,
MARIOTTI P, TONALI PA. Dysfunction of arousal systems
in sleep-related migraine without aura. Cephalalgia
2006;26:857–64.
31. SCHWEDT TJ, KRAUSS MJ, FREY K, GEREAU RWT. Epi-
sodic and chronic migraineurs are hypersensitive to ther-
mal stimuli between migraine attacks. Cephalalgia
2011;31:6–12.
32. LOVATI C, D’AMICO D, BERTORA P et al. Correlation
between presence of allodynia and sleep quality in mi-
graineurs. Neurol Sci 2010;31(Suppl 1):S155–8.
33. JANACKOVA S, SFORZA E. Neurobiology of sleep fragmen-
tation: cortical and autonomic markers of sleep disor-
ders. Curr Pharm Des 2008;14:3474–80.
54
34. STOVNER LJ, HAGEN K. Hypertension-associated hypalge-
sia: a clue to the comorbidity of headache and other pain
disorders. Acta Neurol Scand Suppl 2009;120(Suppl
189):46–50.
35. BONNET MH, DOGHRAMJI K, ROEHRS T et al. The scoring
of arousal in sleep: reliability, validity, and alternatives. J
Clin Sleep Med 2007;3:133–45.
36. GORI S, LUCCHESI C, MORELLI N, MAESTRI M, BONANNI
E, MURRI L. Sleep-related migraine occurrence increases
with aging. Acta Neurol Belg 2012;112:183–7.
37. KEAY KA, BANDLER R. Parallel circuits mediating dis-
tinct emotional coping reactions to different types of
stress. Neurosci Biobehav Rev 2001;25:669–78.
38. SAND T, VINGEN JV. Visual, long-latency auditory and
brainstem auditory evoked potentials in migraine: rela-
tion to pattern size, stimulus intensity, sound and light
discomfort thresholds and pre-attack state. Cephalalgia
2000;20:804–20.
39. SCHULZ KF, GRIMES DA. Multiplicity in randomised
trials I: endpoints and treatments. Lancet 2005;365:
1591–5.