1

Despite these shortcomings, the IHS division of all headaches into 2 major
groups is a simple, intuitive, and clinically applicable approach to headache
diagnosis. Indeed, the distinction between primary and secondary headaches
is, in reality, the first objective of every headache evaluation.
Primary headache types (shown in red here) are by definition benign,
relatively few in number, and, fortunately, represent the vast majority of
headache types seen in clinical practice. Secondary headaches (shown in gold
here), on the other hand, occur as a symptom of any of a long list of possible
underlying diseases.
1.

Classification and diagnostic criteria for headache disorders, cranial neuralgias and
facial pain. Headache Classification Committee of the International Headache
Society. Cephalalgia. 1988;8(suppl 7):1-96.

The cause or type of most headaches can be determined by a careful history

and physical examination. The clinical imperative is to recognize the warning
signals that raise red flags and prompt further diagnostic testing. In the absence
of worrisome features in the history or examination, the task is then to
diagnose the primary syndrome based upon the clinical features. If atypical
features are present or the patient does not respond to conventional therapy, the
diagnosis should be questioned and the possibility of a secondary headache
disorder should be revisited (1,2).
Because migraine and tension-type headache (TTH) account for over 90% of
the primary headache disorders in clinical practice, this discussion will focus
on their clinical features, the warning signals of serious secondary headaches,
and the role of diagnostic testing in the evaluation of headache (3).
1.

Silberstein SD, Lipton RB, Goadsby PJ. Headache in Clinical Practice. London,
England: Martin Dunitz; 2002.

2.

Olesen J, Tfelt-Hansen P, Welch KMA. The Headaches. 2nd ed. Philadelphia, PA:
Lippincott, Williams & Wilkins; 2000.

3.

Rasmussen BK, Jensen R, Schroll M, Olesen J. Epidemiology of headache in a

general population a prevalence study. J Clin Epidemiol. 1991;44:1147-1157.

These helpful headache hints provide a rapid screening approach to sort out
those headaches that can be dealt with on a non-urgent basis e.g migraine, as
opposed to secondary headaches of concern such as a thunderclap type of
headache that might be an indication of intracranial hemorrhage, for example
of subarachnoid origin. They are intended for the busy practitioner, e.g.
emergency room clinician who might desire to take a brief headache history.

Temporal patterns of presentation are major markers of diagnosis of all

headache types. The slide shows four patterns. Fill in the notes for which
headache type each pattern represents.
Fit the diagnosis to the temporal pattern. A=,B=,C=,D=; Choices are Tensiontype =, Migraine =, Cluster =, Intracranial neoplasm =.

More than in any other headache disorder, migraine sufferers identify triggers.
Stress is the trigger most commonly listed by patients. Dietary factors are also
frequently reported triggers, although few have been scientifically validated.
Although the impact of food triggers probably is not great for the population,
their impact could be for the individual. Oversleeping and sleep deprivation are
commonly recognized triggers. Patients should maintain a routine bedtime and
avoid sleeping in.
Hormonal headaches are triggered by variations in female estrogen levels and
possibly other hormonal factors. Noise, bright lights, and fumes are commonly
identified migraine triggers. Physical exertion can cause headache of the
subtype, exercise-induced migraine.
1.

Silberstein SD, Saper JR, Freitag FG. Migraine diagnosis and treatment. In:
Silberstein SD, Lipton RB, Dalessio DE, eds. Wolffs Headache and Other Head
Pain. 7th ed. Oxford, England: Oxford University Press; 2001:121-237.

2.

Silberstein SD, Lipton RB, Goadsby PJ. Headache in Clinical Practice. Oxford,
England: Isis Medical Media; 1998.

10

11

12

This patient had transient episodes of transient ill-defined patchy visual

obscuration preceded by scintillations, and a severe unilateral cervicooccipitally located pulsating headache of acute onset lasting days. The case
illustrates that large arteries are pain sensitive, the pulsating nature of the pain
indicates an arterial origin, and the ischemic embolization to both occipital
cortices can be mistaken as aura. This unilateral pulsating headache with
visual change MIMICS migraine with aura.

14

15

An attempt to elicit these worrisome features should be part of every

new-headache evaluation because their presence may signify an underlying
pathological condition for which diagnostic testing is obligatory.
Systemic symptoms, such as fever, malaise, or weight loss, should suggest an
underlying infectious or systemic inflammatory disorder. Newly acquired
neurologic signs or symptoms should always raise concern.
The mode of onset is perhaps the most important characteristic of a headache
to be delineated. Patients who have a sudden or abrupt headache that peaks in
seconds or minutes require careful assessment to exclude causes such as
subarachnoid hemorrhage (SAH), venous sinus thrombosis, arterial dissection,
or raised intracranial pressure.
Any new or progressive headache that begins in middle age or any headache
that deviates significantly from a previous pattern should be investigated
further.
If these features are addressed, the chance of overlooking a sinister cause for
headache are greatly diminished.
1.

Silberstein SD, Lipton RB, Dalessio DJ. Overview, diagnosis, and classification. In:
Silberstein SD, Lipton RB, Dalessio DJ, eds. Wolffs Headache And Other Head
Pain. 7th ed. Oxford, England: Oxford University Press; 2001:20.

16

Head injury without fracture of the skull was the origin of this subdural
hematoma (arrows). The CT is relatively early in the development of the
subdural collection because you see the increased density of the blood. Later it
can become the density of CSF but will occupy greater space. Note that the
brain is shifted to the opposite side and that the ventricular system on the side
of the hematoma is compressed. If this mass effect is not relieved (by a simple
burr hole in the skull), the patient may herniate. Headache is a prominent
feature of the presentation usually crescendo over days to weeks, and
generalized frontal in location.

How an Epidural Hematoma Can Be FatalNatasha Richardson's Epidural Hematoma: Why Blood Clot in Brain was Fatal
The autopsy report for Natasha Richardson is now in, regarding her "minor fall" at the Mont Tremblant ski resort in Quebec. After
the actress had fallen, ski patrol took her to the bottom of the hill and urged her to see a doctor. Natasha Richardson declined,
seemed normal, and went to
her room. But later, Natasha Richardson developed a severe headache, and was still conscious when the ambulance arrived to
take her to a Montreal hospital Tuesday. Natasha Richardson was then transferred to a NYC hospital.
About four hours lapsed between the time Natasha Richardson fell and was taken to the first hospital. At just what point she
lapsed into an irreversible coma, has not been released. The New York Daily News had reported that since Tuesday morning,
Natasha Richardson had been brain dead and on life support. According to the autopsy report, the fall caused a fracture to Natasha
Richardson's temporal lobe, which is a thin, delicate area of the skull. The fracture ripped a blood vessel that then began to slowly
leak blood into the brain.
This kind of blood clot is called an epidural hematoma, and only 1-3 percent of head injuries result in epidural hematoma, and
most are not fatal. It's not uncommon for the patient to feel no oddball symptoms, or feel anything wrong at all, one or two hours
following a blow to the head, all while the blood vessel slowly leaks blood onto brain tissue. But as the epidural hematoma
grows, symptoms will develop, including severe headache, nausea and vomiting. Natasha Richardson reported not feeling well at
all and suffering from a severe headache, but by then, it was too late.
If left untreated for too long, an epidural hematoma will compress the brain into the spinal cord, causing death to the brain. Once
brain cells die, they cannot be restored. So even though Natasha Richardson was whisked away to a hospital, this brain
compression had already been set in motion by the epidural hematoma, and it was irreversible.
A hit in the head, a head injury, can "shear" at a blood vessel, explained Dr. Nancy Snyderman on a "Today Show" segment that
was taped before Natasha Richardson died. Natasha Richardson was 45.
With an epidural hematoma, every minute literally counts. "Seems like an innocent fall," said Dr. Snyderman regarding Natasha
Richardson's ski accident. Other symptoms of an epidural hematoma: Inability to count backwards from 100, name the current
president and other cognitive abnormalities, weakness on one side of the body, convulsions, and "posturing," which is stiffening
of the body while the victim is unconscious.
Dr. Snyderman warned that if you receive a blow to the head, and someone urges you to see a doctor, there is a reason for that.
They are detecting an aberration that you are not aware of. A doctor will look into your eyes with a light, and if he suspects brain
trauma, will order a CT or MRI scan. A surgeon will "drill" into the skull, said Dr. Snyderman, and "suck" out the knot of
bleeding.
A physician on "On the Record" explained that had Natasha Richardson sought medical help sooner, she could have been walking
out of the hospital today with no residual effects of the accident.
If brain surgery is performed too late, death or permanent brain damage will result. The term epidural hematoma means blood

buildup between the dura mater and the skull; hematoma translates to blood mass.

18

19

Pathological specimen of intracerebral hypertensive hemorrhage. Likely to

present with the first and the worst headache of crescendo headache over
minutes

Although a number of primary headache syndromes exist that can present with
sudden, severe headache, many underlying diseases that can be clinically
indistinguishable, from benign thunderclap headache (TCH) to SAH, also
occur. These include:

Venous sinus thrombosis

Pituitary apoplexy

Arterial dissection

Meningoencephalitis

Acute hydrocephalus

Acute hypertension

Some of these conditions may be difficult to detect on CT scan, which

underscores the need for MRI in patients with bloodless CT and CSF who
present with a sudden-onset severe headache.
1.

de Bruijn SF, Stam J, Kappelle LJ. For the Cerebral Venous Sinus Thrombosis Study
Group. Thunderclap headache as first symptom of cerebral venous sinus thrombosis.
Lancet.1996;348(9042):1623-1625.

2.

de Bruijn SF, Stam J, Vandenbroucke JP. For the Cerebral Venous Sinus Thrombosis
Study Group. Increased risk of cerebral venous sinus thrombosis with thirdgeneration oral contraceptives. Lancet. 1998;351(9113):1404.

21

) Figure 3. Left: Coronal gadolinium-enhanced T1-weighted MR image revealing diffuse

pachymeningeal enhancement (curved arrows) and bowing of the optic chiasm over the pituitary fossa
(straight arrow) preoperatively (left). Right: Coronal image demonstrating resolution of the
pachymeningeal enhancement and restoration of the crowding of the optic chiasm after surgical ligation
of a leaking meningeal diverticulum of the left L-2 nerve root in a 41-year-old man.
Pachymeningeal Enhancement. Magnetic resonance imaging has revolutionized our understanding of
intracranial hypotension and undoubtedly is one of the most important factors responsible for the everincreasing number of patients in whom spontaneous intracranial hypotension can be diagnosed.
Enhancement of the pachymeninges (that is, dura mater) following administration of gadolinium is the
most characteristic neuroimaging feature of intracranial hypotension (Figs. 2 and
3).[2,8,21,25,28,29,45,49,56, 63,64,70,84] This association was first reported in 1991 by Mokri and
colleagues.[47] The pachymeningeal enhancement is diffuse, thin, involves both supra- and
infratentorial compartments, and may extend into the spinal canal.[44] There has been debate as to the
cause of the pachymeningeal enhancement observed on MR imaging. However, careful examination of
meningeal biopsy samples consistently demonstrate a thin layer of fibroblasts in the so-called subdural
zone as well as small thin-walled dilated blood vessels without evidence of inflammation.[48]
Meningeal fibrosis is seen only in patients with long-standing symptoms.[48] These findings strongly
suggest that dural venous dilation is the most likely explanation of the pachymeningeal enhancement in
intracranial hypotension. Similarly, reactive hyperemia of the pituitary gland in patients with
spontaneous intracranial hypotension may mimic a pituitary tumor

Acute thunderclap headache as a presentation of SAH (arrows show blood in

basal cisterns due to rupture of an aneurysm of the anterior Circle of Willis.

The major criteria and associated symptoms required for the IHS diagnosis of
migraine are so well known as to be almost intuitive for many clinicians.
However, these criteria were established to diagnose headaches, not patients.
When used literally, the sensitivity and specificity may be diminished.
Distinguishing between migraine and tension-type headache (TTH) can
sometimes be difficult because the conditions have overlapping features, and
patients have more than one type of headache. Their ability to ascribe
symptoms to a specific headache on recall may be unreliable.
Although TTH is the most common primary headache disorder, it is the least
distinctive, most poorly understood, and most frequently mimicked by
underlying diseases. In fact, its clinical diagnosis is based chiefly on the
absence of the symptoms that characterize migraine.

What we call TTH may be the lower end in a normal distribution of painful
episodic headaches. Whether some TTH is simply a mild migraine or a distinct
entity is still an area of debate.
1.

Cady R, Schreiber C, Farmer K, Sheftell F. Primary headaches: a convergence

hypothesis. Headache. 2002;42(3):204-216.

2.

Schulman EA. Overview of tension-type headache. Curr Pain Headache Rep.

2002;5(5):454-462.

3.

Waters WE. [editorial] The prevalence of migraine. Headache. 1978;18:53-54.

25

In both males and females, the prevalence distribution of migraine is an inverted Ushaped curve. Prevalence rises through early adult life and then falls after midlife.
The second important point to emphasize is that, at all postpubertal ages, migraine is
substantially more common in women than in men.
The prevalence of migraine varies as a function of age. Migraine is a disorder that is
most prevalent between the ages of 25 and 55. Part of the reason the condition has
such a big impact in the workplace is that it affects people during their peak
productive years.
At prepubertal ages, the rate of onset for migraine is actually a little bit higher in boys
than in girls, but at all postpubertal ages, the incidence is higher in girls than in boys.
The incidence of migraines without aura peaks around age 12 in boys and age 15 in
girls. Although half of all migraine onsets begin before the age of 20, migraine can
begin at age 1.
1.
2.
3.

Lipton RB, Stewart WF, Diamond S, et al. Prevalence and burden of migraine in the United
States: data from the American Migraine Study II. Headache. 2001;41(7):646-657.
Lipton RB, Stewart WF. Migraine in the United States: a review of epidemiology and health
care use. Neurology. 1993;43 (suppl 3):S6-S10.
Stewart WF, Linet MS, Celantano DD, et al. Age- and sex-specific incidence rates of
migraine with and without visual aura. Am J Epidemiology. 1991;134:1111-1120.

26

Images taken from migraine art painted by patients. Left frame illustrates
scintillating visual phenomena, middle frame illustrates illusion of teichopsia
(fortification spectra), and right frame shows slowly spreading dysesthesia
(cheiro-oral distribution). Together these frames indicate slowly spreading
neurological dysfunction produced by cortical spreading depression.
Welch KMA. Neurology Supplement 2004

29

MRI of a massive AVM presenting as migraine with visual aura.

34

Although the unique clinical features of cluster headache (CH) have been recognized since the 17 th century,
the striking periodicity was not articulated until the 1940s. The term cluster headache was coined in the
1950s, and since then the International Headache Society (IHS) has identified and classified two major
temporal patterns of CH (1). The episodic type (ECH), by far the most common (90%), is characterized by
discrete attack and remission phases. The chronic type (CCH) is defined by attacks that occur daily for more
than one year without remission or with remission periods lasting less than 14 days.
Cluster headache is rare (about 0.4% of the general population), and it predominates in males, although recent
studies indicate that the rate in females is rising (2). Onset can occur at any age but usually begins between 30
and 50 years of age (3).
In contrast to migraine headache, genetics in cluster headache is not thought to be important, although recent
studies have shown a positive family history in about 7% of patients with cluster headache. When compared
with prevalence of CH in the general population, first-degree relatives have about a 14-fold increased risk of
developing CH. Furthermore, in one study, five sets of monozygotic twins were 100% concordant for CH (4).
A number of related short-lasting headaches, referred to as cluster variants, may be confused with cluster
headache. These less common variants include chronic and episodic paroxysmal hemicranias and shortlasting unilateral neuralgiform with conjunctival injection and tearing (or SUNCT). Cluster variants have a
number of distinguishing features that have therapeutic implications and are important to recognize. These
related syndromes will be reviewed later in this presentation.
1.

International Headache Society Classification and diagnostic criteria for headache disorders, cranial
neuralgias and facial pain. Cephalalgia. 1988;8(Suppl 7):35-38.

2.

Ekbom K, Ahlborg B, Schele R. Prevalence of migraine and cluster headache in Swedish men of 18.
Headache. 1978;18(1):9-19.

3.

Swanson JW, Yanagihara T, Stang PE, et al. Incidence of cluster headaches: a population-based
study in Olmstead County Minnesota. Neurology. 1994;44:433-437.

4.

Russell MB. Genetic epidemiology of migraine and cluster headache. Cephalalgia. 1997;17:683701.

35

The most striking feature of CH the feature from which its name is derived is the
unmistakable periodicity of the attacks. Individual cluster attacks occur during attack
phases known as cluster periods. Most patients have one or two annual cluster periods,
each lasting between one and three months. Some patients have a seasonal propensity for
attacks related to the duration of the photoperiod, with the highest incidence of attacks
occurring in January or July. Intriguingly, the attacks occur soon after the shortest and
longest days of the year. This may have pathophysiologic implications, which we will
discuss later.(1)
Kudrow demonstrated that the most likely times for a cluster period to begin were
associated with the number of daylight hours; that is, more exacerbations occur within two
weeks following the summer and winter solstices, with fewer exacerbations beginning
within two weeks of the onset and offset of daylight savings time.(2)
Cluster periods punctuate longer-lasting remission periods, which usually last six months to
two years. During remission periods, neither spontaneous nor provoked attacks occur.
Although the duration of cluster and remission periods varies among individuals, these
periods remain relatively consistent within the same individual.
1.

Waldenlind E. Biological rhythm in cluster headache. In: Olesen J, Goadsby PJ, eds. Cluster
Headache and Related Conditions. London, England: Oxford University Press;1999:171-178.

2.

Kudrow L. The cyclic relationship of natural illumination to cluster period frequency. Cephalalgia.
1987;7 (Suppl 6):76-77.

36

Cluster headache also has a striking circadian periodicity, with most

individuals having one to three attacks per day, although some have up to eight
attacks daily. In an individual patient, the attacks usually occur at the same
time each day. As shown on this illustration, the most common times for
cluster attack onset are 1 a.m. to 2 a.m., 1 p.m to 3 p.m., and 9 p.m.
In addition, cluster headache is characterized by nocturnal attacks that
generally occur around the same time each night, with a peak incidence
between 1 a.m. and 3 a.m., as seen here in this first clock. This time roughly
correlates with the onset of the first period of rapid eye movement (REM)
sleep. Although this relationship has been well documented, the significance
and exact relationship of this association remains unclear.(1)

There appears to be a relationship in some patients of cluster headache and

obstructive sleep apnea (OSA). One possible trigger for cluster headache
attacks may be the observed hypoxia or hypercapnia normally associated with
OSA.
1.

Trucco M, Waldenlind E. Circadian distribution of episodic cluster headache attacks.

Cephalalgia. 1993;13 (Suppl 13):196.

2.

Chervin RD, Zallek SN, Lin X, Hall JM, Sharma N, Hedger KM. Sleep disorder
breathing in patients with cluster headache. Neurology. 2000;54(12):2302-2306.

37

As distinctive as its periodicity, each individual cluster attack has a highly stereotyped
profile. The attacks are almost exclusively unilateral, and the pain is excruciatingly
severe, located mainly around the orbit and temporal region. Most patients suffer from
strictly unilateral attacks, although the headache may alternate sides between cluster
periods or, more rarely, within the same cluster period. The headache peaks within
minutes and usually lasts between 45 and 90 minutes, although some last up to 8 hours.
In contrast to the quiescent state seen in migraine, the cluster patient prefers to pace in
an agitated and colicky state, where neither position nor rest offers any relief.
Over the past few years, several authors have reported in independent studies that
migrainous symptoms, such as prodromal and premonitory symptoms, nausea, vomiting,
photophobia, phonophobia, and even visual aura, are more commonly associated with
cluster attacks than was previously recognized. Whether this reflects a shared underlying
pathogenesis or a similar phenotype based on a common final pathway for expression is
unclear.
Cranial autonomic symptoms occur in the vast majority of patients and are considered
integral to the diagnosis of this syndrome.
1.

Dodick DW, Campbell JK. Cluster headache: diagnosis, management and treatment. In:
Silberstein SD, Lipton RB, Dalessio DJ, eds. Wolffs Headache And Other Head Pain. 7th ed.
New York, NY: Oxford University Press; 2001:283-309.

38

Autonomic symptoms are present in more than 97% of patients. Among the
local signs of autonomic involvement, lacrimation and conjunctival injection
are the most common, each present in more than 80% of patients. Nasal
stuffiness or rhinorrhea is experienced by 70% to 75% of sufferers. All
autonomic symptoms are transient, lasting only for the duration of the attack,
with the exception of a partial Horners syndrome, which occurs in up to 66%
of cases and may become persistent after long periods of attacks. On occasion,
a patient may notice some degree of facial flushing, sweating or edema (1).
1.

Dodick DW, Campbell JK. Cluster headache: diagnosis, management and treatment.
In: Silberstein SD, Lipton RB, Dalessio DJ, eds. Wolffs Headache And Other Head
Pain. 7th ed. New York, NY: Oxford University Press;2001:283-309.

39

Although a unifying pathophysiologic explanation of cluster headache is not yet available, any attempt to
understand this syndrome must take into account the three cardinal features of the disorder. These include
pain, autonomic features, and stereotyped periodicity.
Seminal observations on the neurobiology of cluster headache have recently been made. To recognize their
significance, a basic understanding of neurovascular anatomy is essential.
First, cephalic pain is relayed to the central nervous system via nociceptive ophthalmic branches of the
trigeminal nerve, which innervates pain-sensitive intracranial structures such as the dura mater and dural
blood vessels. Substance P and calcitonin gene-related peptide (CGRP) are trigeminovascular neuropeptides
that are released in both animals and humans when the trigeminal fibers or ganglion are activated. CGRP, the
most potent vasodilator in the human body and in animal models, when released, leads to the production of
neurogenic inflammation and dilation of dural blood vessels. Activation of the trigeminovascular system in
cluster headache has been corroborated by recent evidence demonstrating markedly elevated blood levels of
CGRP in the external jugular vein (EJV) of patients during a cluster attack.
Second, the autonomic features of CH indicate activation of the cranial parasympathetic fibers. These fibers
originate from first-order neurons within the superior salivatory nucleus, which has a functional brainstem
connection to the trigeminal nucleus caudalis. These fibers travel with the seventh cranial nerve and synapse
in the pterygopalatine ganglia. Post-ganglionic fibers provide vasomotor and secretomotor innervation to the
cerebral blood vessels and the lacrimal and nasal mucosal glands, respectively. Activation of this pathway has
similarly been supported by the finding of dramatically elevated blood levels of vasoactive intestinal
polypeptide (VIP) in the EJV of sufferers during an attack.
The presence of a postganglionic Horners syndrome during attacks of CH indicates involvement of the
carotid sympathetic plexus. The cavernous carotid artery is a likely location since it is at this level where the
parasympathetic, sympathetic, and trigeminal fibers converge. Indeed, evidence from a variety of imaging
sources has demonstrated the presence of arterial dilatation and venous outflow obstruction in the region of
the cavernous sinus. This evidence collectively provides an explanation for the pain and autonomic features
underlying a cluster attack, but does not account for the periodicity of the syndrome (1).

40

1.

Edvinsson L, Goadsby PJ. Neuropeptides in headache. Eur J Neurol. 1998;5:329-341.

40

One study collected data confirming that cluster headache is a CNS disorder and
can be considered a form of neurovascular headache. The data indicate that
dilatation of cranial vessels is not specific to any particular headache syndrome,
but is generic to cranial neurovascular activation and is probably mediated by the
trigeminoparasympathetic reflex.
To examine the neurovascular mechanisms in cluster headache, the investigators
used PET (coregistered with subjects structural MR image) to study 18 CH
patients. When an acute CH attack was triggered with nitroglycerin (NTG),
activation occurred in the ipsilateral posterior inferior hypothalamic gray, the
contralateral ventroposterior thalamus, the anterior cingulate cortex, the ipsilateral
basal ganglia, the right anterior frontal lobe, and both insulae. The authors also
found significant activation (vasodilatation) in the region of the major basal
arteries that was caused in part by NTG, but was also induced by capsaicin
injection into the forehead (1).
1.

May A, Bahra A, Buchel C, Frackowiak RS, Goadsby PJ. PET and MRA findings in
cluster headache and MRA in experimental pain. Neurology. 2000;55(9):1328-1335.

41

One study of 9 patients with chronic cluster headache used PET imaging during acute
nitroglycerin-induced cluster headache attacks and found marked activation in the
hypothalamic grey, an area specific for cluster headache. This activation pattern has not been
observed in migraine or experimental ophthalmic division head pain. Furthermore,
hypothalamic activation was not seen in the control group of cluster patients given
nitroglycerin to induce an attack while they were in remission. This finding implies that the
hypothalamus is involved in the pain process in a permissive or triggering manner, rather than
simply as a response to first division nociception. Furthermore, given that this area is involved
in circadian rhythms and sleep-wake cycling, these data established an involvement of the
hypothalamic area in the genesis of an acute cluster attack (1).
Another study that also supports the idea of involvement of the hypothalamic area in CH
pathogenesis provides for the first-time tantalizing evidence that primary headache disorders
may be associated with abnormal brain structure as well as function.(2) Voxel-based
morphometric MR imaging, an objective and automated method of analyzing changes in brain
structure, was used to study the brain structure of patients with cluster headache. A significant
structural difference was found in the hypothalamic gray compared with controls. This
structural anomaly correlated with the area of activation demonstrated in the PET studies (2).
1.

May A, Bahra A, Buchel C, et al. Hypothalamic activation in cluster headache attacks. Lancet.
1998;352:275-278.

2.

May A, Ashburner J, Buchel C, et al. Correlation between structural and functional changes in
brain in an idiopathic headache syndrome. Nat Med. 1999;5(7):836-838.

43

47