Curr Neurol Neurosci Rep (2014) 14:437 DOI 10.1007/s11910-014-0437-9
HEADACHE (R HALKER, SECTION EDITOR)
Thunderclap Headache Esma Dilli
Published online: 20 March 2014 # Springer Science+Business Media New York 2014
Abstract Thunderclap headache (TCH) is a sudden severe headache that peaks to maximum intensity within 1 minute. Subarachnoid hemorrhage is the most commonly identified etiology for this headache, however, other secondary etiologies should be considered. Sentinel headache, reversible cerebral vasoconstriction syndrome, arterial dissection, cerebral venous sinus thrombosis, pituitary apoplexy, intracranial hemorrhage, ischemic stroke, reversible posterior leukoencephalopathy, spontaneous intracranial hypotension, colloid cyst, and intracranial infections are other possible causes of TCH. Investigations for the etiology of TCH begin with noncontrast CT head and lumbar puncture. MR brain, CT angiogram, MR angiogram, or CT/MR venogram may need to be performed if the initial investigations are negative. Treatment and prognosis depend on the etiology of the TCH. Keywords Thunderclap headache . Subarachnoid hemorrhage . Aneurysm Reversible cerebral vasoconstriction syndrome . Reversible posterior leukoencephalopathy . Sentinel headache . Cerebral venous sinus thrombosis . Arterial dissection . Pituitary apoplexy . Spontaneous intracranial hypotension . Colloid cyst
Introduction Thunderclap headache (TCH) describes a sudden severe headache that peaks to maximum intensity within 1 minute. Day and Raskin first used the term TCH to describe a sentinel
headache associated with an unruptured intracranial aneurysm [1]. Since then, other secondary etiologies associated with TCH have been described (Table 1). Primary TCH is a diagnosis of exclusion when other etiologies have been ruled out by diagnostic evaluation; however, most headache specialists question the existence of Primary TCH and believe secondary causes were missed due to insensitive, incomplete testing, or poor timing of investigations. TCH is a neurologic emergency and should be evaluated and treated urgently. TCH is associated with high morbidity and mortality. Therefore, prompt evaluation is required to rule out serious secondary etiologies such as subarachnoid hemorrhage (SAH), cerebral venous sinus thrombosis (CVST), carotid dissection, and reversible cerebral vasoconstriction syndrome (RCVS).
Epidemiology The incidence of thunderclap headache is unknown due to the various possible etiologies. In a prospective study the incidence was reported to be 43 per 100,000 adults per year [2]. In this study, 28 % of patients presenting with TCH had a history of migraine [2]. Recurrent attacks of TCH occurred in 24 % of patients with non-SAH thunderclap headache, whereas none of the patients with SAH related TCH had recurrent attacks [2]. At an emergency headache center, approximately 120 out of 8000 patients each year presented with TCH [3•].
This article is part of the Topical Collection on Headache
Clinical Presentation
E. Dilli (*) Department of Medicine, Division of Neurology, University of British Columbia, 8219-2775 Laurel Street, Vancouver, BC V5Z 1M9, Canada
Patients typically present with severe “worst headache” of life with peak intensity within 1 minute. Time to peak intensity helps differentiate TCH from other severe headache
437, Page 2 of 12 Table 1 Differential diagnosis of thunderclap headache Secondary headache Vascular etiology Subarachnoid hemorrhage Sentinel bleed related to unruptured aneurysm Arterial dissection Reversible cerebral vasoconstriction syndrome Cerebral venous sinus thrombosis Intracranial hemorrhage Ischemic stroke Pituitary apoplexy Reversible posterior leukoencephalopathy Arterial hypertension Retroclival hematoma Nonvascular intracranial disorders Spontaneous intracranial hypotension Third ventricle colloid cyst Intracranial infection
conditions. It is more important to ask how long it took for “the worst headache of your life” to reach maximal intensity instead of the severity alone. TCH may last minutes to days and may be single or recurrent. The location and type of pain is nonspecific. Spontaneous onset of the pain is commonly reported. However, there may be precipitants such as Valsalva, exertion, and sexual activity [2]. TCH may be associated with photophobia, nausea, vomiting, neck pain, visual symptoms, altered level of consciousness, and focal neurologic symptoms and signs. These symptoms and/or signs may help to determine the etiology of the TCH.
Differential Diagnosis Subarachnoid Hemorrhage Subarachnoid hemorrhage (SAH) is defined as “bleeding into the subarachnoid space (space between the arachnoid membrane and pia matter of the brain or spinal cord)” [4]. SAH is the most common cause of secondary TCH; 11 %–25 % of patients with TCH have a SAH as the etiology [2, 5]. Approximately 70 % of patients with SAH present with a headache of whom 50 % of have a TCH [5–7]. Possible mechanisms for the headache include stretching of the vessel wall, elevation of intracranial pressure or irritation from blood products. The location of pain varies; however, the most common site is the occipital region with neck pain. The headache usually lasts few a days; and rarely less than 2 hours [8].
Curr Neurol Neurosci Rep (2014) 14:437
Although TCH is the most common presentation of SAH, 50 % of patients have other clinical features [9]. In a retrospective study of 109 patients with SAH, Fontanarosa reported altered awareness (53 %), meningismus (35 %), photophobia, seizures, impaired cognition, nausea/vomiting (77 %), visual abnormalities, stroke, hypertension, and electrocardiogram changes [10]. Fundal hemorrhage or disc swelling can be seen in 20 %– 50 % of patients particularly if the patient has altered level of consciousness [11–13]. The extent of SAH appears to be correlated to the onset of vitreous hemorrhage and not retinal hemorrhage [12]. No specific clinical features can differentiate TCH related to SAH from other etiologies, however, the presence of any combination of age >40, neck pain or stiffness, loss of consciousness or headache onset during exertion may be a useful clinical predictor of non-traumatic SAH [14]. In the study by Mark et al these clinical features had a sensitivity of 97 % and specificity of approximately 23 % for the diagnosis of SAH [14]. Investigations for TCH are focused on assessing for a SAH due to its morbidity and mortality of approximately 50 % [15]. SAH is most commonly due to an intracranial aneurysm rupture (85 %), followed by nonaneurysmal perimesencephalic hemorrhage (10 %), cerebral arteriovenous malformation (AVM), dural arteriovenous fistula (AVF), mycotic aneurysm, and transmural arterial dissection [16]. The incidence of aneurysmal SAH (aSAH) is 2–23 per 100,000 patient years [16–20]. A Norwegian study reported that a modest increase in the incidence of aneurysmal SAH over the years may be related to improved diagnoses with no change in mortality over a 23 year follow-up [17]. Approximately 5 %–12 % of SAH are missed during an emergency room visit and are diagnosed with either migraine or neck pain as the etiology for their headache [21, 22]. Risk factors for aneurysm SAH include hypertension, smoking, heavy alcohol consumption, family history of aneurysms, and genetic conditions for autosomal dominant polycystic kidney disease and Ehlers Danlos type IV disease [17, 23–26]. The risk of SAH is four times greater in first degree family members with a SAH and six times greater in smokers than the general population [27]. The overall lifetime risk of SAH in a person with 1 affected first degree family member is 1 % at 50 years and 2 % at 70 years [28]. Females are about 1.9 times more likely to have a SAH than males [17]. Patients with suspected SAH should undergo computed tomography (CT) scan without contrast as quickly as possible after the onset of the symptoms as the sensitivity of the CT scan to detect SAH decreases over time [9]. During the first 12 hours after SAH the sensitivity of CT is approximately 98 %. The sensitivity decreases to 86 %–93 % at 24 hours, 76 % at 48 hours, and 58 % at 5 days [29]. Newer multidetector CT scanners are insufficient in ruling out SAH alone [30].
Curr Neurol Neurosci Rep (2014) 14:437
In a prospective study of 592 patients presenting with nontraumatic acute headache and normal neurologic examination to a tertiary emergency room, the specificity, sensitivity, and negative likelihood ratio of combination of CT and lumbar puncture for detecting a SAH was 98 %, 67 %, and 0.02, respectively [31]. CT angiogram (CTA) is a reliable test to rule out aneurysms when CT head is negative or imaging shows perimesencephalic hemorrhage [32, 33]. In a systematic review, the pooled sensitivity and specificity of a CTA for diagnosing an aneurysm in patients with acute SAH were 98 % and 100 %, respectively [34]. In a study of 116 patients with suspected SAH, CTA showed aneurysms in 6 CT negative patients; 3 of whom had positive lumbar puncture (LP) [35]. All patients with TCH should have a LP if head imaging is normal [9]. Cerebrospinal fluid (CSF) xanthochromia should be checked if the CSF shows red blood cells. In a retrospective review of 152 patients presenting to the ER with TCH with 2 weeks of onset and negative noncontrast CT scan, the sensitivity, specificity, positive predictive value, and negative predictive value of CSF xanthochromia by visual inspection for detection of cerebral aneurysms was reported to be 93 %, 95 %, 72 %, and 99 %, respectively, [36]; however, clinical follow-up was the only basis to determine an absence of aneurysm in 63 of the CSF negative patients. A retrospective view of 19 patients with documented SAH, xanthochromia was identified in only 47 % of patients while 53 % had colorless supernatant on visual inspection [37]. In a prospective study of 253 patients with normal CT scans and suspected SAH, spectrophotometry had sensitivity, specificity, and positive predictive value of 100 %, 75.2 %, and 3.3 %, respectively, in diagnosing SAH and therefore, the authors suggested spectrophotometry had limited clinical utility [38]. MRI is more sensitive than CT in detecting a SAH in the acute phase unless the cisterns are tight [39, 40]. MRI FLAIR imaging shows hyperintensity in the subarachnoid space. In the subacute phase, MRI T2* images show hypointensity in the subarachnoid space and is the most sensitive test [41] with a sensitivity of 94 % in the acute and 100 % in the subacute phase [42]. In a systematic review of population based studies from 1960 1992, case -fatality rate after SAH was found to have decreased during the last 30 years [15] possibly due to earlier identification and improved management including treatment of complications such as rebleeding and vasospasm. Unruptured Intracranial Aneurysm—“Sentinel Headache” Sentinel headache (SH) refers to a “warning headache” that occurs days to weeks prior to a SAH [43]. Patients with SH typically present with a TCH without meningismus or altered level of consciousness [44].
Page 3 of 12, 437
The incidence of a sentinel headache was reported in a systematic review to be 10 %–43 % in patients with aneurysmal SAH [45]; however, this may not reflect the true incidence due to potential recall and referral bias [46]. Also, patients may not be able to report the history of headache when the SAH resulted in a high morbidity or mortality. In a study of 237 patients with SAH, approximately 17 % of the patients reported a sentinel headache [47]. According to prospective autopsy and angiographic studies, 3.6 % and 6 % of the population have an intracranial aneurysm [46]. However, it is unknown whether there is a direct correlation between TCH and unruptured aneurysm. Proposed explanations for the pain include stretching/expansion of the arterial wall or minor bleeding from a leak of an aneurysm that is not detected in imaging and/or lumbar puncture. Sentinel headache is typically diagnosed retrospectively where the patient had TCH days to weeks prior to the intracranial aneurysmal rupture with SAH. In a prospective review, 75 % of patients with a sentinel headache, had the TCH within 2 weeks of the SAH [48] with a peak incidence at 24 hours. There is a minor peak in the 7–14 day period that may reflect the associated rebleeding risk of a SAH [44]. The odds ratio for rebleeding prior to aneurysm repair was ten times in those patients with sentinel headache when controlling for age, timing, aneurysm size and number, and Hunt-Hess grade [47].
Cervical Artery Dissection Headache and/or neck pain are the most common symptoms of a cervical artery dissection (CeAD). Headache is described in 60 %–95 % and 70 % of patients with carotid and vertebral artery dissections, respectively [49]. Headache or neck pain may precede other neurologic symptoms and/or signs by a median time of 4 days for a carotid dissection or 14.5 hours for a vertebral artery dissection [49]. Thunderclap headache (9.2 % vs 3.6 %; P=0.001) and neck pain were more common (65.8 % vs 33.5 %; P
HEADACHE (R HALKER, SECTION EDITOR)
Thunderclap Headache Esma Dilli
Published online: 20 March 2014 # Springer Science+Business Media New York 2014
Abstract Thunderclap headache (TCH) is a sudden severe headache that peaks to maximum intensity within 1 minute. Subarachnoid hemorrhage is the most commonly identified etiology for this headache, however, other secondary etiologies should be considered. Sentinel headache, reversible cerebral vasoconstriction syndrome, arterial dissection, cerebral venous sinus thrombosis, pituitary apoplexy, intracranial hemorrhage, ischemic stroke, reversible posterior leukoencephalopathy, spontaneous intracranial hypotension, colloid cyst, and intracranial infections are other possible causes of TCH. Investigations for the etiology of TCH begin with noncontrast CT head and lumbar puncture. MR brain, CT angiogram, MR angiogram, or CT/MR venogram may need to be performed if the initial investigations are negative. Treatment and prognosis depend on the etiology of the TCH. Keywords Thunderclap headache . Subarachnoid hemorrhage . Aneurysm Reversible cerebral vasoconstriction syndrome . Reversible posterior leukoencephalopathy . Sentinel headache . Cerebral venous sinus thrombosis . Arterial dissection . Pituitary apoplexy . Spontaneous intracranial hypotension . Colloid cyst
Introduction Thunderclap headache (TCH) describes a sudden severe headache that peaks to maximum intensity within 1 minute. Day and Raskin first used the term TCH to describe a sentinel
headache associated with an unruptured intracranial aneurysm [1]. Since then, other secondary etiologies associated with TCH have been described (Table 1). Primary TCH is a diagnosis of exclusion when other etiologies have been ruled out by diagnostic evaluation; however, most headache specialists question the existence of Primary TCH and believe secondary causes were missed due to insensitive, incomplete testing, or poor timing of investigations. TCH is a neurologic emergency and should be evaluated and treated urgently. TCH is associated with high morbidity and mortality. Therefore, prompt evaluation is required to rule out serious secondary etiologies such as subarachnoid hemorrhage (SAH), cerebral venous sinus thrombosis (CVST), carotid dissection, and reversible cerebral vasoconstriction syndrome (RCVS).
Epidemiology The incidence of thunderclap headache is unknown due to the various possible etiologies. In a prospective study the incidence was reported to be 43 per 100,000 adults per year [2]. In this study, 28 % of patients presenting with TCH had a history of migraine [2]. Recurrent attacks of TCH occurred in 24 % of patients with non-SAH thunderclap headache, whereas none of the patients with SAH related TCH had recurrent attacks [2]. At an emergency headache center, approximately 120 out of 8000 patients each year presented with TCH [3•].
This article is part of the Topical Collection on Headache
Clinical Presentation
E. Dilli (*) Department of Medicine, Division of Neurology, University of British Columbia, 8219-2775 Laurel Street, Vancouver, BC V5Z 1M9, Canada
Patients typically present with severe “worst headache” of life with peak intensity within 1 minute. Time to peak intensity helps differentiate TCH from other severe headache
437, Page 2 of 12 Table 1 Differential diagnosis of thunderclap headache Secondary headache Vascular etiology Subarachnoid hemorrhage Sentinel bleed related to unruptured aneurysm Arterial dissection Reversible cerebral vasoconstriction syndrome Cerebral venous sinus thrombosis Intracranial hemorrhage Ischemic stroke Pituitary apoplexy Reversible posterior leukoencephalopathy Arterial hypertension Retroclival hematoma Nonvascular intracranial disorders Spontaneous intracranial hypotension Third ventricle colloid cyst Intracranial infection
conditions. It is more important to ask how long it took for “the worst headache of your life” to reach maximal intensity instead of the severity alone. TCH may last minutes to days and may be single or recurrent. The location and type of pain is nonspecific. Spontaneous onset of the pain is commonly reported. However, there may be precipitants such as Valsalva, exertion, and sexual activity [2]. TCH may be associated with photophobia, nausea, vomiting, neck pain, visual symptoms, altered level of consciousness, and focal neurologic symptoms and signs. These symptoms and/or signs may help to determine the etiology of the TCH.
Differential Diagnosis Subarachnoid Hemorrhage Subarachnoid hemorrhage (SAH) is defined as “bleeding into the subarachnoid space (space between the arachnoid membrane and pia matter of the brain or spinal cord)” [4]. SAH is the most common cause of secondary TCH; 11 %–25 % of patients with TCH have a SAH as the etiology [2, 5]. Approximately 70 % of patients with SAH present with a headache of whom 50 % of have a TCH [5–7]. Possible mechanisms for the headache include stretching of the vessel wall, elevation of intracranial pressure or irritation from blood products. The location of pain varies; however, the most common site is the occipital region with neck pain. The headache usually lasts few a days; and rarely less than 2 hours [8].
Curr Neurol Neurosci Rep (2014) 14:437
Although TCH is the most common presentation of SAH, 50 % of patients have other clinical features [9]. In a retrospective study of 109 patients with SAH, Fontanarosa reported altered awareness (53 %), meningismus (35 %), photophobia, seizures, impaired cognition, nausea/vomiting (77 %), visual abnormalities, stroke, hypertension, and electrocardiogram changes [10]. Fundal hemorrhage or disc swelling can be seen in 20 %– 50 % of patients particularly if the patient has altered level of consciousness [11–13]. The extent of SAH appears to be correlated to the onset of vitreous hemorrhage and not retinal hemorrhage [12]. No specific clinical features can differentiate TCH related to SAH from other etiologies, however, the presence of any combination of age >40, neck pain or stiffness, loss of consciousness or headache onset during exertion may be a useful clinical predictor of non-traumatic SAH [14]. In the study by Mark et al these clinical features had a sensitivity of 97 % and specificity of approximately 23 % for the diagnosis of SAH [14]. Investigations for TCH are focused on assessing for a SAH due to its morbidity and mortality of approximately 50 % [15]. SAH is most commonly due to an intracranial aneurysm rupture (85 %), followed by nonaneurysmal perimesencephalic hemorrhage (10 %), cerebral arteriovenous malformation (AVM), dural arteriovenous fistula (AVF), mycotic aneurysm, and transmural arterial dissection [16]. The incidence of aneurysmal SAH (aSAH) is 2–23 per 100,000 patient years [16–20]. A Norwegian study reported that a modest increase in the incidence of aneurysmal SAH over the years may be related to improved diagnoses with no change in mortality over a 23 year follow-up [17]. Approximately 5 %–12 % of SAH are missed during an emergency room visit and are diagnosed with either migraine or neck pain as the etiology for their headache [21, 22]. Risk factors for aneurysm SAH include hypertension, smoking, heavy alcohol consumption, family history of aneurysms, and genetic conditions for autosomal dominant polycystic kidney disease and Ehlers Danlos type IV disease [17, 23–26]. The risk of SAH is four times greater in first degree family members with a SAH and six times greater in smokers than the general population [27]. The overall lifetime risk of SAH in a person with 1 affected first degree family member is 1 % at 50 years and 2 % at 70 years [28]. Females are about 1.9 times more likely to have a SAH than males [17]. Patients with suspected SAH should undergo computed tomography (CT) scan without contrast as quickly as possible after the onset of the symptoms as the sensitivity of the CT scan to detect SAH decreases over time [9]. During the first 12 hours after SAH the sensitivity of CT is approximately 98 %. The sensitivity decreases to 86 %–93 % at 24 hours, 76 % at 48 hours, and 58 % at 5 days [29]. Newer multidetector CT scanners are insufficient in ruling out SAH alone [30].
Curr Neurol Neurosci Rep (2014) 14:437
In a prospective study of 592 patients presenting with nontraumatic acute headache and normal neurologic examination to a tertiary emergency room, the specificity, sensitivity, and negative likelihood ratio of combination of CT and lumbar puncture for detecting a SAH was 98 %, 67 %, and 0.02, respectively [31]. CT angiogram (CTA) is a reliable test to rule out aneurysms when CT head is negative or imaging shows perimesencephalic hemorrhage [32, 33]. In a systematic review, the pooled sensitivity and specificity of a CTA for diagnosing an aneurysm in patients with acute SAH were 98 % and 100 %, respectively [34]. In a study of 116 patients with suspected SAH, CTA showed aneurysms in 6 CT negative patients; 3 of whom had positive lumbar puncture (LP) [35]. All patients with TCH should have a LP if head imaging is normal [9]. Cerebrospinal fluid (CSF) xanthochromia should be checked if the CSF shows red blood cells. In a retrospective review of 152 patients presenting to the ER with TCH with 2 weeks of onset and negative noncontrast CT scan, the sensitivity, specificity, positive predictive value, and negative predictive value of CSF xanthochromia by visual inspection for detection of cerebral aneurysms was reported to be 93 %, 95 %, 72 %, and 99 %, respectively, [36]; however, clinical follow-up was the only basis to determine an absence of aneurysm in 63 of the CSF negative patients. A retrospective view of 19 patients with documented SAH, xanthochromia was identified in only 47 % of patients while 53 % had colorless supernatant on visual inspection [37]. In a prospective study of 253 patients with normal CT scans and suspected SAH, spectrophotometry had sensitivity, specificity, and positive predictive value of 100 %, 75.2 %, and 3.3 %, respectively, in diagnosing SAH and therefore, the authors suggested spectrophotometry had limited clinical utility [38]. MRI is more sensitive than CT in detecting a SAH in the acute phase unless the cisterns are tight [39, 40]. MRI FLAIR imaging shows hyperintensity in the subarachnoid space. In the subacute phase, MRI T2* images show hypointensity in the subarachnoid space and is the most sensitive test [41] with a sensitivity of 94 % in the acute and 100 % in the subacute phase [42]. In a systematic review of population based studies from 1960 1992, case -fatality rate after SAH was found to have decreased during the last 30 years [15] possibly due to earlier identification and improved management including treatment of complications such as rebleeding and vasospasm. Unruptured Intracranial Aneurysm—“Sentinel Headache” Sentinel headache (SH) refers to a “warning headache” that occurs days to weeks prior to a SAH [43]. Patients with SH typically present with a TCH without meningismus or altered level of consciousness [44].
Page 3 of 12, 437
The incidence of a sentinel headache was reported in a systematic review to be 10 %–43 % in patients with aneurysmal SAH [45]; however, this may not reflect the true incidence due to potential recall and referral bias [46]. Also, patients may not be able to report the history of headache when the SAH resulted in a high morbidity or mortality. In a study of 237 patients with SAH, approximately 17 % of the patients reported a sentinel headache [47]. According to prospective autopsy and angiographic studies, 3.6 % and 6 % of the population have an intracranial aneurysm [46]. However, it is unknown whether there is a direct correlation between TCH and unruptured aneurysm. Proposed explanations for the pain include stretching/expansion of the arterial wall or minor bleeding from a leak of an aneurysm that is not detected in imaging and/or lumbar puncture. Sentinel headache is typically diagnosed retrospectively where the patient had TCH days to weeks prior to the intracranial aneurysmal rupture with SAH. In a prospective review, 75 % of patients with a sentinel headache, had the TCH within 2 weeks of the SAH [48] with a peak incidence at 24 hours. There is a minor peak in the 7–14 day period that may reflect the associated rebleeding risk of a SAH [44]. The odds ratio for rebleeding prior to aneurysm repair was ten times in those patients with sentinel headache when controlling for age, timing, aneurysm size and number, and Hunt-Hess grade [47].
Cervical Artery Dissection Headache and/or neck pain are the most common symptoms of a cervical artery dissection (CeAD). Headache is described in 60 %–95 % and 70 % of patients with carotid and vertebral artery dissections, respectively [49]. Headache or neck pain may precede other neurologic symptoms and/or signs by a median time of 4 days for a carotid dissection or 14.5 hours for a vertebral artery dissection [49]. Thunderclap headache (9.2 % vs 3.6 %; P=0.001) and neck pain were more common (65.8 % vs 33.5 %; P
