Spontaneous Intracranial Hypotension Presenting as Mental Deterioration

Correspondence and Clinical Notes Clinical Notes Intracranial Hypertension Induced by Rofecoxib Saiju Jacob, MD, MRCP; Yusuf A. Rajabally, MD

Several secondary causes are implicated in the etiology of idiopathic intracranial hypertension. Rofecoxib is a selective cyclo-oxygenase (COX)-2 inhibitor, now being increasingly used in place of nonselective nonsteroidal anti-inflammatory drugs (NSAIDs). We report a case of intracranial hypertension in a 69-year-old man 3 weeks after the commencement of rofecoxib therapy with reversal of clinical findings on drug withdrawal. Key Words: intracranial hypertension, rofecoxib

(Headache 2005;45:75-88)

A 69-year-old gentleman with a long-standing history of rheumatoid arthritis was admitted with a 1-week history of headache and double vision. The past history was unremarkable except for the rheumatoid arthritis. Three weeks prior to admission, he was commenced on rofecoxib for his arthritis. He had not been on any other medications except nonselective NSAIDs, which were replaced by rofecoxib. The headache was described as bifrontal in distribution with a severity of 8-9 out of 10. The headaches worsened on coughing and sneezing without clear postural variation, however. Diplopia on lateral gaze had been noticed 5 days after the onset of the headaches. There was no nausea, vomiting, photophobia, or other neurological symptoms. Neurological examination revealed bilateral papilledema with few hemorrhages and bilateral lateral rectus palsies. There was no other focal neurology. Blood pressure was raised at 178/110 mmHg, the rest of the cardiovascular examination being within normal limits. CT and MRI scans of the brain with and without contrast exclusively showed cerebrovascular ischemic changes. Venous sinuses were patent. Other investigations which were normal or negative included blood counts, renal and liver function, thyroid function, CRP, ESR, auto-antibodies, D-dimers, cholesterol, and ECG. The CSF examination showed an opening pressure of 31 cm with no other significant abnormalities. On withdrawal of rofecoxib, the

headaches, diplopia, and lateral rectus palsies gradually resolved within 2 weeks. The systemic hypertension persisted, requiring antihypertensive therapy.

DISCUSSION Idiopathic intracranial hypertension (IIH) has been previously referred to as “pseudotumor cerebri” or “benign intracranial hypertension.” The diagnostic criteria have been proposed by Smith and co-workers1 in 1985 and later by Friedman2 in 2002. Both these criteria define IIH as the occurrence of symptoms and signs of increased intracranial pressure (headache, nausea, vomiting, transient visual obscurations, papilledema with or without lateral rectus palsies), an elevated CSF pressure >25 cm of water, normal CSF content and the absence of hydrocephalus, compressive/structural lesion or venous thrombosis on imaging. IIH typically occurs in young, overweight women of the childbearing age. Development of IIH in children, elderly, thin individuals, or males is relatively rare. In such atypical patients, a secondary cause of intracranial hypertension should be suspected. Many of the secondary causes are excluded by normal neuroimaging (including magnetic resonance venography) and CSF composition. Several medications including antibiotics, Vitamin A, anabolic steroids as well as the withdrawal of corticosteroids have been implicated as secondary causes of intracranial hypertension. The list is vast and some authors classify these cases as intracranial hypertension of identified etiology. 3 Rofecoxib is a selective cyclo-oxygenase (COX)-2 inhibitor being increasingly used in view of its reduced gastrointestinal adverse events compared to other NSAIDs.4 Headaches have been described as a rare side effect of

From the University Hospitals of Leicester, Neurology, Leicester United Kingdom, (Drs. Jacob and Rajabally). Address all correspondence to Dr. Saiju Jacob, University Hospitals of Leicester, Neurology, Leicester, United Kingdom. Accepted for publication August 20, 2004.

75

76

January 2005

rofecoxib,4 but the exact pathogenesis of this is uncertain. An increased incidence of edema and blood pressure destabilization is being recently recognized as a side effect of rofecoxib.5-7 The significant increase in systolic blood pressure was even greater in patients aged 65 years and older.6 To our knowledge, this is the first report of rofecoxib causing symptomatic intracranial hypertension. It could be hypothesized that fluid retention, possibly secondary to renovascular influence, while resulting in systemic hypertension, could also contribute to the development of intracranial hypertension.

REFERENCES 1. Smith JL. Whence pseudotumor cerebri? J Clin Neuroophthalmol. 1985;5:55-56. 2. Friedman DI, Jacobson DM. Diagnostic criteria for idio-

3.

4. 5.

6.

7.

pathic intracranial hypertension. Neurology. 2002;59:14921495. Digre KB, Corbett JJ. Idiopathic intracranial hypertension (pseudotumor cerebri): a reappraisal. Neurologist. 2001;7:267. Scott LJ, Lamb HM. Rofecoxib. Drugs. 1999;58:499-505. Brinker A, Goldkind L, Bonnel R, Beitz J. Spontaneous reports of hypertension leading to hospitalisation in association with rofecoxib, celecoxib, nabumetone and oxaprozin. Drugs Aging. 2004;21:479-484. Cho J, Cooke CE, Proveaux W. A retrospective review of the effect of COX-2 inhibitors on blood pressure change. Am J Ther. 2003;10:311-317. Wolfe F, Zhao S, Pettitt D. Blood pressure destabilisation and edema among 8538 users of celecoxib, rofecoxib and nonselective nonsteroidal anti-inflammatory drugs (NSAID) and nonusers of NSAID receiving ordinary clinical care. J Rheumatol. 2004;31:1143-1151.

Spontaneous Intracranial Hypotension Presenting as Mental Deterioration Ping-Huang Tsai, MD; Shuu-Jiun Wang, MD; Jiing-Feng Lirng, MD; Jong-Ling Fuh, MD A 55-year-old woman had new onset of postural headache followed by change of mental status 3 weeks later. Magnetic resonance imaging (MRI) of the brain and whole spine showed typical spontaneous intracranial hypotension (SIH) findings, bilateral subdural hematoma, and cerebrospinal fluid leakage over the T 7–T 9. Her headache and mentality improved after epidural blood patches. Early recognition and correct diagnosis are crucial for successful treatment in patients with SIH presenting with mental confusion. Key words: coma, orthostatic headache, spontaneous intracranial hypotension Abbreviations SIH spontaneous intracranial hypotension, CSF spontaneous cerebrospinal fluid, MRI magnetic resonance imaging, MMSE mini-mental state examination, AVLT, auditory verbal learning test

Spontaneous intracranial hypotension (SIH) is a syndrome characterized by postural headache, spontaneous cerebrospinal fluid (CSF) leaks, and subsequent hypoliquorrhea. Bilateral subdural effusions or hematomas are not rare complications. Additional clinical features are protean and

From the Neurological Institute, Taipei Veterans General Hospital, Taipei, Taiwan (Drs. Tsai, Wang, and Fuh); Department of Neurology, National Yang-Ming University School of Medicine, Taipei, Taiwan (Drs. Tsai, Wang, and Fuh); Department of Radiology, Taipei Veterans General Hospital, Taipei, Taiwan (Dr. Lirng); and Department of Radiology, National YangMing University School of Medicine, Taipei, Taiwan (Dr. Lirng). Address all correspondence to Dr. Ling Fuh, The Neurological Institute, Taipei Veterans General Hospital, Taipei, Taiwan. Accepted for publication August 13, 2004.

include neck stiffness, nausea, vomiting, diplopia, blurred vision, third, fourth, and sixth cranial nerve palsy, dizziness, and hearing change.1 Behavioral and mental disturbances are very rare in patients with SIH and are easily misdiagnosed. Here we present a patient with manifestations of newonset daily headache and progressive mental and behavioral change.

CASE REPORT A 55-year-old woman, a junior high school teacher, was well until 1 month before admission. She developed new severe persistent, non-throbbing, bilateral frontal headaches that worsened when she was in an upright or sitting position and were relieved when she was lying down. In the beginning, the headache was accompanied by rhinorrhea and mild cough. About 20 days after symptom onset, her husband noted that she became loquacious, and at a high

Headache

77

Fig 1.—Brain MRI 43 days after headache onset (A–C). (A, B) T2-weighted images show a thin subdural hematoma in the bilateral frontotemporal regions and medioinferior displacement of the bilateral thalami obliterating the third ventricle and deforming the midbrain (arrowheads). (C) Sagittal T1-weighted image shows brain sagging with deformed midbrain and pons, obliteration of the CSF space in the interpeduncular cistern (arrow), and tonsillar herniation. Brain MRI 54 days after headache onset (D–F) shows slightly increased CSF space between the dorsal aspect of the clivus and the ventral surface of the brainstem (arrow). Brain MRI 3 months after headache onset (G–I) shows normal contour of the midbrain, normal position of the cerebellar tonsil, and a normal CSF space between the dorsal aspect of the clivus and the ventral surface of the brainstem (arrow).

school reunion manifested intermittent outbursts of bizarre speech. Then she stayed home, spent up to 12 hours a day in bed, coping with severe nausea and vomiting without fever or neck stiffness. She visited several clinics, where only

painkillers were prescribed. Her mental deterioration progressed. Disorientation, severe confusion, irritability, and inability to follow simple commands were noted 1 month after the onset of her headache. She was admitted to a hospital

78 where brain CT showed bilateral subdural hematoma. Brain magnetic resonance imaging (MRI) revealed bilateral subdural hematoma, diffuse dural enhancement, and marked descent of the brain. The CSF studies were normal except for low CSF pressure (60 mm H2 O). However, glycerol infusion was prescribed under the impression of increased intracranial pressure, possibly due to meningitis. However, her fading consciousness finally progressed to profound obtundation. After being transferred to our hospital, her consciousness remained clouded (day 43 after headache onset). On the basis of medical history and repeated similar brain images (Figure 1A–C), SIH was impressed. With SIH treatment including intravenous fluid challenge, aminophylline, and fludrocortisone, the patient’s condition improved and she could obey simple commands 2 days later. After lumbar epidural blood patch treatments with 20 ml of autologous blood, her mental status improved dramatically, but her orthostatic headaches and memory impairment persisted. The patient scored 24 points on the Mini-Mental State Examination (MMSE) at that time (full score = 30). Her comprehension and repetition were normal. The patient’s memory curve was 5/6/5/5/7 with 50% retention at 15 minutes for the 12-item auditory verbal learning test (AVLT). The patient, however, recognized 100% of the learning material (after four lessons). The patient could identify 14 animals in 1 minute and draw simple two-dimension figures, but had difficulty drawing three-dimension figures. One week later (54 days after headache onset), brain MRI (Figure 1D–F) was repeated. We measured the distance of the iter to the incisural line, the distance of the cerebellar tonsillar tip to the foramen magnum, and the distance from the basion to the medulla to evaluate the degree of the brain descent. Comparing the brain MRI before and after blood patch treatment revealed no significant change in the amount of subdural hematoma and in the degree of the brain descent, except for the slight recovery of CSF volume in the interpeduncular cistern (Figure 1D–F). These findings implied a reversal in the midbrain’s descent and compression. MRI myelography of the whole spine disclosed the presence of fluid in the right intervertebral foramina and paraspinal regions at the levels of T7–T9. A second blood patch given at day 61 after symptom onset improved both headache and mental status further. The patient became headache free and had normal mental status 2 days later. She scored 30 points on the MMSE. Her memory curve was 9/9/10/10/12 with 100% retention at 15 minutes during the 12-item AVLT test. Follow-up brain MRI 1 month later (Figure 1G–I) revealed only a very small amount of the subdural effusion and pachymeningeal en-

January 2005 hancement. The contours of midbrain and diencephalons were largely restored. She was doing very well at 8-months’ follow-up.

COMMENTS Mental status change is a very rare presentation of SIH, as in our patient. The underlying mechanism is not clear at present. Previous reports as well as ours showed that empirical treatment for SIH was effective for this rare complication. The Table lists our and previously reported six patients of SIH presenting with mental decline.2-7 Four are men and three are women. The mean age was 51 years with a range of 40–66. All of them had postural headache and mental status change. The mean CSF opening pressure was 69 mm H2 O (range from 60 to 95). Six patients were treated with epidural blood patches and three with intrathecal normal saline infusion. Mental status was partially restored in one patient and completely restored in six. The ascending reticular activating system in the midline structure is responsible for consciousness. Increase in intracranial pressure after treatment provides the buoyancy to relieve the compression of the mid-line structures, including midbrain and diencephalons. Initial neuroimaging in patients with SIH and mental deterioration revealed herniation,2,3,5-7 diencephalic compression,2,3 flattening of the pons,2,5 and midbrain compression,4,6 all of which returned to normal after successful treatment. The consciousness change was assumed to relate to diencephalic or midline compression.2-7 The interval between the initial and followup MRI in our patient was only 1 week, but her mental status improvement during that week was striking. The only difference between the two MRI findings was the slight restoration of the space between the midbrain and sphenooccipital synchondrosis. Therefore, we agree that midbrain compression is related to mental status change. Bilateral subdural hematoma was also a possible cause for mental status change. Bilateral subdural effusions/hematomas are common in patients with SIH. Five of the seven patients listed in the Table also showed subdural hematoma. Two of them received an operation to remove the subdural effusions/hematomas.3,6 The consciousness of one patient improved transiently, but deteriorated later.3 The other patient did not show any improvement.6 Their consciousness returned after intrathecal saline infusion. Surgical treatment of subdural hematoma is not indicated when clinical symptoms are absent, hematoma thickness is less than 15 mm, volume is less than 30 cc, or a midline shift is less than 5 mm.8 Although the above two patients met the criteria for

Headache

79

Table.—The Patients of Spontaneous Intracranial Hypotension Presented With Consciousness and/or Mental Deterioration in the English literature

Authors

Open Operation for Pressure Subdural Subdural Effect of Sex/Age (mm H2 O) Hematoma Hematoma Operation

Pleasure et al2

M/51

60

Y

N

–

Hong et al4 Blinder et al6

F/66 M/43

60 No

N Y

N Y

Beck et al5 Whiteley et al3

F/40 M/62

95 68

Y Y

N Y

Evans et al7

M/42

No

Y

N

– No improvement – Transient improvement and then deteriorated –

Tsai et al (current patient)

F/55

60

Y

N

–

Treatment for SIH

Final Outcome

Intrathecal saline infusion, EBP operation for repair the leakage Prednisolone EBP, intrathecal saline infusion Hydration, EBP Intrathecal saline infusion, EBP EBP

Total recovery

Hydration, aminophylline, EBP

Total recovery Total recovery Total recovery Minimally impaired executive function Total recovery except for a resolving third nerve palsy Total recovery

M: male; F: female; EBP: epidural blood patch; SIH: spontaneous intracranial hypotension.

surgery, the surgery did not improve their mental status. In our patient, consciousness was obtunded (an indication for surgery) and the sizes of the bilateral subdural effusions were 10.7 mm on the right side and 5.2 mm on the left. The mental status of our patient was restored to normal after epidural blood patch treatment and no reduction of the bilateral subdural hematoma. The size of the subdural hematoma decreased 1 month later in response to conservative treatment. In addition, one recent report showed acute mental deterioration after craniotomy in one patient with SIH.9 Therefore, surgical drainage might not be the treatment of choice for these patients. A recent study showed that more than one-half of patients with SIH are misdiagnosed initially.10 Meningitis was misdiagnosed in our patient initially. Glycerol was prescribed to reduce the intracranial pressure, which resulted in even less consciousness. Early clinical recognition of SIH in patients with mental status change is important for management. What is more, we consider that surgical drainage of subdural hematoma or effusion might not be necessary in patients with SIH.

CONCLUSIONS Spontaneous intracranial hypotension complicated with mental deterioration is very rare. The compression of diencephalon or midbrain is a possible underlying pathophysiology. Early recognition of this rare complication and increase of the CSF pressure by hydration or blood patches are mandatory in treatment of these patients.

REFERENCES 1. Mokri B. Headache caused by decreased intracranial pressure: diagnosis and management. Curr Opin Neurol. 2003;16:319-326. 2. Pleasure SJ, Abosch A, Friedman J, Ko NU, Barbaro N, Dillon W. Spontaneous intracranial hypotension resulting in stupor caused by diencephalic compression. Neurology. 1998;50:1854-1857. 3. Whiteley W, Al-Shahi R, Myles L, Lueck CJ. Spontaneous intracranial hypotension causing confusion and coma: a headache for the neurologist and the neurosurgeon. Br J Neurosurg. 2003;17:456-464. 4. Hong M, Shad GV, Adams KM, Turner RS, Foster NL.

80

January 2005

Spontaneous intracranial hypotension causing reversible frontotemporal dementia. Neurology. 2002;58:1285-1297. 5. Beck CE, Rizk NW, Kiger LT, Spencer D, Hill L, Adler JR. Intracranial hypotension presenting with severe encephalopathy. J Neurosurg. 1998;89:470-473. 6. Binder DK, Dillon WP, Fishman RA, Schmidt MH. Intrathecal saline infusion in the treatment of obtundation associated with spontaneous intracranial hypotension: technical case report. Neurosurgery. 2002;51:830-837.

7. Evans RW, Mokri B. Spontaneous intracranial hypotension resulting in coma. Headache. 2002;42:159-160. 8. Winn RH, Youmans JR. Youman’s Neurological Surgery, 5th edition. Philadelphia: W.B. Saunders, 2003. 9. Kelley GR, Johnson PL. Sinking brain syndrome: craniotomy can precipitate brainstem herniation in CSF hypovolemia. Neurology. 2004;62:157. 10. Schievink WI. Misdiagnosis of spontaneous intracranial hypotension. Arch Neurol. 2003;60:1713-1718.

Cluster-Like Headache Aura Status Marc Langedijk, MD; Joukje van der Naalt, MD, PhD; Gert Jan Luijckx, MD, PhD; Jacques De Keyser, MD, PhD We describe a patient with successive attacks (40 to 90 minutes) of cluster-like headache associated with aphasia, and contralateral hemihypesthesia and hemiplegia. The condition can best be described as cluster-like headache aura status. Key words: cluster headache, trigeminal autonomic cephalgia, aura, migraine aura status

Aura symptoms are occasionally reported in patients with cluster headache. In a prospective study in 230 patients with cluster headache, symptoms consistent with migraine aura were identified in 14%.1 Seventy percent of this group experienced visual symptoms, 16% hemimotor, and 13% hemisensory symptoms; 1% experienced both visual and hemisensory symptoms. The symptoms occurred either during or within 60 minutes before or after the cluster headache attack. Siow and colleagues described four cases of cluster headache with hemiparesis, for which they proposed the name cluster hemiparesis.2 We report here a patient with a variant form that can best be described as clusterlike headache aura status. A 24-year-old healthy man experienced paresthesia starting in the right foot and gradually spreading up over his right trunk, arm and face, together with severe leftsided retro-orbital stabbing pain. The pain was accompanied by ipsilateral lacrimation, conjunctival injection, and ptosis. Sensory symptoms were followed within minutes by a right-sided hemiplegia and expressive aphasia. Forty minutes later all symptoms, including the headache, resolved. On arrival at the hospital he developed a second identical attack that lasted 60 minutes. Subsequently, he had three From the Department of Neurology, University Hospital Groningen, Groningen, The Netherlands. Address all correspondence to Dr. Jacques De Keyser, Department of Neurology, University Hospital Groningen, 9713 GZ Groningen, The Netherlands. Accepted for publication May 28, 2004.

other similar attacks, of which the longest lasted 90 minutes, with symptom-free intervals of 15 to 30 minutes. On each attack right-sided sensory symptoms started with headache onset and were within minutes followed by expressive aphasia and right-sided hemiparesis. All neurological symptoms resolved completely when the headache subsided. There was no history of migraine or auras, and the patient had never experienced this type of headache before. Magnetic resonance imaging (MRI; T1, T2, Flair) of the brain and MR angiography of neck and intracranial vessels were normal. An electrocardiogram showed sinus arrhythmia. A cardiac ultrasonographic study was normal. One day later he was discharged with no residual neurological symptoms. He had no further attacks up to 18 months after discharge. His mother, cousin, and aunt, and his father’s sister and her daughter were known to have had migraine in childhood. Genetic analysis of the brain-specific P/Q-type calcium channel gene CACNA1A, which is one of the genes associated with familial hemiplegic migraine, revealed no abnormality. In analogy with migraine aura status, the condition of this patient can best be described as cluster-like headache aura status. In the series with cluster headache and aura studied by Bahra and colleagues, 36% had a history of migraine with or without aura and 15% had a family history of migraine.1 Our patient also had a family history of migraine, which lends further support to the hypothesis that migraine aura occurs in patients who carry an aura-susceptibility gene.3 The relationship between cluster headache and migrainous aura remains uncertain. Activation of the

Headache

81

trigeminovascular pathways, resulting in the stimulation of the cranial parasympathetic outflow, is considered to form the anatomical basis for the expression of trigeminal pain and ipsilateral autonomic symptoms in cluster headache.4 Functional MRI and positron emission tomography studies suggest that the initiator site of classical cluster headache is located in or near the posterior hypothalamus.4 An intriguing question in patients with cluster aura is whether in some cases the initiator site could be located in the cerebral cortex. The focal neurological manifestations may be due to cortical spreading depression that is characterized by slowly propagating waves of depolarizing cortical neurons and is believed to be the underlying mechanism for a migraine aura. Cortical spreading depression has been shown to activate trigeminovascular afferents,5,6 and this might be the underlying mechanism for the cluster-like headache. Recognizing cluster-like headache aura status as an entity is important to prevent the patient from useless and potentially harmful investigations.

REFERENCES 1. Bahra A, May A, Goadsby PJ: Cluster headache: a prospective clinical study with diagnostic implications. Neurology. 2002;58:354-361. 2. Siow HC, Young WB, Peres MF, Rozen TD, Silberstein SD: Hemiplegic cluster. Headache. 2002;42:136-139. 3. Goadsby PJ. Migraine, aura, and cortical spreading depression: why are we still talking about it? Ann Neurol. 2001;49:46. 4. Goadsby PJ: Pathophysiology of cluster headache: a trigeminal autonomic cephalgia. Lancet Neurol. 2002;1:251-257. 5. Bolay H, Reuter U, Dunn AK, Huang Z, Boas DA, Moskowitz MA: Intrinsic brain activity triggers trigeminal meningeal afferents in a migraine model. Nat Med. 2002;8:136-142. 6. Moskowitz MA, Nozaki K, Kraig RP: Neocortical spreading depression provokes the expression of c-fos proteinlike immunoreactivity within trigeminal nucleus caudalis via trigeminovascular mechanisms. J Neurosci. 1993;13:11671177.

Cluster Headache With Obstructive Sleep Apnea and Periodic Limb Movements During Sleep: A Case Report Zerrin Pelin, MD; Melda Bozluolcay, MD We report the case of a man with episodic cluster headache who suffered from severe obstructive sleep apnea (OSA) as well as periodic limb movements during sleep (PLMS). His attacks of headache occurred primarily during sleep being timely to REM sleep as 90 to 120 minutes interval. OSAs were more frequent and prolonged during REM sleep and oxygen saturation decreased to 81% during this sleep period. Periodic limb movements were also observed in our patient that were more frequent during the first half of the polysomnographic recordings. This case is one of the few reporting cases with CH who had both OSA and PLMS. Key words: cluster headache, sleep apnea, periodic limb movements, REM sleep,

Cluster headaches are characterized by recurrent severe unilateral head pain associated with ipsilateral cephalic autonomic changes, such as conjunctival injection, rhinorhea, forehead and facial sweating, miosis, ptosis, and eyelid edema.1 The association between CH and sleep has long

From the EEG and Sleep Disorders Unit, Department of Neurology, Pendik State Hospital, Pendik, Istanbul (Dr. Pelin); and Istanbul University, Cerrahpasa Medical School, Department of Neurology, Cerrahpasa, Istanbul (Dr. Bozluolcay). Address all correspondence to Dr. Zerrin Pelin, EEG and Sleep Disorders Unit, Department of Neurology, Pendik State Hospital, Pendik, Istanbul. Accepted for publication August 12, 2004.

been recognized because of the predilection of attacks to occur during nocturnal sleep.2 Studies including polysomnography revealed sleep apnea in cluster headache patients whose nocturnal attacks are preceded by oxyhemoglobin desaturation and related to rapid eye movement sleep.2,3 Periodic limb movements during sleep (PLMS) is a sleep disorder characterized by repetitive contractions of limb muscles that usually occur in the legs and are often associated with a partial arousal or awakening. There are very few reports of patients with CH who had both OSAs and PLMS. Here we report a patient with CH who suffered from severe OSAs as well as PLMS. A 43-year-old man, a businessman, presented with a chief complaint of sudden, excruciating headaches that had

82 began 20 years ago. His headaches began in spring and autumn months and tended to cluster in 9 to 10 weeks period. Sharp, right supraorbital, and temporal pain was accompanied by ipsilateral rhinorhea, sinus congestion, and tearing. During the period of attack, headaches began during nocturnal sleep and they recurred three to four times during a night with an interval of 90 to 120 minutes. Sometimes such attacks also occurred during daytime. He reported any benefit of analgesic drugs such as indomethacin, diclofenac, naproxene sodium. The cranial magnetic resonance imaging performed 3 years ago was reportedly normal. The patient also reported snoring very loudly and difficulty in breathing during sleep. He fell asleep easily while reading, watching television, after meal in the afternoon, and driving a car for a long time. He had nocturia (one or two times per night) and excessive sweating during sleep. Past medical history included nasal operation for septal deviation 2 years ago. An examination revealed a blood pressure of 140/85 mmHg and excess pharyngeal soft tissue. A detailed neurological examination was found to be normal. The polysomnographic studies included EEG (C4-A1, C3-A2, O2-A1, O1-A2), left and right electro-oculogram, submental and anterior tibialis muscles electromyograms, electrocardiogram, oro-nasal thermistors for the evalua-

January 2005 tion of airflow, thoracic and abdominal excursion, snoring, body position, and finger oximetry (Embla, Flaga, Iceland). Audio-video recording was performed throughout the night. Apnea was defined as complete airflow cessation for ≥10 seconds during sleep. Hypopnea was defined as a reduction of airflow at least 50% according to baseline for ≥10 seconds with a more than 3% reduction of oxygen saturation. Periodic leg movements (PLMs) were scored according to 1993 American Sleep Disorders Association’s (ASDA) criteria in order to establish an index of these events per hour of sleep. PLMS were also classified into groups by the presence or absence of an arousal according to the 1992 ASDA’s report. If the periodicity of leg movements existed, simultaneous leg movements with the end of apneas were accepted as leg movements, otherwise leg movements were excluded. The polysomnographic variables are shown in the Table. The evaluation of study demonstrated that obstructive sleep apneas (OSAs) and hypopneas were particularly prominent during REM sleep. It also showed that periodic limb movements were sometimes simultaneously occurring with the termination of apneas (Figure). PLMS were observed more frequently during the first half of the polysomnographic recording. Apnea-hypopnea index

Periodic limb movements and obstructive sleep apneas recorded from patient are shown on the eighth and ninth channel, respectively.

Headache

83

Table.—Polysomnographic Measures of the Patient

Measure

Time in bed (min) Total sleep time (min) Stage 1 (%) Stage 2 (%) Stage 3–4 (%) REM (%) Number of stage shifts Number of apneas Number of hypopneas PLMs with arousal PLMs without arousal

478.5 459.5 7 69.3 8.1 15.6 147 229 104 149 90

(total apnea-hypopneas/total hours of sleep) was calculated as 43.5, minimum oxygen saturation was found as 81%. Periodic leg movement index was found to be 30.4. Biochemical analysis, including complete blood count, erythrocyte sedimentation rate, glucose, urea, creatinine, iron, and ferritin level, was normal. We report one case fulfilling the criteria for episodic cluster headache with a possible association of OSAS and PLMS. Evidence that OSAS triggering CH includes case reports that treatment of OSAS in patients with cluster headache can lead to marked improvement or complete resolution in headaches. But there is only one report that was very recently published, which showed only two patients with CH who had both abnormal PLM score and OSA.3 Therefore, we thought that reporting this case might add some information about the relationship between CH and sleep. The prevalence of PLMS in general population is not known. The disorders appear to be rare in children and progresses with advancing age to become a relatively common finding, described in up to 34% of patients over the age of 60. Nobre et al found a high percentage (37.5%) of PLM in patients with CH and it was evaluated as a nonsignificant finding comparing with their control group (28%).3 If the mean age of their patients was thought to be 40 years, it might be suggested that it was a higher percentage of PLMS among their patients with CH. The circadian and circannual variation of CH is attributed to a hypothalamic generator in the suprachiasmatic nucleus. Melatonin secretion is regulated by the suprachiasmatic nucleus and is decreased in cluster headache patients. Melatonin is effective in the prophylaxis of episodic cluster headache.4 Melatonin is also effective in the treat-

ment of PLMS.5 In our patient, periodic leg movements were detected more frequently in the first half of the night that was similar to timing pattern of CH as reported by Chervin et al.2 This timing pattern and melatonin effectiveness in CH as well as PLMS might suggest an involvement of the circadian timing system in the pathophysiology of both disorders. The presence of a higher percentage of OSAS among patients with CH was shown in previous studies.2,3 More frequent, prolonged, and severe desaturation during REM sleep was shown as an explanation for the causal relationship between CH and OSAS.2,3 It was also proposed that during REM sleep, in which the chemoreceptor activity remains abnormal due to the changes in autonomic activity, hypoxemic stimuli acting over a deregulated threshold would cause a hyperactive reflex chemoreceptor response.3 As a consistent finding, three REM periods were observed during the recordings and oxygen saturation reached the lowest value during REM sleep as 81% in our patient. Periodic limb movements during sleep are a common finding in patients with OSA syndrome (OSAS).6 It is possible that our finding might be an incidental coincidence of CH and OSAS as well as PLMS. However, to achieve a better understanding of the interplay of sleep and CH, we thought that each new datum about the sleep disorders and headache might extend the point of view.

REFERENCES 1. Headache Classification Committee of the International Headache Society. Classification and diagnostic criteria for headache disorders, cranial neuralgias and facial pain. Cephalalgia. 1988;8(suppl 7):1-96. 2. Chervin RD, Zallek SN, Lin X, Hall JM, Sharma N, Hedger KM. Timing patterns of cluster headaches and association with symptoms of obstructive sleep apnea. Sleep Res Online. 2000;3:107-112. 3. Nobre ME, Filho PFM, Dominici M. Cluster headache associated with sleep apnea. Cephalalgia. 2003;23:276-279. 4. Leon M, D’Amico D, Moshiano F, Fraschini F, Bussone G. Melatonin versus placebo in the prophylaxis of cluster headache: a double-blind pilot study with parallel groups. Cephalalgia. 1996;16:494-496. 5. Kunz D, Bes F. Exogenous melatonin in periodic limb movement disorder: an open clinical trial and a hypothesis. Sleep. 2001;24:183-187. 6. Carelli G, Krieger J, Calvi-Gries F, Macher JP. Periodic limb movements and obstructive sleep apnea before and after continuous positive airway pressure treatment. J Sleep Res. 1999;8:211-216.

84

January 2005

Peripheral Postganglionic Sympathicoplegia Mimicking Cluster Headache Attacks Jan Hannerz, MD, PhD; Snjolaug ´ Arnardottir, MD, PhD; Hanne Pernille Bro Skejø, MD; Jan Anders Lilja, MD, PhD; Kaj Ericson, MD, PhD After dissection with complete occlusion of the internal carotid artery, a 58-year-old man started to suffer from intense cluster headache-like attacks. Magnetic imaging showed signs of nonsymptomatic cerebral emboli, which could be dated to have occurred in temporal relation to the start of the attacks, all on the right side. This case and two similar ones indicate that peripheral postganglionic sympathicoplegia can cause attacks with similar pain characteristics, accompanying symptoms, duration, and regularity as in cluster headache. Key words: cluster headache, carotid artery dissection

A 58-year-old man, nonsmoker, started to suffer from unilateral headache attacks localized behind the right eye. The pain was accompanied by ipsilateral conjunctival injection, lacrimation, and bilateral nasal congestion but no rhinorrhea. The duration of the attacks was 2 to 3 hours. The patient did not experience photo- or phonophobia or nausea during the attacks. The intensity of the pain was reported to be 8-9 on a pain scale of 0-10 (0 = no pain, 10 = maximal pain). During the attacks, the patient could not lie still in bed but moved around in the room. Between the attacks, he experienced a slight pain (2/10) ipsilateral to the attacks. The attacks appeared every night after 3-4 hours of sleep. Ten days after the start of the attacks they also appeared during the day, although he never experienced more than five attacks per 24 hours. From the 8th day after the start of the attacks, he was treated with suppositories of 1 mg ergotaminetartrate up to three times a day without satisfactory effects. When he came to Karolinska Hospital 19 days after the start of the attacks, he had a more pronounced Horner’s syndrome on the right side than usually found in cluster headache patients after an attack. No other neurologic deficiencies were found in the clinical investigations. His MRI with MRA showed dissection of and complete occlusion of the internal carotid artery ipsilateral to the Horner’s syndrome. MRI of the brain showed cortical changes from multiple emboli on the right side of the brain. Diffusion-weighted MR showed several areas with restricted diffusion in parietal and temporal

From the Department of Neurology (Drs. Hannerz and Arnardottir) and Department of Neuroradiology (Drs. Skejø, Lilja, and Ericson), Karolinska Hospital, Stockholm, Sweden. Address all correspondence to Dr. Jan Hannerz, Karolinska Hospital, Department of Neurology, Stockholm, Sweden. Accepted for publication May 19, 2004.

regions of the right hemisphere considered to be less than 710 days old and others were about a week older (Figure (A)). Pulse-gated single shot spin-echo echoplanar sequence with a b value of 1000 s/mm2 , 20 nonangulated slices covering the whole brain with slice thickness of 5 mm with 2 mm gap and a field view of 24 cm and a matrix of 128 × 128 were obtained, pulse gating with a trigger delay of 300 milliseconds was used according to our previous experience,1 tetrahederal diffusion gradient of Conturo was used,2 obtained as four iterations of a set of diffusion-weighted images with one measurement at b value of 0 s/mm2 followed by four measures in each direction at a b value of 1000 s/mm2 , images processed online to one set of isotropic diffusion weighted images, one set of calculated diffusion weighted images (without T2 shine through effect), and one set of trace ADC map. Perfusionweighted imaging (PWI) confirmed extensive disturbance of cerebral blood flow, cerebral blood volume, and mean transit time in the affected areas. T2-weighted gradient echo echoplanar sequence with TR of 2000 milliseconds and TE of 55, nine nonangulated slices covering the whole brain with slice thickness of 5 mm and a 9 mm gap with an FOV of 24 cm and a matrix size of 128 × 128, the slice position determined by using the DWI, 40 phases obtained corresponding to one image every 1.5 seconds in each position during injection of 15 ml of gadobutrol (Gadovist® 1.0, Schering Diagnostics, Germany) with an injection rate of 5 ml/sec followed by 50 ml of isotonic saline with the same injection rate, the contrast agent delivered by an MRI compatible power injector (Medrad, Inc.), the analysis of bolus tracking raw data processed off-line according to the principles described by Østergaard et al3,4 with calculation maps of relative cerebral blood volume, relative cerebral blood flow, and mean transit time. The patient was treated with heparin and warfarin. After 3 days of medication, the attacks stopped and were replaced by right-sided hemicrania with an intensity

Headache

85

A, Diffusion-weighted MR at the time of diagnosis showing recent infarctions in right hemisphere. B, Diffusion-weighted MR at follow-up 6 months later showing consolidated infarctions in right parietal and temporal regions.

of 2-4/10, which increased at strong exercise. No conjunctival injection, lacrimation, rhinorrhea, or nasal stuffiness was registered during this kind of headache. Six months after the start of the disease, he no longer suffered from headache and did not display signs of Horner’s syndrome, and normalization of the sympathetic neuropathy, tested with phenylephrine, had occurred. Doppler investigations of the carotid arteries and magnetic resonance imaging arteriography still showed complete occlusion of the right internal carotid artery. Diffusion-weighted MR showed consolidated, old infarcts in temporal and parietal regions of the right hemisphere, and a well-delineated area with reduced blood flow and blood volume in the right parietal area (Figure (B)). The patient did not have a relapse of cluster-like headache during the next 3 years of observation. Dissection within the wall of the internal carotid artery often results in ipsilateral hemicrania and Horner´s syndrome.5 Although the internal carotid artery was completely obstructed in the patient of the present study, the headache

was ipsilateral to the carotid disease. This indicates that the pain was caused by the sympathicoplegia which may cause dilation of the ophthalmic artery, as seen in most angiograms performed during cluster headache attacks,6-8 increasing blood flow to this vessel from other intracranial arteries with intact sympathetic regulation. At maximal pain in episodic cluster headache attacks common carotid artery blood flow9 and the diameters of the intracranial internal carotid and basilar arteries decrease compared to baseline conditions.10 The constriction of the large intracranial arteries and the resulting blood flow decrease have been considered to be caused by the pain for the attack to cease. Such a notion that constriction of intracranial arteries brings the attack to stop is supported by the efficacy of sumatriptan on cluster headache attacks, a drug known to constrict intracranial arteries. The pain in the patient in the present report was as intense as in episodic cluster headache attacks and may thus cause similar constriction of the large intracranial arteries in this patient as in cluster headache patients. When the patient is free from pain, the

86 constriction of the large arteries ceases after some hours and a new headache attack may start. This may explain the periodicity and the duration of the attacks of this patient with dissection of the internal carotid artery. Unilateral nasal congestion ipsilateral to the pain is usually reported in cluster headache attacks. In the attacks of the patient of the present study, bilateral nasal congestion was reported. Such bilaterality of this symptom, however, has been found to occur in 7% of regular episodic cluster headache attacks.11 Lacrimation is usually not seen in patients with internal carotid dissection with moderate hemicrania. In cluster headache attacks, Drummond found a relationship between pain intensity and lacrimation.12 Unilateral subcutaneous injection of capsaicin in the forehead of eight normal subjects, a potent pain provocation, caused lacrimation in six but ipsilaterally only in two. Bilateral conjunctival injection was observed in three and bilateral nasal congestion in one, ie, none had unilateral manifestations of the two latter “autonomous” symptoms.13 However, the capsaicin-induced pain only lasted for 60 seconds and may not be appropriate to use for comparison with the long-lasting pain of cluster headache. On the other hand, ipsilateral miosis occurred in four of the eight controls, symptoms which could be considered to be activated by the trigeminal-parasympathetic reflexes secondary to the capsaicin-induced pain. The findings of nonunilaterality in the capsaicin experiments suggest a vascular cause for the symptoms accompanying the pain in cluster headache and in the attacks of the patient in the present study rather than a trigeminal-parasympathetic reflex. Two cases with carotid dissection and ipsilateral cluster headache-like attacks have been reported earlier.14,15 These two patients did not have complete occlusion of the carotid artery in contrast to the patient in the present study. The second patient had two carotid dissections, one on each side with an interval of 1 year and with similar symptoms both the times. In the first of these two prior patients, the headache and the Horner’s syndrome persisted in contrast to the second patient and in the now reported patient in whom both Horner’s syndromes and headaches disappeared. This may support the notion of a relationship between the sympathicoplegia and the headache. The mechanisms involved in cluster headache are unknown. The medical histories of the three patients with dissection of the internal carotid arteries show that the symptomatology of cluster headache attacks, including the accompanying symptoms, the duration, and the clock-like regularity of the attacks may be caused by peripheral postganglionic sympathicoplegia.

January 2005 REFERENCES 1. Skare S, Andersson JRL. On effects of gating in diffusion imaging of the brain using single shot EPI. MRI. 2001;19:1125-1128. 2. Conturo TE, McKinstry RC, Akbudak E, Robinson BH. Encoding of anisotropic diffusion with tetrahedral gradients: a general mathematical diffusion formalism and experimental results. Magn Reson Med. 1996;35:399-412. 3. Østergaard L, Weisskoff RM, Chesler DA, Gyldensted C, Rosen BR. High resolution measurement of cerebral blood flow using intravascular tracer bolus passages. Part I: Mathematical approach and statistical analysis. Magn Reson Med. 1996;36:715-725. 4. Østergaard L, Sorensen AG, Kwang KK, Weisskoff RM, Gyldensted C, Rosen BR. High resolution measurement of cerebral blood flow using intravascular tracer bolus passages. Part II: experimental comparison and preliminary results. Magn Reson Med. 1996;36:726736 5. Bousser MG, Good J, Kittner SJ, Silberstein SD. Headache associated with vascular disorders. Wolff´s Headache and other head pain. 7th ed. Oxford University Press; 2001:360364. 6. Ekbom K, Greitz T. Carotid angiography in cluster headache. Acta Radiol (Diagn). 1970;10:177-186. 7. Hannerz J, Hellstrom ¨ G, Klum T, Wahlgren N-G. Cluster headache and “dynamite headache”: blood velocities in the middle cerebral artery. Cephalalgia. 1990;10:31-38. 8. Waldenlind E, Ekbom K, Torhall J. MR-angiography during spontaneous attacks of cluster headache: a case report. Headache. 1993;33:291-295. 9. Hannerz J, Jogestrand T. Pain induces decrease in blood flow in the common carotid arteries in cluster headache attacks. Cephalalgia. 1993;13:102-107. 10. Hannerz J, Greitz D. MRI of intracranial arteries in nitroglycerin induced cluster headache attacks. Headache. 1992;32:485-488. 11. Lance JW, Anthony M. Migrainous neuralgia or cluster headache? J Neurol Sci. 1971;13:401-414. 12. Drummond PD. Dissociation between pain and autonomic disturbances in cluster headache. Headache. 1990;30:505508. 13. Frese A, Evers S, May A. Autonomic activation in experimental trigeminal pain. Cephalalgia. 2003;23:67-68. 14. Rosenbraugh CJ, Griebel DJ, Di Pette DJ. A case report of carotid artery dissection presenting as cluster headache. Am J Med. 1997;102:418-419. 15. Frigerio S, Buhler ¨ R, Hess CW, Sturzenegger M. Symptomatic cluster headache in internal carotid artery dissection—consider anhidrosis. Headache. 2003;43:896900.

Headache

87

Severe Headaches Following Smallpox Vaccination James Sejvar, MD; Roumiana Boneva, MD, PhD; J. Michael Lane, MD, MPH; John Iskander, MD, MPH Headaches are common following smallpox vaccination; the re-introduction of civilian vaccination necessitates better understanding of the clinical features and outcome of postvaccination headache. We identified patients reporting headache following vaccination from among those reported to the U.S. Vaccine Adverse Events Reporting System to characterize demographic and clinical features. One-hundred and eight reports were obtained from among 627 smallpox vaccine-related reports, including 15 hospitalized persons. None had neurologic dysfunction or acute laboratory abnormalities; headache resolved in all except 2 hospitalized patients within 3 months. Severe headache following smallpox vaccination is generally transient, but debilitating headache may occur and further characterization is needed. Key words: headache, smallpox vaccination, adverse events

The recent re-introduction of vaccination against smallpox among selected civilian public health and health care volunteers has renewed concerns about vaccine-associated adverse events. Headaches have been reported in 44% of recipients of smallpox vaccine1 ; however, the clinical features and outcome of patients suffering severe headache have not been characterized. We initiated an investigation to characterize patients experiencing postvaccinial headache. We identified patients developing headache following smallpox vaccination from among reports submitted to the Vaccine Adverse Event Reporting System (VAERS) up to 2 weeks after vaccination. VAERS, a vaccine safety surveillance system operated by the U.S. Centers for Disease Control and Prevention (CDC) and the Food and Drug Administration, has been described previously.2 We defined severe headache as hospitalization where headache was a chief or major complaint. We abstracted information on demographics, past history, associated signs and symptoms, laboratory and neuroimaging from medical records, and gathered additional clinical and functional outcome data using a standardized questionnaire administered by telephone. Between January 24 and May 2, 2003, 34,752 civilians received smallpox vaccine. Among all 627 smallpox-vaccine related VAERS reports received by May 16, 108 (17.2%)

From the Division of Viral and Rickettsial Diseases (Dr. Sejvar), Division of HIV, STD, and TB Laboratory Research (Dr. Boneva); National Center for Infectious Diseases, Epidemiology and Surveillance Division, National Immunization Program (Drs. Lane and Iskander); and the Centers for Disease Control and Prevention, Atlanta, Georgia. Address all correspondence to Dr. James J Sejvar, Division of Viral and Rickettsial Diseases, National Center for Infectious Diseases, Centers for Disease Control and Prevention, 1600 Clifton Road, Mailstop A-39, Atlanta, GA 30333. Accepted for publication July 30, 2004.

included headache, of which 15 involved hospitalized persons. Females comprised 79% of reported headache patients and 65% of vaccinees overall. Persons aged 19 to 29 years (presumed primary vaccinees) accounted for 4.6% of all vaccinees, 4.3% of nonsevere, and 20% of severe headache cases. Associated signs and symptoms in the 15 patients with severe headache included nausea or vomiting (9, 60%), stiff neck (6, 40%), fever (6, 40%), myalgia (4, 27%), chest pain (4, 27%), and elevated blood pressure (2, 13%). Neurologic examination was normal in all cases. Cerebrospinal fluid was obtained in 4 patients; 1 was reported as normal, another had “mildly elevated protein and glucose,” and 2 had elevated protein levels (49 and 57 mg per dL; normal