Cluster headache: pupillometric patterns as a function of the degree of anisocoria

Fifty-three patients with cluster headache, mean age 42.6 years, were examined by means of pupillometry. Pharmacological stimulation was carried out by the instillation of eye drops; the sympathomimetic agents hydroxyamphetamine (a norepinephrine releaser) and phenylephrine (an agent acting directly on the postsynaptic receptors) were applied. Pupillary dilatation was measured at set time intervals, comparing the responses of the symptomatic and non-symptomatic sides. The material was divided into groups according to the degree of basal anisocoria. The subgroup with the most pronounced basal miosis of the symptomatic side demonstrated a uniform pattern of deficient symptomatic side dilatation after OH-amphetamine and supersensitivity to phenylephrine. The other groups demonstrated the same general pattern, but to a far lesser degree. In cluster headache, the extent of nonresponsiveness to OH-amphetamine and of phenylephrine supersensitivity on the symptomatic side thus, at least partly, seems to be a function of the degree of anisocoria. The response pattern in cluster headache seems to differ from that of 3rd neuron Horner's syndrome with an anisocoria of the same extent.     

Sweat gland and pupillary responsiveness in Horner''s syndrome

Eight patients with Horner's syndrome (five with a 1st neuron lesion and three with a 2nd neuron lesion) were examined for their pupillary responses to pharmacologic stimulation with tyramine (2%) and with phenylephrine (1%) eye drops. The same patients were also evaluated for their forehead sweating pattern on stimulation with body heating and pilocarpine injection, using the Evaporimeter. Five patients had a brain stem (1st sympathetic neuron) lesion, while three patients had had a traumatic C8-Th1 root avulsion and hence had a preganglionic neuron lesion. The average response with the phenylephrine eye test and the pilocarpine sweat test differed markedly between the two groups; only the central neuron lesion group had a supersensitivity reaction to both drugs. These procedures may be of diagnostic value in localizing the lesion in patients with a Horner's syndrome of unknown etiology. Patients with 3rd neuron lesion have not been examined with this combination of techniques.     

Latent dysautonomic pupillary lateralization in cluster headache. A pupillometric study

Forty-five patients with cluster headache in the asymptomatic phase were studied by electronic pupillography, testing autonomic function of both pupils pharmacologically. Topical sympathetically-acting mydriatics, tyramine and cocaine and the cholinoceptor blocker, homatropine, induced defective mydriatic responses on the symptomatic side, indicating latent impairment of sympathetic function. The abnormality was found in interattack intervals of the cluster period or during intercluster phases. The tyramine test can be proposed for objective diagnosis of cluster headache. We postulate that cluster attacks are triggered and lateralized by a permanent latent unilateral sympathetic dysfunction. Lithium reduced the mydriatic response to tyramine of the pupil contralateral to the pain, thus restoring the equilibrium between both pupils; this therapy may correct the asymmetric sympathetic function by attenuating the activity in the asymptomatic side.     

Cervicogenic headache

Eleven female patients with cervicogenic headache (mean age, 43 years; range, 25-59 years) have been examined with the pupillometer. The pupillary diameter was examined in the basal state (that is, the status before pharmacologic stimulation) and after topically administered tyramine (2%), phenylephrine (1%), and hydroxyamphetamine (1%). A total of 51 tests were performed, 35 in the asymptomatic period and 16 during pain attacks. In a control group consisting of 26 age-matched women a total of 39 tests were carried out. Before pharmacologic stimulation (that is, in the "basal state") the pupils were smaller in the asymptomatic (pain-free) period than during pain attacks in the patients and also as compared with that of control individuals. The anisocoria (the difference in pupillary size in the same individual) observed was not significantly different between the patient group and control individuals either in the basal state (before pharmacologic stimulation) or after pharmacologic stimulation. The mydriasis resulting from the instillation of the three sympathicomimetic drugs was symmetrical in both controls and patients both during and between the pain attacks. This finding is in clear contrast to what is found in cluster headache, in which there is a "Horner-like" syndrome on the symptomatic side. These two headaches thus seem to differ essentially with regard to this variable.     

The Enhanced Ciliospinal Reflex in Asymptomatic Patients With Cluster Headache is Due to Preganglionic Sympathetic Mechanisms

An amplified ciliospinal reflex response has been documented in patients with cluster headache, lacking a Horner like syndrome. The mechanism is unknown, Tentatively, it may be due to an increased release of monoamines from post-ganglionic sympathetic nerve endings or an increased density of postsynaptic adrenergic receptors in the dilatator muscle of the iris.
The instillation of a 1% phenylephrine solution into the conjunctival sac induces mydriasis by stimulating postsynaptic adrenergic receptors in the dilatator muscle of the iris, while the instillation of a 2% tyramine solution causes mydriasis by releasing noradrenaline from the presynaptic sympathetic nerve terminals in the iris.
According to these premises, a positive correlation shouId be expected between the ciliospinal reflex response and the pupillary response to tyramine, if the enhanced ciliospinal so-flex response was due to an increased presynaptic release of monoamines. No such correlation was found. Nor was there any positive correlation between the ciliospinal reflex response and the pupillary response to phenylephrine, contradicting an increased density of postsynaptic monoaminergic receptors in the dilatator muscle of the iris as the explanation. However, there was a significant positive correlation between the pupillary responses to phenylephrine and tyramine, ruling out any functionally caused "denervation" hypersensitivity in the dilatator muscle of the iris.
It is concluded that the amplified ciliospinal reflex response in cluster headache patients (lacking a Horner-like syndrome) reflects compensatory pathophysiological mechanisms proximal to the third-order sympathetic neuron.     

Pupil responsiveness in cluster headache: A dynamic TV pupillometric evaluation

In this study the variations in pupil diameter induced by different stimuli (dark-light adaptation, light reflex, electric stimulation of the sural nerve) were investigated in episodic (in the active or remission phases) and in chronic cluster headache (CH) patients. Pupil size monitoring was performed with a monocular, infrared TV pupillometer, and sural nerve stimuli were applied after the pain threshold had been measured as the flexion reflex threshold of the biceps femoris muscle (RIII reflex). The results were compared with those obtained in patients with "peripheral" (third neuron) Horner's syndrome and in healthy sex- and age-matched controls. On the symptomatic side we found an impairment of pupil response to light flashes and nociceptive stimuli; similar findings were sometimes evident on the pain-free side, too. These results substantiate previous observations that in cluster headache a dysfunction of the integrative central nervous system pathways also exists intercritically and mostly bilaterally, involving both autonomic regulation and pain perception mechanisms.     

Ciliospinal Reflex Response in Cluster Headache

The ciliospinal reflex response is mainly mediated by second- and third-order sympathetic nerves to the dilatator muscle of the iris. As the pupillary response to various pharmacological agents indicates a sympathetic dysfunction in patients with cluster headache, the ciliospinal reflex was studied in 25 patients. Five of these patients with cluster headache exhibited a Horner-like syndrome (miosis, ptosis) on the symptomatic side. The pupillary responses to phenylephrine and tyramine showed that the Horner-like syndrome was due to postganglionic sympathetic nerve dysfunction. Their ciliospinal reflex response on the symptomatic side was significantly less than in controls and in other patients with cluster headache, lacking a Horner-like syndrome. This also applied to the nonsymptomatic side compared to the majority of cluster headache patients without any clinical evidence of sympathetic nerve dysfunction.
These findings seem to delineate those patients with a Horner-like syndrome as a subgroup, distinctly separated from the majority of cluster headache patients. Furthermore, the findings indicate that the Horner-like syndrome is not a consequence of repeated attacks of headache over many years, but is a manifestation of bilateral cephalic sympathetic dysfunction being more marked on the symptomatic side.
In 18 (72%) of our 25 patients, an asymmetric and lower ciliospinal reflex response on the symptomatic side was seen. In 3 (12%) patients, there was no difference in the response. In 4 patients (16%), the incorrect side was indicated by an asymmetric reflex response. Two of these patients (8%) had suffered from cluster headache on alternating sides.
In summary, the findings support the concept that dysfunction of the sympathetic nervous system, whether peripheral or central, is involved in the pathophysiology of cluster headache.     

Cluster headache: Forehead sweating pattern during heating and pilocarpine tests

Thirty-one patients with cluster headache were examined with regard to their forehead sweating pattern, by means of the Evaporimeter. Sweating was stimulated in two different ways: by body heating and by parenterally administered pilocarpine. The resulting increase in evaporation was frequently measured at different positions on both sides of the forehead, and the possibility of variations in the pattern related to the passage of time was specifically scrutinized. Some typical patterns emerged. The previously reported, marked asymmetries of response (deficient heat-induced sweating and pilocarpine supersensitivity of the symptomatic side) at the medial positions in the forehead were confirmed. However, the asymmetries invariably faded to some extent with the passage of time. Patients with cluster headache show gross similarities with, but also some minor differences from, the sweat pattern of patients with brain stem lesions causing a Horner's syndrome. A subdivision of the material into groups in accordance with the pupillometric pattern after sympathomimetic stimulation made it clear that the cases of definite evaporimetric asymmetries ("typical reactions") belonged to the group with a typical pupillometric pattern. These results suggest that from an "autonomic" point of view, subpopulations may exist within the clinical entity of cluster headache.     

Cluster Headache in Two Sisters. Pupillary Response to Phenylephrine and Tyramine

Two sisters with cluster headache were studied with respect to the pupillary responses to instillation into the conjunctival sac of a single drop of a 1% solution of phenylephrine and a 2% solution of tyramine. The changes in pupillary diameters were documented by photographic pupillometry prior to and at 15, 30, 60, and 90 minutes after the instillations.
Of the two sisters, one (case A) was examined during a symptom-free interval, when she had been free from cluster headache attacks for 2 1/2 years. When the cluster headaches recurred, retesting was performed. The other sister (case B) had been free from cluster headaches for 9 years, when she was examined.
The findings indicate hypofunction within the postganglionic sympathetic nerve fibers during a cluster headache period. The hypofunction is bilateral, and thus, can not be a consequence of the unilateral cluster headache attacks. During remissions, tyramine induces a marked mydriasis, particularly on the symptomatic side, tentatively indicating an excessive release of stored monoamines.     

Horner''s syndrome Sweat gland and pupillary responsiveness in two cases with a probable 3rd neurone dysfunction

Two patients with a Horner's syndrome due to a probable 3rd neurone lesion were examined with regard to the pupillometric and evaporimetric patterns. The results are compared with those found in patients with a Horner's syndrome due to a 1st or 2nd neurone lesion, previously described by our group. Concurring with observations by others, the pupil on the symptomatic side did not dilate at all when stimulated with hydroxy-amphetamine eye drops. Postganglionic dysfunction may thus be distinguished from 1st and 2nd neurone dysfunction. Supersensitivity was present on the side of the Horner's syndrome, both on pupillometry (phenylephrine stimulation) and in the medial part of the forehead at evaporimetry (pilocarpine stimulation). The lateral part of the forehead, however, did not demonstrate deficient evaporation in these postganglionic sympathetic lesions.     

Coexistence of pupillary and heart sympathergic asymmetries in cluster headache

Ten cluster headache patients and 10 healthy controls were subjected to electrocardiographic and pupillometric procedures in a search for cardiac and pupillary sympathergic asymmetry. Sympathergic stimulation was provoked by hyperventilation and by instilling tyramine into both eyes. In the control group, hyperventilation changed neither the T-wave form and polarity nor the QTc. Tyramine provoked an equal mydriasis on the two sides. In cluster headache sufferers, hyperventilation produced changes in the T-wave form and polarity as well as an increase of the QTc due to a disproportionate shortening of the R-R and Q-T intervals. An unequal mydriasis was noted after tyramine instillation due to less marked response on the symptomatic side. The observed electrocardiographic abnormalities are considered an expression of an asynchronous repolarization attributed to a sympathergic asymmetry. It is postulated that both the cardiac and pupillary sympathetic imbalance associated with cluster headache are central in origin.     

Cluster headache: Further observations on the dissociation of pain and autonomic findings

A 51-year-old male cluster headache patient had during five bouts in the course of 11 years always had the headache attacks on the left side. Autonomic abnormalities were, however, present on the right side. Pupillometrically, there was thus a Horner-like syndrome on the right (non-symptomatic) side, with miosis and a relatively more marked dilatation of that eye subsequent to topical application of a directly working sympathomimetic agent (phenylephrine) than after an indirectly working one (hydroxyamphetamine), whereas this was not the case on the symptomatic side. The findings on evaporimetry were not as clear-cut as the pupillometric findings; however, even facial sweating was consistent with a pathologic condition on the right (non-symptomatic) side. A primary dichotomy of pain and autonomic signs (that is, not due to change of side of pain localization) thus seems to be present in this case.     

Chronic paroxysmal hemicrania

Pupillometric studies were carried out in eight patients with chronic paroxysmal hemicrania (CPH) and in age- and sex-matched controls in the basal condition and after instillation of 2% tyramine (CPH, n = 5; controls, n = 17), 1% OH-amphetamine (CPH, n = 6; controls, n = 12), and 1% phenylephrine (CPH, n = 6; controls, n = 17). The pupil on the symptomatic and non-symptomatic sides in CPH patients was significantly smaller in the basal condition than in controls, particularly on the symptomatic side. The mydriatic responses to pharmacologic stimulation were essentially similar on the symptomatic and non-symptomatic sides. An evaporimetric study of the forehead sweat glands, using the body heating and pilocarpine tests, was also carried out in these patients and in age- and sex-matched controls. "Early", "intermediate", and "late" measurements demonstrated symmetry of forehead sweating. The findings for both methods of examination thus contrast with those in cluster headache patients. Pupillometric and forehead sweating patterns therefore suggest differences in the pathogenesis of the two headache entities. These tests may be used to distinguish CPH and cluster headache clinically.     

Chronic paroxysmal hemicrania: From the index patient to the disease

The first patient with chronic paroxysmal hemicrania (CPH), a 41-year-old woman, first seen in 1961, was followed until an adequate treatment was found, 12 years later. Clinically, attack frequency and duration differed widely from the general pattern of cluster headache. Ocular variables, such as intraocular pressure and corneal indentation pulse amplitudes, also differed in our case (clear symptomatic side increment during attacks) and cluster headache. Pupil reactions to directly and indirectly acting sympathicomimetic drugs were also vastly different in our case and cluster headache: no signs of Horner’s syndrome in our patient, while cluster headache exhibits a "Horner-like pattern." In cluster headache, there is a relative hypohidrosis in the forehead on the symptomatic side if body temperature is increased, and a clear hyperhidrosis on direct parasympathomimetic stimulation. This was not so in our case. Indomethacin was highly effective in our case, while "cluster headache drugs," such as ergotamine/sumatriptan, were ineffective. Indomethacin was inactive in cluster headache. Accordingly, our case seemed to differ decisively from cluster headache: CPH had been discovered.     

Dysfunction of the sympathetic nervous system in cluster headache

Ocular sympathetic function was studied in 13 cluster headache patients during and between attacks and several weeks or months after attacks had subsided. The pupillary response to tyramine eyedrops and facial sweating and flushing in response to body heating and to the taste of chilies were also investigated during remission. Pupillary dilatation lag on the symptomatic side persisted between bouts and correlated significantly with loss of thermoregulatory sweating in the lower part of the forehead. In six patients in remission, pupillary dilatation in response to tyramine eyedrops was impaired on the symptomatic side, whereas five patients showed no sign of ocular sympathetic deficit. These findings indicate that incomplete sympathetic deficit persisted on the symptomatic side in a subgroup of cluster headache patients during remission. In most of this subgroup the pattern of sympathetic deficit was consistent with impaired function of postganglionic cervical sympathetic fibres.     

Cervical Sympathetic Deficit in Unilateral Migraine Headache

Pupil diameter was measured during the headache-free interval in 38 migraine sufferers selected from the general community. In each case, at least 70 percent of attacks recurred on the same side. Anisocoria was greater than in 40 control subjects, but miosis was not consistently greater on the usual side of headache. Average pupil diameter was similar in migraine sufferers and controls. In patients with pupillary dilation lag on the usual side of headache, miosis persisted after 4% cocaine eyedrops. These findings suggest that cervical sympathetic outflow was lower on the usual side of headache in a subgroup of migraine sufferers. Pupillary dilatation to tyramine eyedrops was greater in control subjects than in migraine sufferers, consistent with decreased function of post-ganglionic cervical sympathetic fibres. Pupillary dilatation to 1% phenylephrine eyedrops did not differ consistently between the headache and headache-free sides, and was similar in migraine sufferers and controls. Thus, adrenergic supersensitivity of the pupils was not evident in this community sample of migraine sufferers. Vasodilatation or swelling of the arterial wall in the carotid canal could cause minor cervical sympathetic deficit in patients with frequent or severe attacks of migraine. Loss of sympathetic vascular tone could increase vasodilatation and pain during attacks.     

The sweating anomaly in cluster headache

We describe a patient with a typical history of cluster headache for more than 18 years. During the first approximately 10 years of his disease, the pain was right-sided, and pupillometric and evaporimetric measurements indicated a sympathetic deficiency on this same side. However, for the next >6 years, his pain was consistently left-sided, although the signs of sympathetic dysfunction still were more marked on the right side. This was also true for the findings obtained during the interictal period and for the heating test performed within an attack. The implications of this interesting case are discussed. The view that two separate lines of symptom production lead to the pain and the autonomic phenomena seems to be supported by this case history. The cluster headache syndrome may also be a bilateral disorder, with only the weight of balance pointing one way or the other. Finally, the autonomic test results of this patient could reflect an autonomic "scar" in the previous headache side.     

Clinical Subtypes of Cluster Headache and Response to Lithium Therapy

SYNOPSIS
Based on clinical features, a group of 31 patients with cluster headache was divided into 1) episodic cluster headache, with months or years of headache free intervals, 2) chronic cluster headache without significant headache free intervals. Chronic type could be subdivided into primary and secondary varieties, secondary being a transformation from episodic cluster headache. A clinical trial of Lithium Carbonate was undertaken. Blood levels of lithium were determined at regular intervals to monitor the therapeutic dosage. Lithium was found to be an effective prophylactic agent in both episodic and chronic cluster headache patients. The percentage of improvement based on headache index was as follows: 55% of patients had more than 90% improvement; 10% of patients showed 60–90% improvement; 15% of patients showed 25–60% improvement; and 20% no improvement. Effectiveness of lithium was evident in less than a week after the initiation of treatment in those who responded. 55% of patients showed mild side effects such as tremor, nausea, diarrhea, abdominal discomfort and lethargy. Only one patient had serious side effects which needed discontinuation of therapy.The beneficial effect of lithium on cluster headaches appear to be independent of its anti-depressant action. Mechanism of action of lithium in cluster headache is not clear.     

Iris Adrenergic Impairment in Idiopathic Headache

SYNOPSIS
Adrenergic iris neurons were pharmacologically studied in 30 migraineurs, 11 cluster headaches and 30 controls.
The pupillary diameter was measured by photographic technique. When compared with controls, migraineurs showed minor fenfluramine mydriasis and greater, more prolonged guanethidine miosis. No significant differences between the affected and unaffected side were observed by testing with guanethidine in unilateral migraineurs.
A poor concentration and synthesis of noradrenaline in the pupil terminals is hypothesized. Moreover, increased mydriasis observed with phenylephrine suggests alpha adrenoceptor supersensitivity from a defect of neuronal transmitter.
In cluster headaches, only a minor tyramine mydriasis was found in the affected pupil. This suggests low availability of neuronal transmitter or a defect of the synaptic tyramine re-uptake in the affected cluster pupil. A disorder of the iris adrenergic neurons could mirror an impairment in the central aminergic systems of idiopathic headaches.     

Oculosympathetic alterations in migraine patients

Oculosympathetic function was studied in 20 headache-free migraine patients and in 20 controls. Pupillary investigation was performed under basal conditions, and after instillation of tyramine (2%) and phenylephrine (1%) eyedrops. Each test was performed twice shortly after a spontaneous attack and then repeated after 7 and 15 days. In the patients, the normal mydriatic response induced by tyramine was significantly ( p <0.001) reduced and phenylephrine instillation caused a significant ( p <0.01) pupillary dilatation in both the assessments performed shortly after the attack. These abnormal responses were bilateral in all patients and slightly anisocoric in some. They were significantly ( p <0.001 ) more pronounced in the patients who had pain and pronounced vascular features. The reduced oculosympathetic response to tyramine, as well as the hypersensitivity to phenylephrine, was less evident 7 days after the attack and absent after 15 days. A transient and bilateral post-ganglionic oculosympathetic hvpofunction, with adrenoceptor hypersensitivity, was found to be temporally related to the migraine attack, regardless of the side or predominant side of pain.