Medical treatment of trigeminal neuralgia and cluster headache

Document Sample
Medical treatment of trigeminal neuralgia and cluster headache Powered By Docstoc
					MARK           W. GREEN,                 M.D.

Medical                                                     treatment    of trigeminal
neuralgia                                                     and cluster headache
Thgeminal             neuralgia          and      cluster      headache         are both     facial       cutaneous           triggers         that are in different         divisions,  and in
pain syndromes              that are paroxysmal                    in nature,     and are com-            some       cases      the triggers      were      located    quite far from the facial
monly       confused         in clinical          practice.       This discussion          willde-        pain.      In fact,     there is a case reported             ofa trigger in a toe that
scribe these two syndromes            and their treatment   as well as                                    could      reproducibly           trigger      a paroxysm        in the face.
comparing       and contrasting    them. Since they are extraordi-
narily  painful    conditions,   early diagnosis     leading to effec-                                    Etiology
tive treatment     is very important.                                                                     There has been a great deal ofdiscussion       and argument                                  over
                                                                                                          the etiology oftrigeminal  neuralgia.    Heredity   certainly                               does
Thgeminal     neuralgia                                                                                   not seem to play a role, and there is a strong     body of cvi-
Most fibers of the trigeminal        nerve are sensory     and inner-                                     dence to suggest that the site of pathology   is located in the
vate the face and the frontal portions ofthe nasal cavity. Tri-                                           nerve root entry zone, the spot where myelin         changes    char-
geminal neuralgia,      or tic douloureux,     is an excruciatingly                                       acter from peripheral    to central myelin.2     Therefore,      most
painful paroxysmal      form of facial pain that typically begins                                         cases seem to be caused by distended         vascular      loops that
in patients over the age of 40. ‘ It tends to affect the second                                           compress   the nerve root entry zone at this location.       Some of
and then the third division     of the nerve and then a combina-                                          the other etiologies              (Table 1), if they are relevant,     probably
tion of the two; involvement         of the first division    alone is                                    cause theirpathology                in the same anatomic     location.    Forex-
quite     rare.      The pain        is unilateral,           and characteristically           con-       ample,       in multiple          sclerosis,       which     has long been included
sists of repetitive   superficial  jabs                       of abrupt onset, common-                    in the differential diagnosis                    oftrigeminal      neuralgia, patients
ly precipitated     by very superficial                         cutaneous   stimuli, such                 who were studied,      although                     they may     have had plaques      in
as shaving,   touching  the face, or chewing.     It was not rare in                                      multiple       areas,     all had plaques at the nerve                 root entry         zone.3
the days before treatment    was available    for patients  with se-                                      Dry      sockets       or temporomandibular          joint             dysfunction            are
vere     forms       of trigeminal              neuralgia        triggered      by chewing           to   probably           not important   causes   of trigeminal                   neuralgia.
starve     themselves             orat    least     lose a great       deal of weight,         rath-
er than       eat.    The painful          paroxysms             usually     do not last more             Examination
than a few seconds                   to a few minutes.                 Spontaneous         remis-         The neurologic                 examination           of patients       with        trigeminal
sions are common                   and so is clustering                of attacks,       which is         neuralgia          is typically        normal,       although        some       have      argued
responsible           for some           of the confusion              between         trigeminal         over extremely     sensitive ways of evaluating      facial sensation
neuralgia   and cluster headache.                                                                         and these show some alterations.       If the bedside examination
   The trigger areas are typically,   but not necessarily,    locat-                                      is not normal, or if patients experience       hypalgesia    or hyper-
ed in the same distribution     of the trigeminal     nerve as the                                        algesia  or background      pain between     paroxysms      of trigem-
pain (Figure 1). Patients   have been reported    with superficial                                        inal neuralgia,           then    it is reasonable          and customary              to under-

DECEMBER              1985’VOL26’NO                   12 (SUPPLEMENT)                                                                                                                                        19
Medical             treatment

take a search                 for symptomatic                 trigeminal          neuralgias          in
which        the pain          is a manifestation               of other        neurologic         dis-
                                                                                                                              Table 1-Lesions    That Can
                                                                                                                                   Chronically Irritate
                                                                                                                           Trigeminal Nerve and Result in
Mechanism                                                                                                                        Trigeminal Neuralgia
How irritation             of the trigeminal              nerve      in this location           might
cause       these        typical       repetitive        paroxysms             has been         much
speculated          upon.          There is probably   a combination                       of ectop-                 Arterial      loop                                     Carotid       syphon
ic action        potentials          in the area of demyelinization                        with con-                 Vascular           malformation                        Acoustic         neuroma
current        failure      of segmental             inhibition       that ultimately            leads               Epidermoid             cyst                            Glioma
to a discharge             within      the trigeminal             nucleus      and an attack         of
                                                                                                                     Multiple       sclerosis          plaque               Chronic oral or
typical      tic douloureux              (Figure        2).
                                                                                                                     Fibrous       bands        across                      dental disease
                                                                                                                     petrous       ridge
Traditional          analgesics          are ineffective            and are contraindicat-
ed in the treatment                 of trigeminal    neuralgia.   Most                   of the med-
ications that work                  have an intrinsic    anticonvulsant                      activity.     mon in patients over the age of 40-the      same group of pa-
It is interesting,             however,         that phenobarbital,                  an effective          tients most likely to have trigeminal     neuralgia-and    this
anticonvulsant,   has no effect whatsoever  on trigeminal   neu-                                           makes it more important   to look for other, less toxic agents
ralgia. The most common        agent to be used is carbamaze-                                              that could            be used. Another    problem  is that carbamazepine
pine, in a dosage that produces therapeutic   blood levels. It is                                          stimulates             its own metabolism     and may make steady state
effective        in about          90% of patients,           depending           on what study            management                   ultimately        more     difficult.        Phenytoin         is also
is read.4 The most serious                     complications             of carbamazepine,                 used, being effective   in about 60% of patients,56 although      it
particularly             the hematologic              ones,       seem       to be more         com-       is clear that to obtain relief with phenytoin     requires levels
                                                                                                           that are in the higher therapeutic                        range and this is a range in
                                                                                                           which patients are more likely                          to develop  ataxia and other
                                                                                                           side effects.
                                                                                                               There       is a small literature  in the use ofbaclofen, which is
                                                                                                           otherwise         used in spasticity.7    Several authors have spoken
                                                                                                           of the use of chlorphenesin,’                         which          is used   otherwise       as a
                                                                                                           skeletal       muscle           relaxant,       but my        own      experience       with   this
                                                                                                           agent has been disappointing.
                                                                                                              All of the studies of clonazepam                                  to date have       involved
                                                                                                           relatively        numbers
                                                                                                                                small                      of patients.  Court and Kase9 did a
                                                                                                           drug trial on 25 patients,                    using dosages of6 to 9 mg/d. They
                                                                                                           reported that 23 patients                     had significant    relief, including 10
                                                                                                           who had complete      relief.                  Interestingly,   in the same study 16
                                                                                                           of the 25 patients  had been refractory    to carbamazepine
                                                                                                           treatment. The major side effects reported   were gait distur-
                                                                                                           bance and somnolence.
                                                                                                                            reported a similar study of 119 trigeminal
                                                                                                               Chandra’#{176}                                             neu-
                                                                                                           ralgia patients,     with 68% success-success       being total or
                                                                                                           partial      relief     of tic symptoms.              Swerdlow”           did a study of par-
                                                                                                           oxysmal   pain and reported  similar     efficacy of carbamaze-
                                                                                                           pine and clonazepam.   However,     patients were given the op-
                                                                                                           tion of choosing   a second drug if they preferred,   and if these
                                                                                                           data are evaluated   in that regard the side effects of carbama-
                                                                                                           zepine were so much higher than those of clonazepam            that
                                                                                                           ultimately            a much       higher      percentage           of patients     chose don-
FIGURE            1-Trigger            areas        in trigeminal           neuralgia.                     azepam         over carbamazepine                    (Table     2).

20                                                                                                                                                                                   PSYCHOSOMATICS
Cluster         headache                                                                                       cally retro-orbital                       or temporal,  in contradistinction                                        to tn-
Cluster  headache    is another     facial pain syndrome.    It has                                            geminal    neuralgia,                      which tends to be in the second                                       or third
been described    in the literature     under a variety of names,                                              divisions    of the trigeminal    nerve.   However,     there is a sub-
which       really       indicate         the author’s             speculations          as to the ori-        group     of cluster   headaches,     known    as   ‘lower    half syn-                     ‘

gin ofthe        pain. Many call it a neuralgia-greater                                    superficial         drome,     in which most of the pain is maxillary.
                                                                                                                               ‘ ‘                                    This latter
petrosal        neuralgia,  Vidian   neuralgia,        Vail’s                            neuralgia-            syndrome     is very commonly  confused    with trigeminal   neu-
while      others        have felt that cluster                    headache         was just       a var-      ralgia.
iant      of migraine              and therefore      a vascular    headache   syn-                                The pain of cluster headache    is severe, so much so that
             2 Horton’3            called it histamine    cephalalgia    on the ba-                            Horton          found a substantial  suicide   rate in his series                                                   of pa-
sis of his observation                      that he could             reliably       precipitate         at-   tients.        The pain builds up rapidly,   but over minutes,                                                     not in-
tacks      with small amounts    of histamine.   However,                                        we have       stants,          and          is sometimes                   described                 as       “knife-like”               or
since      known that any vasodilator     can do the same.                                      The term       ‘   ‘boring.          ‘ ‘    Patients          will    often        say        that      they       feel     as though
cluster      headache             was introduced               by Kunkle             and associates            they      ‘   ‘have a hot poker in the eye’                               ‘    when they develop an at-
in 1952.’                                                                                                      tack.         Lacrimation   and rhinonrhea                                    of the ipsilateral side are
    Cluster headaches   occur predominately          in men, with a                                            common.                     They        also     exhibit         a different                    type       of behavior
male: female ratio of about 8:1. They tend to begin after the                                                  from          that seen in trigeminal                         neuralgia.              Trigeminal               neuralgia
age of 21 or so. Sufferers   experience     clustering    of attacks                                           patients are very protective                                   of their           face.          Men often grow
during which time they tend to have one or more attacks per                                                    beards for fear of shaving;                                    women               often         wear veils over
day,      day    after        day for a period               of two       to eight        weeks,         but   their faces and do not go out on windy                                                   days. Cluster                head-
there is a great deal of variation.   There is also great variabil-                                            ache patients tend to pace up and down                                                 for the duration                of an
ity in the duration   of the attacks,    though they tend to last                                              attack.         This also distinguishes                             them from               migraineurs,                who
from       30 to 90 minutes.                     The    pain is always            unilateral,       typi-      typically              prefer to lie down and go to sleep. The pain may
                                                                                                               terminate               abruptly  or subside gradually.  Between clusters
                                                                                                               patients              are       typically             free     of     headaches                    for      periods        of
                                                                                                               months     or even years.
                                                                                                                   Triggers of attacks, besides     vasodilators                                             such as alcohol,
                       Chronic           irritation      of trigeminal        nerve        ]                   include     REM sleep, over-sleeping,
                                                                                                               but attacks will occur with or without
                                                                                                                                                                triggers    during     the
                                                                                                                                                                                                           daytime  napping,

                                                                                                               cluster period.  Patients   with cluster   headaches      have a much
                       Ectopic                                [        Failure      of                         higher incidence     of ulcer disease    than the general        popula-
                       action                                          segmental                               tion,         and this incidence     is highest in patients with the lower

                L      potentials                                      inhibition
                                                                                                                                           in cluster
                                                                                                                                                         artery disease
                                                                                                                                                                          also has a higher
                                                                                                                                                                                   patients             than          in the general
                                                                                                               population.                 Additionally,                    a substantial number                            of cluster
                                                                                                               headache                patients develop                     marked bradycardias                             and coro-
                     Paroxysmal              discharge  of interneurons
                     in trigeminal            nucleus oralis
                                                                                                               nary sinus rhythms                          during           the course               of an        tta’6          Table 3
                                                                                                               outlines              and compares                some ofthe                  clinical          characteristics            of
                                                                                                               cluster headache       and trigeminal     neuralgia.
                                                                                                                   Sjaastad and associates’7 described           another                                              variant  called
                               Firing of nociceptive                                                           chronic     paroxysmal        hemicrania.        Chronic                                                paroxysmal
                               trigeminothalamic     neurons                                                   hemicrania      is seen largely    in women,       whereas                                              typical    clus-
                                                                                                               ter headache is seen largely in men. The attacks are other-
                                                                                                               wise identical, except that they are briefer  and more fre-
                                                                                                               quent.          Sufferers               may experience                        ten to 20 attacks                 per day,
                                   Attack         of tic douloureux
                                                                                                               lasting an average of 15 minutes each. Corneal       indentation
                                                                                                               pulse patterns8   are generally elevated in patients with clus-
                                                                                                               ter headaches    and are therefore  considered       a marker  of this
FIGURE               2-Proposed                   mechanism:             Combination               of in-      condition.    And these are elevated     in patients    with chronic
creased         affferent          activity,           decreased         segmental          inhibition         paroxysmal    hemicrania,                               just     as they              are in other              forms      of
leads      to paroxysms                 of trigeminal          neuralgia.                                      cluster headache.

 DECEMBER              1985      VOL
                               #{149}      26’    NO     12 (SUPPLEMENT)                                                                                                                                                                  21
Medical        treatment

Pathology     site                                                                             tion     would       also explain       why       patients          are so sensitive           to va-
Localizing      the site of pathology      in cluster    headache       has                    sodilators.
been difficult.     One could speculate,        however,       that it lies                        Ekbom          and Greitz’9       performed            an angiogram              on a patient
within the carotid system.        This would explain the partial                               having  a cluster headache attack, and were able to demon-
Homer’s     syndrome      that typically     accompanies         a cluster                     strate what they felt was a constriction within the canal of
headache.    This partial Homer’s        syndrome     spares sweat fi-                         the carotid         artery,      although     others        said that this could               repre-
bers, which would ascend along the external             carotid,    so one                     sent a paroxysmal     edema of a vessel wall and compress                                           the
could anatomically       place a lesion     here and claim that the                            ascending   sympathetic    plexus on that basis.
pain is due to stimulation     of the sphenopalatine                    ganglion,                     Other     investigators        say that the problem                   could     be as well
which is adjacent   to the external   carotid.’2    Such               a localiza-             explained          by a parasympathetic                  paroxysm   at the level of the
                                                                                                                              brainstem,                 rather than a sympathetic
                                                                                                                                   block.       Such       a paroxysm           might     cause        a
                                                                                                                                   Homer’s             syndrome,         although        it would
                            Table 2-Response     to First Dru9
                                                                                                                                   not explain            why      we always          see a par-
                           Received for Trigemlnal Neuralgia
                                                                                                                                   tial Homer’s   syndrome.      Stimulation
                                                                                                                                   of the greater   superficial      petrosal
                                                                                 No. transferred to                                nerve could explain    the tearing,     nasal
                                                                                     other drugs                                   congestion,             and      injection        of the eye
                                           No.                No.                                                                  that occur commonly       as concomitants
                                       receiving          relieved           Because             Because                           to the pain. Furthermore,       stimulation
                                        as first           by first           of side            of inef-                          of the vagus             nerve       could       explain        the
      Drug                                drug            drug (%)          eftects(%)           ficacy(%)
                                                                                                                                   bradycardia      and the elevated  gastrin
                                                                                                                                   levels,     which are now recognized      as
      Carbamazepine                         37             11(30)               19(51)                7(19)
                                                                                                                                   being     responsible             for the increased             in-
      Clonazepam                            35             23(66)                 6(17)               6(17)
                                                                                                                                   cidence   ofulcer disease.     These gastrin
      Phenytoin                             47             22(47)                13(28)          12(26)                            levels are often extraordinarily     high-
      Valproate                             51             20(39)               14(27)           17(33)                            in the range seen in the Zollinger-Elli-
      Total                               170              76                   52               42                                son syndrome.    An acetylcholine-like
      Reproduced     with permission     from Swerdlow”                                                                            substance  has been found in the spinal
                                                                                                                                   fluid     immediately                following             a dis-
                                                                                                                                   charge        of cluster         headache.

                                                                                                                                   A number     of treatments                        previously
                           Table 3-Clinical  Characteristics
                                                                                                                                   touted for cluster headache                        have been
                                 of Cluster Headache
                                vsTrigemlnal  Neuralgia                                                                            shown         to be          ineffective.          Antihista-
                                                                                                                                   mines      alone are ineffective,      as is hista-
                                                                                                                                   mine      desensitization.      Analgesics       are
      Cluster      headache                                         Trigeminal       neuralgia
                                                                                                                                   ineffective            and       contraindicated.               Er-
      Dull retro-orbital                                            Paroxysmal     shooting            pain,                       gotamine  administered      in                       the same
      or boring pain                                                confined   to trigeminal                                       way as it is for a migraine                            attack is
                                                                    territory                                                      generally,  though not always,    ineffec-
      Builds in intensity
      to plateau;   lasts 15 mm                                     Pain has instantaneous                                         tive, and since these attacks  are so fre-
      to two hr; slow                                               onset, brief duration;    may                                  quent,        the     safe      weekly       dosage        of er-
      defervescence                                                 occur as repetitive    jabs                                    gotamine  would easily be exceeded.
                                                                                                                                   At the same time, various prophylac-
      No cutaneous          trigger                                 Cutaneous        trigger
                                                                                                                                   tic agents           seem       to be the most             effec-
      Associated       autonomic                                    No associated         autonomic
                                                                                                                                   tive treatments.                Methysergide           is prob-
      symptoms                                                      symptoms
                                                                                                                                   ably the most                   widely  used            agent.
                                                                                                                                   Methysergide’s                  most feared            side ef-

22                                                                                                                                                                      PSYCHOSOMATICS
fects, retroperitoneal, pleural, pericardial, and subendocar-                                                      region       and inner       canthus       ofthe      eye.     Clonazepam             was effec-
dial fibrosis, do not occur within the two- to three-month                                                         tive in preventing             pain    in fourofsix            patients.      The usual          dos-
treatment           period         in cluster          headaches              as they        might with            age required     was 1 .5 to 3 mg, which was well tolerated                                           in
longer-term              treatment              for migraine.           Propranolol            is not ef-          these patients.    Caccia     did a small study on the efficacy                                      on
fective        in treating   cluster    headaches   and probably      is con-                                      clonazepam      in tnigeminal    neuralgia,   finding it effective                                    in
traindicated         because    of the bradycardia    produced     by the at-                                      five of six patients.    At the same time,  he studied    two pa-
tacks.        In acute attacks,       100% oxygen      inhaled   at 7 to 12                                        tients with cluster   headaches   and found that in itself don-
L’min         acts as a rapid vasoconstrictor      and is highly success-                                          azepam         was ineffective,            but when          an unidentified            antihisti-
ful in aborting          attacks.               Lidocaine,         4%, nose drops to anes-                         minic drug was added,   clonazepam     was rapidly                                     effective.
thetize        the sphenopalatine                   ganglion        may also be useful.#{176}                      As mentioned  before, except for cyproheptadine,                                        antihista-
    Lithium carbonate                      has probably    become                  the treatment     of            mines        alone     are clearly       not of any use in cluster                  headaches.
choice for prophylaxis                       ofcluster  headaches.2’                  It was origin-
ally      used      as a treatment                for a variant          of cluster           headaches            Relationship             to trigeminal             neuralgia
called chronic cluster, in which                               the attacks          do not occur              in   In 1978, Green and Apfelbaum        described     several    patients
clusters,  but which is otherwise                               identical          to cluster            head-     who had clear crossover   syndromes       between    cluster     head-
ache.2’          Most     patients         respond          to 300 to 900           mg/d.          The     cal-    ache     and trigeminal             neuralgia.         Raskln          presented         an inter-
cium channel      blockers  currently     available                                are not very             ef-    esting series around the same time of what he called “ice-
fective in treating cluster    headaches.                                                                          pick’  headaches,
                                                                                                                            ‘              which referred to almost half of his typi-
       In chronic         cluster         headache           lithium          is again       the    drug      of   cal migraine      headaches.    If questioned,  patients reported
choice.           Methysergide                  is rather      ineffective.              Interestingly,            frequent   bursts-typical                 neuralgic          pains-around              the orbits
indomethacine                  is nearly            100%       effective           in blocking              the    associated    with and                occasionally            unassociated            with their
headaches    ofchronic     paroxysmal     hemicrania.22     T’pically                                        it    migraine         attacks.
will increase     the frequency      of migraines      in migraine                                         pa-        Thus,        there clearly appears                 to be a relationship               between
tients.        It has been              tried     in women             with      otherwise          typical        vascular         headache    syndromes                 and typical     brief           neuralgic
cluster        headaches,               and seems       more likely to work in women                               pains.       Whether        or not clonazepam                is an effective         drug     in the
than        with episodic
          men              cluster    headache.                                                                    vascular   headache    syndromes,      particularly  cluster                                  head-
    Sminne and Scarlato’      treated    patients    with Sluder     syn-                                          ache, remains   to be seen, but clearly it is a major agent                                   in the
drome, which was really what was described                 by others as                                            prophylaxis            of   trigeminal           neuralgia          and     other      neuralgic
lower   half syndrome,    pain located       largely  in the maxillary                                             headache         syndromes.                                                                           D

 1 . cushing        H: The major         trigeminal  neuralgias         and their surgical treatment                   1956.
      based       on experiences             with 332 Gaserian          operations.   Am J Med Sci                 14. Kunkle E, Pfeifter JB, Wilhott W, et al: Recurrent                brief headaches         in ‘clus-
        160(2):157-184,        1920.                                                                                   ter’ pattem. Trans Amer Neuro/ Assoc 77:240, 1952.
 2.    Jannetta    P: Arterial compression          ofthe trigeminal        nerve at the pons in pa-               15. Ekbom K; Patterns       ofcluster     headache      with a note onthe relationsto             angi-
       tients withtrigeminal       neuralgia.     J Neurosurg        26:159-162,       1967.                           na pectoris and peptic ulcer. Acta Neuro/Scand46:225-237,                             1970.
 3.    Jenson TS, Rasmussen            P, Reske-Nielsen           E, et al: Association       of trigem-           16. Ekbom K: Heart rate, blood pressure              and electrocardiographic               changes
        inal neuralgia    with multiple sclerosis.         clinical    and pathological         features.              during provoked       attacks ofcluster      headache.         Acta Neuro/Scand46:215-
        Acta NeurolScand65:182-1              89, 1982.                                                                224, 1970.
 4.    Blom S: Trigeminal         neuralgia:      Its treatment       with a new anticonvulsant                    17. Sjaastad     0, Apfelbaum        R, caskey      W, et al: chronic          paroxysmal        hemi-
        drug. Lancet 1 :839-840,         1962.                                                                         crania(CPH).       Ups J Med Sci(suppl)31           :27-33, 1980.
 5.    Braham J, Sare A: Phenytoin            in the treatment       of trigeminal      and other neu-             18. Broch A, Horven I, Nornes H, et al: Studies on cerebral                    and ocular circula-
        ralgias. Lancet2:892-893,          1960.                                                                       tion in a patientwith     cluster headache.         Headache          10:1-8, 1970.
 6.    BIum S: Tic douloureux          treated with a new anticonvulsant.                  Arch Neurol             19. Ekbom K, Greitz J: Carotid          angiography       in cluster headache.           Acta Radio/
       9:285-290,      1963.                                                                                           (Diagn) 10:177-186,        1970.
 7.    Fromm G, Terrence cF, chattha                A: Treatment of face pain with baclofen.                       20. Kittrelle JP, Grouse D, Seybold M, et aI: Cluster headache,                      local anesthe-
        Trans Am Neurol Assoc 105:486-488,                  1980.                                                      tic abortive agents. Arch Neuro/42:496-498,                  May1985.
 8.    Dalessio     D: Chlorphenesin          for trigeminal        neuralgia.      JAMA 225;1659,                 21. Kudrow L: Lithium prophylaxis             in chronic      cluster headache.           Headache
        1973.                                                                                                          17:15-1 8, 1977.
 9.    court JE, Kase c: Treatment of tic douloureux                    with a new anticonvulsant                  22. Sjaastad    0, Dale I: A new headache              entity, ‘chronic         paroxysmal       hemi-
        (clonazepam).       JNeuro/ Neurosurg           Psychiatry      39:297-299,       1976.                        crania.’ Acta Neuro/Scand54:140-159,                  1976.
10.    Chandra      B: The use of clonazepam             in the treatment       of trigeminal     neural-          23. Smirne 5, Scarlato G: Clonazepam              in cranial neuralgias.          MedJ Aust 1:93-
       gia. Mod Med Asia 11:8-9, 1975.                                                                                 94, 1977.
11.    Swerdlow      M: Anticonvulsant         drugs used in the treatment               of lancinating            24. caccia     MR: clonazepam          in facial neuralgia        and cluster headache.             Eur
       pain: A comparison.        Anaesthesia        36:1129-1132,          1981.                                      Neuro/ 13:560-563,        1975.
12.    Kunkle E: Significance        of minor eye signs on headache                  of migraine type.             25. Green M, Apfelbaum            A: Cluster-tic      syndrome,         abstracted.       Headache
       Arch Ophthalmol65:504-508,               1961.                                                                  18(suppl):1     12, 1978.
13.    Horton B: Histaminic          cephalalgia.        Proc Staff Meet Mayo C/in 31:325,                         26. Raskin NH: Icepick pain. Neurology              29:550, 1979.

 DECEMBER               1985     VOL 26’ NO 12 (SUPPLEMENT)
                               #{149}                                                                                                                                                                                      23