Journal of Neuro-Ophthalmology, September 2002, Volume 22, Issue 3

Oculomotor Ophthalmoplegic Migraine

Oculomotor ophthalmoplegic migraine is a rare episodic childhood condition in which a unilateral oculomotor palsy is preceded by headache. I describe six new cases that had magnetic resonance imaging signal abnormalities during the acute phase, consisting of a thickened and enhancing ipsilateral oculomotor nerve at its exit from the midbrain. During the quiescent phase, when the headache had resolved, the signal abnormalities were still present but less dramatic. Seventeen similar cases have been previously reported. The pathophysiology may be a trigeminovascular migraine epiphenomenon that is dependent on the unique oculomotor nerve anatomy and porous blood-nerve barrier at the emergence of the oculomotor nerve from the brainstem and the sequelae of demyelination. Early high-dose corticosteroid treatment is recommended to rapidly resolve an acute episode and to potentially prevent permanent abnormal oculomotor nerve signs.

THE HOYT LECTURE r William F. Hoyt, MD Editor's Note: William Fletcher Hoyt, MD, professor emeritus of Ophthalmology, Neurology, and Neuro­surgery, University of California, San Francisco, was born and raised in Berkeley, California. He took his undergraduate education at the University of Cali­fornia, Berkeley, and his medical education at the University of California, San Francisco ( UCSF). After a year's study at the Wilmer Institute, Johns Hopkins University, under the mentorship of Frank B. Walsh, MD, he returned to UCSF in 1958 to found the neuro- ophthalmology service. During a 36- year aca­demic career- all of it at UCSF- he authored 266 journal articles, co- authored ( with Frank B. Walsh, MD) the biblical third edition of Clinical Neuro-ophthalmology, and trained 71 neuro- ophthalmology fellows. In 1983, he received the title of Honorary Doctor of Medicine from the Karolinska Institute. He is widely acknowledged as one of the titans of twentieth century neuro- ophthalmology. In recognition of his contributions, the North American Neuro- Ophthalmology Society ( NANOS), in conjunc­tion with the American Academy of Ophthalmology, in 2001 initiated the Hoyt Lecture to be delivered each year at the Annual Meeting of the American Academy of Ophthalmology. Oculomotor Ophthalmoplegic Migraine: Is It Really Migraine? Thomas J. Carlow, MD Oculomotor ophthalmoplegic migraine is a rare episodic childhood condition in which a unilateral oculomotor palsy is preceded by headache. I describe six new cases that had magnetic resonance imaging signal abnormalities during the acute phase, consisting of a thickened and enhancing ipsilateral oculomotor nerve at its exit from the midbrain. During the quiescent phase, when the headache had re­solved, the signal abnormalities were still present but less dramatic. Seventeen similar cases have been previously re­ported. The pathophysiology may be a trigeminovascular migraine epiphenomenon that is dependent on the unique oculomotor nerve anatomy and porous blood- nerve barrier at the emergence of the oculomotor nerve from the brain­stem and the sequelae of demyelination. Early high- dose corticosteroid treatment is recommended to rapidly resolve an acute episode and to potentially prevent permanent ab­normal oculomotor nerve signs. ( JNeuro- Ophthalmol 2002; 22: 215- 221) Professor of Neurology, Ophthalmology, and Radiology, University of New Mexico, School of Medicine, Albuquerque, New Mexico. Address correspondence to Thomas J. Carlow, MD, University of New Mexico, School of Medicine, 302 Juniper Hill Road NE, Albuquerque, NM 87122, USA; E- mail: tjcarlow@ swcp. com Dr. William F. Hoyt's dedication to excellence, his pre­cise scholarship and distinguished bibliography are universally recognized. Sir William Osier aptly de­scribed another dimension of Dr. Hoyt with this quote: " No bubble is so iridescent or floats longer than that blown by the successful teacher" ( 1). Dr. Hoyt's contribution to neuro- ophthalmology is unprecedented when you con­sider the number of fellows and residents that he has taught, motivated, and inspired to enter academic medicine. Seventy- one one- year fellows from the United States, Canada, and 23 other countries throughout the world have trained with Dr. Hoyt; 52 are full professors of neurology, ophthalmology, or neurosurgery. He has educated 50 fel­lows for periods of less than a full year, and 86 neurology, ophthalmology, and neurosurgical residents from outside the University of California, San Francisco program. Eleven of his former fellows have held chairs in neurology, neurosurgery, and ophthalmology. His students include a president of the American Academy of Ophthalmology, a president of the American Neurological Association, four of the past six presidents of the North American Neuro- Ophthalmology Society ( NANOS), and numerous neurology and ophthalmology journal editors and associate editors. When one recognizes that these are just the first _ J Neuro- Ophthalmol, Vol. 22,, Na^ 3, Copyright © Lippincott WIIIIE ,2002 „ , „ , . , . . . , DOI: 10.1097/, 01. WNP, 0000030330.98308.33, 215 , ams & Wilkins. Unauthorized reproduction of this article is prohibited. JNeuro- Ophthalmol, Vol. 22, No. 3, 2002 THE HOYT LECTURE Thomas J. Carlow, MD, born in Cincinnati, Ohio, received his undergraduate degree from Xavier University and his medical degree from the University of Cincinnati School of Medicine, where he was admitted to the Alpha Omega Alpha Society. He completed a medical internship at the Henry Ford Hospital, Detroit, and 2 years in the United States Public Health Service before starting his training in neurology at the University of Michigan. After a year's neuro- ophthalmology fellowship under the aegis of Joel S. Glaser, MD, and Robert B. Daroff, MD at the University of Miami, he joined the faculty of the University of New Mexico in Albuquerque, where he is currently Professor of Neurology, Ophthalmology, and Radiology. He has con­tributed important journal articles on chronic progressive ophthalmoplegia, acute multifocal pigment epitheliopathy, lupus erythematosus, pseudotumor cerebri, migraine, ver­tigo, carotid- cavernous fistulas, stroke, and headache. In 1975, he began an annual neuro- opthalmology meeting that became the Rocky Mountain Neuro- Ophthalmology Society, and eventually, the North American Neuro- Ophthalmology Society ( NANOS). He was president of NANOS from 1980 to 1992, board chairperson from 1992 to 1996, and has served as executive vice president since 1996. Through his efforts and those of his wife Susan, NANOS has set the tone for other neuro- ophthalmological societies around the world. generation of Hoyt scholars, his legacy is unparalleled. It is a distinct personal and professional honor to be invited by the American Academy of Ophthalmology and NANOS to deliver the first William F. Hoyt Lecture. Definitions Notta contributed the first clinical description of oph­thalmoplegic migraine in 1854 ( 2,3). In 1884, Mobius termed it " periodic oculomotor paralysis" ( 4). In 1890, Charcot coined the term " ophthalmoplegic migraine" ( 5). Mobius required that a structural process involve the oculomotor nerve, whereas Charcot did not. The concept of ophthalmoplegic migraine has been controversial since the original Mobius and Charcot debate ( 3). In a three- volume text in 1969, Walsh and Hoyt stated, " At the root of the problem has been a lack of strict criteria for the clinical diagnosis of ophthalmoplegic mi­graine and insufficient knowledge of the pathophysiologic events that occur during a migraine attack" ( 6). I intend to document that children with the clinical criteria for ocu­lomotor ophthalmoplegic migraine also have specific magnetic resonance imaging ( MRI) findings, to propose a hypothesis to explain the pathophysiology, and to sug­gest treatment. The generally accepted clinical criteria for oculomo­tor ophthalmoplegic migraine are: 1) childhood onset; 2) headache preceding and ipsilateral to the third nerve pare­sis; 3) a commonly dilated pupil; 4) ophthalmoplegia that may be permanent and rarely accompanied by aberrant ocu­lomotor regeneration; 5) a minimum of two episodes; and 6) no evidence for a structural lesion ( 7,8). The incidence of the condition, as so defined, is estimated to be 0.7 in 1,000,000 ( 9). My Series Over the past 7 years, I have examined six children who fulfilled these clinical criteria. The exception was that all children had abnormal MRI scans at the oculomotor midbrain exit, suggesting a focal pathologic process. In my series, five of six patients were girls. The age of onset varied from 16 months to 9 years, with a mean of 3.7 years. Head­ache preceded the onset of a third nerve paralysis by up to 11 days. The pupil was dilated in 5. Each child had at least two and as many as 12 events. Permanent oculomotor pare­sis or paralysis was present in four of the six children after multiple episodes, and two developed aberrant regenera­tion. Four of the six children had headaches unassociated with oculomotor paresis. A family history of migraine was present in five. MRI brain signal abnormalities were present in all cases. In the scans obtained during the acute phase, when the children had headache and ophthalmoplegia, non-contrast Trweighted images documented thickened ipsilat­eral oculomotor nerves at the midbrain exit that were isointense with brain ( Fig. 1 A). Contrast Trweighted MRI scans showed enhancement of the ipsilateral oculomotor nerves at the midbrain exit ( Figs. IB and 2). A trapezoid ^ 216 „.. , „ . . © 2002 Lippincott Williams & WUkins , Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. THE HOYT LECTURE JNeuro- Ophthalmol, Vol. 22, No. 3, 2002 Fig. 1. Axial T-, MRI before ( A) and after contrast ( B) through the midbrain at the exit of the oculomotor nerve in a child with right oculomotor ophthalmoplegic migraine. Precontrast study ( A) shows thickened right oculomotor nerve ( arrow) that enhances intensely after contrast ( B, arrow). configuration was typically seen, with the widest area adja­cent to the midbrain. MRI scans were also acquired during the quiescent phase when the children no longer had headache or ophthalmoplegia, between one and eight months ( average = 3.5 months) after an acute episode. All noncontrast T r weighted MRI scans continued to show thickened ipsilateral third cranial nerves at the brainstem exit ( Fig. 3). The con­trast scans demonstrated persistent oculomotor nerve en­hancement that was less intense than during the acute phase ( Figs. 4 and 5). Cases with Abnormal MRI Scans In addition to my six cases, 17 patients have previ­ously been reported to have MRI abnormalities for a total of 23 cases ( 10- 19). All have exhibited ipsilateral third nerve MRI abnormalities at the midbrain exit. Combining my cases and previously reported cases, there have been 17 Fig. 2. Sagittal Tn postcontrast MRI through the oculomo­tor nerve brainstem exit in a child during an acute phase of ophthalmoplegic migraine. The clinically involved oculo­motor nerve shows marked contrast enhancement at its midbrain exit ( arrow). girls and six boys, with an age of onset ranging from 7 months to 19 years ( average = 6 years). A partial clinical history is available in 17 of the 23 children who have had abnormal MRI scans ( 10- 14,16- 19) ( Table 1). Headache occurred simultaneously or preceded the onset of the oculomotor paresis by 11 days in 15 cases ( average = 3.3 days). Six children have had a persistent partial or complete third nerve paresis. In three of the six permanent palsy cases, the ophthalmoplegia developed after the first episode; in the remaining three cases, it devel­oped after multiple episodes. Two children from my series eventually demonstrated oculomotor nerve aberrant Fig. 3. Coronal Tn precontrast MRI through the brainstem exit of the oculomotor nerves in a child that had developed right oculomotor ophthalmoplegic migraine nearly three months earlier. No clinical abnormalities were evident at the time of this scan. The right oculomotor nerve ( large arrow) is thickened and is several times larger than the nor­mal left third nerve ( small arrow). Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. JNeuro- Ophthalmol, Vol. 22, No. 3, 2002 THE HOYT LECTURE J • A ' \ jfr V" < w \ jp* 1 ^ * N « , • . Fig. 4. Axial T-, contrast- enhanced MRI through the brain stem exit of the oculomotor nerves in a child with right oculomotor ophthalmoplegic migraine. ( A) Acute phase. The right oculomotor nerve enhances intensely ( arrow). ( B) Quiescent phase. Six weeks after the acute episode, the right oculomotor nerve enhances less dramatically ( arrow). regeneration, which was not reported in the other 15 cases. The time to resolve the oculomotor paresis, when it did clear, ranged from four days to twelve weeks ( average = 4.1 weeks). The pupil was dilated in 15 of 16 cases. There was a family history of migraine in 10 of 15; headaches occurred without ophthalmoplegia in 13 of the 15 cases that docu­mented these findings. Twenty- two of the 23 reported cases ( including my own) with MRI abnormalities have had abnormal Tj - weighted MRI scans during the acute phase. Twenty re­ceived contrast and all showed enhancement of the oculo­motor nerve at the midbrain exit. Two children had only noncontrast MRI scans, and both showed a thickened third nerve at the brainstem exit. Twenty children had abnormal MRI scans during the quiescent period, from 5 weeks to 8 months ( average = 2.6 months) after the acute phase. Of the 17 who had contrast MRI scans, all demonstrated reduced residual oculomotor nerve enhancement at the midbrain exit. Three children had only noncontrasted MRI scans, which showed a thickened oculomotor nerve at the brainstem exit. In my opinion, the diagnostic criteria for oculo­motor ophthalmoplegic migraine should include an MRI contrast- enhanced thickened third nerve at the oculomotor midbrain exit during the acute phase with less enhancement during a quiescent phase. If these MRI findings are not present, other diagnostic etiologies must be excluded ( 20). Nine children, including two from my series, have had recurrent oculomotor paresis without headache. They had all been initially diagnosed as having an " ophthal­moplegic migraine variant" ( 12,16,21- 23). The average age of onset was 20 months. Seven of the 9 children later developed headache episodes followed by an oculomotor paresis, at an average age of 5 years. Five of these seven children had contrast MRI scans, all demonstrating Fig. 5. Axial T-, contrast enhanced MRI through the ocu­lomotor nerve brainstem exit in a child with right oculo­motor ophthalmoplegic migraine. ( A) Acute phase. The right oculomotor nerve enhances intensely at the brainstem exit ( arrow). ( B) Quiescent phase. Four months after an acute episode, the right oculomotor nerve enhances less dramatically ( arrow). _ Zlo. , . © 2002 Lippincott Williams & Wilkins , Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. THE HOYT LECTURE JNeuro- Ophthalmol, Vol. 22, No. 3, 2002 TABLE 1. Clinical information in Case if 1 2 3 4 5 6 7- 12 13 14 15 16 17 18 19 20 21 22 23 Study Stommel et al. ( 10) Straube et al. ( 11) Ostergaard et al. ( 12) Ostergaard et al. ( 12) Aers et al. ( 13) Wong and Wong ( 14) Mark et al. ( 15) tt Ramelli et al. ( 16) Prats et al. ( 17) Prats et al. ( 17) Hupp et al. ( 18) Lance and Zagami ( 19) Carlow Carlow Carlow Carlow Carlow Carlow Sex M F F F F M 4F, 2M F F M F F F M F F F F reported Age at onset 18y 19y 18 mo 7 mo 14y 6y 20 mo ny 3y 5y 9 mo 2y 3y 18 mo 18 mo 9y 5y cases of ophthalmoplegic i Headache to 3r d , d 2 4 3- 4 1 1 4 4 6 5 NR 3- 4 11 2 2- 3 3 Simultaneous onset 1 Time to resolve ord* NR 6 wks 4.5- 6 wks 4 wks 1 wk 12 d 2- 4 wks 4 NR 8 wks 3 mo 4 wks 4 wks 4 wks 2 wks 2 wks migraine and abnormal MRI scans Persistent 3rd NR Yes No Yes No No No No No NR No No Yes Yes Yes Yes No Aberrant 3rd regeneration NR NR NR NR NR NR NR NR NR NR NR No Yes No Yes No No Pupil dilated Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes ( 3r d episode) Yes Yes Yes No Yes Headache without 3rd Yes Yes Yes ? Yes Yes Yes Yes NR Yes Yes Yes No Yes No Yes Yes Family history migraine Yes NR No No Yes Yes Yes No No NR Yes Yes Yes Yes No Yes Yes * When the third nerve paresis did resolve. f No Clinical History Available. J Includes one case from ( 20). NR, Not Reported. oculomotor nerve enhancement at the brainstem exit. ( They are included among the 23 abnormal MRI cases previous­ly discussed.) Treatment Multiple drug regimens have been used to treat or prevent ophthalmoplegic migraine with variable to poor re­sults. Corticosteroids have been prescribed for 12 reported children ( 10- 12,14,17,21,22,24,25), only six of whom have had MRI scans. The response to treatment was not docu­mented or considered of no benefit in six; however, the dose or length of treatment was not noted. A beneficial re­sponse was reported in six, including one child from my series ( 12,14,22,24). Prompt resolution of the headache and ophthalmoplegia appeared related to the early institution of corticosteroids ( prednisone ormethylprednisone) and to the amount prescribed. The minimum effective dose was equivalent to 2 mg of prednisone/ kg/ d given over several weeks with a slow taper. In the six patients who improved with corticosteroids, the resolution time was typically mea­sured in days. Among those not treated, resolution occurred in weeks to months. Six of the 12 corticosteroid- treated children had MRI scans; all displayed a thickened third nerve at the midbrain exit. Three of the six children with abnormal MRI scans had a beneficial response to corticosteroids when prescribed at the onset of the ophthalmoplegia and headache. In the three cases with abnormal MRI scans that did not respond to 219 Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited JNeuro- Ophthalmol, Vol. 22, No. 3, 2002 THE HOYT LECTURE treatment, corticosteroids were given for only a few days or after the headache had resolved. Pathogenesis Why do these children have a thickened oculomotor nerve at the brainstem exit that enhances after contrast on MRI during the acute phase and becomes less intense after contrast during the quiescent phase? Possible pathogenetic mechanisms include compression ( 7), ischemia ( 26), a vas­cular anomaly ( 27), a Tolosa- Hunt variant ( 9,11,22), or de-myelination ( 19). I propose a hypothesis for the pathophysiology of oculomotor ophthalmoplegic migraine based on the trigeminovascular theory of migraine, the unique oculomo­tor nerve anatomy at the brainstem exit, the blood- nerve barrier, and the pathology of demyelination. Current theory suggests that there is a trigeminovas­cular basis for migraine ( 28). At the onset of a migraine, the ophthalmic division of the trigeminal nerve is triggered to release neuropeptides at intracranial sites innervated by the ophthalmic branch of the trigeminal nerve ( 28). A sterile inflammatory vascular response is induced. The greatest in­tracranial trigeminal receptor density is located in the arter­ies that comprise the circle of Willis and the proximal ad­jacent arteries emanating from the circle ( 29). The oculomotor nerve is the only cranial nerve adja­cent to the circle of Willis at its exit from the brainstem. The vascular supply to the oculomotor nerve is derived from the posterior aspect of the circle of Willis and adjacent vessels, which are abundantly innervated, by the ophthalmic divi­sion of the fifth nerve ( 30). That the oculomotor nerve is frequently penetrated by the circumflex mesencephalic ar­tery or a perforating vessel of the posterior cerebral artery may also be significant ( 31). The blood- nerve barrier is dependent on tight closure of capillary endothelial vessels. Neuropeptides do not nor­mally cross tight blood- brain barrier or blood- nerve barrier junctions. The blood- nerve barrier junction is, however, relatively porous at nerve root junctions ( 32). Schwann cells begin to myelinate the oculomotor nerve approximately 0.6 mm from the brainstem exit ( 33). Demyelination causes Schwann cell proliferation and onion bulb formation. With repeated episodes of demyelination and remyelination, there is nerve hypertrophy and axonal loss ( 34). These pathologic findings have been reported in diabetic third nerve pareses ( 35- 37). I propose that the oculomotor nerve enlargement in oculomotor ophthalmoplegic migraine is initiated by a migraine stimulus of the trigeminovascular system. Neuropeptides are secreted at the level of the circle of Willis and adjacent vessels that cross a relatively open blood- nerve barrier junction at the oculomotor nerve exit. A sterile inflammation is induced that further opens the blood- nerve barrier. Demyelination results in Schwann cell proliferation and edema in the oculomotor nerve as it emerges from the brainstem. A contrast- enhanced third nerve at the midbrain exit is then visible on MRI. With re­peated demyelination and remyelination, the oculomotor nerve enlarges focally. Less contrast enhancement is seen during the quiescent period when the blood- nerve bar­rier returns to a relatively normal state and the edema re­solves. Subsequent third nerve compression from nerve hy­pertrophy and scar formation after repeated episodes of de­myelination and remyelination could result in permanent oculomotor paralysis or aberrant regeneration. Postcontrast MRI demyelinative nerve root enhancement has been re­ported in the Guillain- Barre syndrome ( 38) and chronic inflammatory demyelinative polyneuropathy ( CIDP) ( 39,40). Proximal spinal nerve root demyelination with hy­pertrophy ( 41) and gadolinium enhancement is seen in CIDP. Contrast enhancement of the spinal nerve root re­solves during remission ( 40). Corticosteroids may decrease matrix metalloproteinases at the oculomotor- midbrain junction and tighten the relatively porous blood- nerve bar­rier ( 42, 43). Prompt resolution of a third nerve paresis would be the ultimate result. In conclusion, I would formally like to thank Dr. Hoyt for the tremendous influence he has had on my life and career, for being an incredible role model, for making NANOS possible, and for his life's work in neuro- ophthalmology. REFERENCES 1. Osier W. Address at the opening of the Pathological Institute of the Royal Infirmary in Glasgow, Oct 4,1911. Quoted by Cushing H, ed. Life of Sir William Osier. Vol 2. London: Oxford at the Clarendon Press, 1925: 295. 2. Notta A. Memoire sur les lesions functionneles qui sont sous la dependence des neuralgies. Arch gen med 5 me ser 1854; 4: 1- 29,290- 316. 3. Bruyn GW. Complicated migraine. In: Vinken PJ, Bruyn GW, eds. Handbook of Clinical Neurology. New York: American Elsevier, 1968: 59- 95. 4. Mobius PJ. Cber periodische wiedertechrende Oculomotoriuslah-mung. Berl klin Wschr 1884; 21: 604- 8. 5. Charcot JM. Sur un cas de migraine ophtalmoplegique ( paralysie oculo- motrice periodique). Prog Med ( Paris) 1890; 31: 83- 6; 32: 99- 102. 6. Walsh FB, Hoyt WF, eds. Clinical Neuro- Ophthalmology. 3rd ed. Baltimore: Williams & Wilkins 1969: 1674. 7. Walsh JP, O'Doherty DS. A possible explanation of the mechanism of ophthalmoplegic migraine. Neurology 1960; 10: 1079- 84. 8. Headache Society Classification Committee of the International Headache Society. Classification and criteria for headache disor­ders, cranial neuralgias and facial pain. Cephalalgia 1988; 8( suppl 7): 1- 96. 9. Hansen SL, Borelli- Moller L, Strange P, et al. Ophthalmoplegic mi­graine: diagnostic criteria, incidence of hospitalization and possible etiology. Acta Neurol Scand 1990; 81: 54- 60. 10. Strommel EW, Ward TN, Harris RD. MRI findings in a case of ophthalmoplegic migraine. Headache 1993; 33: 234- 7. 11. Straube A, Bandmann O, Buittner U, et al. A contrast enhanced lesion of the III nerve on MR of a patient with ophthalmoplegic ^ 220. , „ . . , „ . . © 2002 Lippincott Williams & WUkins , Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. THE HOYT LECTURE JNeuro- Ophthalmol, Vol. 22, No. 3, 2002 migraine as evidence for a Tolosa- Hunt syndrome. Headache 1993; 33: 446- 8. 12. Ostergaard JR, MollerHU, Christensen T. Recurrent ophthalmople­gia in childhood: diagnostic and etiologic considerations. Cephala­lgia 1996; 16: 276- 9. 13. Aers I, VanZandijcke M, Dehaene I, et al. Magnetic resonance im­aging in a case of migraine with ophthalmoplegia. Eur J Neurol 1997; 4: 85- 9. 14. WongV, Wong WC. Enhancement of oculomotor nerve: a diagnos­tic criterion for ophthalmoplegic migraine? Pediatr Neurol 1997; 17: 70- 3. 15. Mark AS, Casselman J, Brown D, et al. Ophthalmoplegic migraine: reversible enhancement and thickening of the cisternal segment of the oculomotor nerve on contrast- enhanced MR images. AJNR Am JNeuroradiol 1998; 19: 1887- 91. 16. Ramelli GP, Vella S, Lovblad K, et al. Swelling of the third nerve of a child with transient oculomotor paresis: a possible cause of oph­thalmoplegic migraine. Neuropediatrics 2000; 31: 145- 7. 17. Prats JM, Mateos B, Garaizar C. Resolution of MRI abnormalities of the oculomotor nerve in childhood ophthalmoplegic migraine. Cephalalgia 1999; 19: 655- 9. 18. Hupp SL. Migraine. In: Miller NR, Newman NJ, eds. Walsh and Hoyt's Clinical Neuro- ophthalmology. 5th ed. Vol 3. Baltimore: Williams & Wilkins, 1998: 3681- 7. 19. Lance JW, Zagami AS. Ophthalmoplegic migraine: a recurrent de-myelinating neuropathy? Cephalalgia 2001; 21: 84- 9. 20. Blake PY, Mark AS, Kattah J, et al. MR of oculomotor nerve palsy. AJNR Am JNeuroradiol 1995: 16: 1665- 72. 21. Durkan GP, Troost BT, Slamovits TL, et al. Recurrent painless ocu­lomotor palsy in children. A variant of ophthalmoplegic migraine? Headache 1981; 21: 58- 62. 22. Kandt RS, Goldstein GW. Steroid- responsive ophthalmoplegia in a child. Diagnostic considerations. Arch Neurol 1985; 42: 589- 91. 23. Chabriat H, Levasseur M, Schaison M, et al. Migraine Ophthal-moplegique. Rev Neurol ( Paris) 1990; 146: ll, 682- 6. 24. Terrence CF, Samaha, FJ. The Tolosa- Hunt syndrome ( painful oph­thalmoplegia) in children. Develop Med Child Neurol 1973; 15: 506- 9. 25. Woody RC, Blaw ME. Ophthalmoplegic migraine in infancy. Clin Pediatrics 1986; 25: 82^ 1. 26. Vijayan N. Ophthalmoplegic migraine: ischemic or compressive neuropathy? Headache 1980; 20: 300- 4. 27. Imes RK, Monteiro ML, Hoyt WF. Ophthalmoplegic migraine with proximal posterior cerebral artery vascular anomaly. J Clin Neuro Ophthalmol 1984; 4: 221- 3. 28. Moskowitz MA. The neurobiology of vascular head pain. Ann Neu­rol 1984; 16: 157- 68. 29. Sanchez del Rio M, Moskowitz MA. The trigeminal system. In: Olesen J, Tfelt- Hansen P, Welch KM, eds. The Headaches. 2nd ed. Philadelphia: Lippincott Williams & Wilkins, 2000: 141- 9. 30. Cahill M, Bannigan J, Eustace P. Anatomy of the extraneural blood supply to the intracranial oculomotor nerve. Brit J of Ophthalmol 1996; 80: 177- 81. 31. Milsavljevi M, Marinkovi S, Loli- Dragani V, et al. Oculomotor, trochlear, and abducens nerves penetrated by cerebral vessels. Mi­croanatomy and possible clinical significance. Arch Neurol 1986; 43: 58- 61. 32. Olsson Y. Topographical differences in the vascular permeability of the peripheral nervous system. Acta Neuropathol 1968; 10: 26- 33. 33. Tarlov IM. Structure of nerve root. II. Differentiation of sensory from motor roots; observations on identification of function in roots of mixed cranial nerves. Arch Neurol Psychiatry 1937; 37: 1338- 55. 34. Thomas PK, Landon DN, King RHM. Diseases of the peripheral nerves. In: Graham DI, Lantos PL, eds. Greenfield's Neuropathol­ogy. 6th ed. Vol 2. London: Oxford University Press, 1997: 367- 487. 35. Dreyfus PM, Hakim S, Adams RD. Diabetic ophthalmoplegia: re­port of a case, with postmortem study and comments on vascular supply of human oculomotor nerve. Arch Neurol Psychiat 1957; 77: 337^ 19. 36. Asbury AK, Aldredge H, Hershbery R, et al. Oculomotor palsy in diabetes mellitus: a clinico- pathologic study. Brain 1970; 93: 555- 66. 37. Weber RB, Daroff RB, Mackey EA. Pathology of oculomotor nerve palsy in patients with diabetes. Neurology 1970; 20: 835- 8. 3 8. Gorson KC, Ropper AH, Muriello MA, et al. Prospective evaluation of MRI lumbosacral nerve enhancement in acute Guillain- Barre syndrome. Neurology 1996; 47: 813- 7. 39. Morgan GW, Barohn RJ, Bazan III C, et al. Nerve root enhancement with MRI in inflammatory demyelinating polyneuropathy. Neurol­ogy 1993; 43: 618- 20. 40. Kuwabara S, Nakajima M., Matsuda S, et al. Magnetic resonance imaging at the demyelinative foci in chronic inflammatory demye­linating polyneuropathy. Neurology 1997; 48: 874- 7. 41. De Shilva RN, Willson HJ, Doyle D, et al. Nerve root hypertrophy in chronic inflammatory demyelinating polyneuropathy. Muscle Nerve 1994; 17: 168- 70. 42. Kieseier BC, Seifert T, Giovannoni G, et al. Matrix metalloprotein-ases in inflammatory demyelination: targets for treatment. Neurol­ogy 1999; 53: 20- 5. 43. Rosenberg GA, Dencoff JE, Correa N Jr, et al. Effect of steroids on CSF matrix metalloproteinases in multiple sclerosis: relation to blood- brain barrier injury. Neurology 1996; 46: 1626- 32. Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.