Articles

CME Internuclear ophthalmoplegia as an

isolated or predominant symptom
of brainstem infarction
Jong S. Kim, MD

Abstract—Objective: To describe the clinical features, MRI findings, and pathogenesis of strokes producing internuclear
ophthalmoplegia (INO) as an isolated or predominant clinical manifestation. Methods: Thirty patients presenting with
INO without (n ⫽ 12) or with (n ⫽ 18) minimal other neurologic signs were studied. MRI and angiogram (conventional in
3, MR angiogram in 27) studies were performed in all of them. Results: Twenty-four presented with INO, whereas six
initially presented with one-and-a-half syndrome that changed into INO. Dorsal brainstem infarcts responsible for the
INO were located in the caudal pons in 4, rostral pons in 8, rostral pons and isthmus in 2, isthmus area in 14, isthmus and
midbrain in 1, and midbrain in 1 patient. Vascular lesions were found in the posterior circulation in 14 patients (47%): at
the basilar artery (BA) in 4, superior cerebellar artery (SCA) in 4, posterior cerebral artery (PCA) in 4, and both the PCA
and the SCA in 1. The INO eventually disappeared in all patients, tending to last longer (p ⫽ 0.05) when it was associated
with other neurologic signs. Conclusions: Isolated or predominant INO is a unique clinical stroke syndrome caused by
small dorsal brainstem infarction. The pathogenesis, however, is heterogeneous, including distal occlusion of small
penetrating arteries, atheromatous branch occlusion from the BA, SCA, or PCA, or major BA occlusion. The functional
prognosis of these patients is excellent, although the INO tends to last longer when there are other neurologic deficits.
NEUROLOGY 2004;62:1491–1496

Internuclear ophthalmoplegia (INO) is a disorder of eye
movements, characterized by adduction impairment com-
bined with contralateral dissociated abduction nystag-
mus.1 Adduction may be preserved during convergence,
and skew deviation, hypertropia on the side of the lesion,
is often present.1,2 INO is caused by a lesion involving the
medial longitudinal fasciculus (MLF) at the brainstem
that connects with ocular nuclei.
The etiology of INO includes multiple sclerosis
(MS), tumor, infection, hydrocephalus, trauma, nu-
tritional or metabolic disorders, and vascular diseas-
es.2,3 Among them, dorsal pontine infarction has been
recognized as an important cause of INO.2-8 However,
infarction presenting INO as an isolated or predomi-
nant symptom has not been studied using a large
number of patients. Moreover, the vascular patho-
physiology leading to isolated INO has not yet been
studied. The purpose of the current study was to
elucidate the clinical, radiologic, and pathophysio-
logic findings in patients with strokes causing INO
as an isolated or predominant neurologic sign.
Patients and methods. Between April 1995 and August 2003,
of 5,315 patients admitted with ischemic stroke, the author iden-
tified 25 consecutive patients (0.47%) presenting with INO with-
out or with minimal other neurologic deficits in the Asan Medical
Center. Because the purpose of this study was to characterize the
patients with INO as a sole or predominant clinical feature, pa-
tients with relatively large brainstem lesions associated with con-
comitant hemiparesis or quadriparesis were excluded. However,
patients with mild ataxia, dysarthria, or sensory symptoms in
restricted body parts were included.
There were four additional patients with isolated INO probably
caused by a vascular event in whom MRI showed no responsible
lesions. Because the etiology of these patients was not certain,
they were excluded. In addition, five patients who visited the
author’s outpatient clinic presenting with isolated INO were in-
cluded. Thus, 30 patients who had MRI-identified lesions became
the subjects of this study. All the patients were initially examined,
and 25 among them were followed up by the author. Five patients
were followed up by other physicians, and their follow-up data
were obtained by chart review.
All patients underwent conventional T2- and T1-weighted MRI
1 to 5 days after the onset of symptoms. Angiograms were also
performed in all patients: MR angiogram in 27 and conventional
angiogram in 3. Diffusion-weighted MRI (single-shot echo planar
spin echo sequence, 1,000 s/mm2, repetition time/echo time 5,000/
139 milliseconds) was additionally performed in 16 patients. In
two of them (Patients 22 and 23), the lesion was identified by
diffusion-weighted MRI but not by T2-weighted MRI, whereas the
lesion was identified by T2-weighted MRI but not diffusion-
weighted MRI in Patient 28. For further analysis, the pons was
divided into caudal (the level showing the facial nerve) and rostral
(the level showing the trigeminal nerve) pons.9 The level of the
pontomesencephalic junction where the small triangular or pen-
tagonal fourth ventricle is seen was designated as the “isthmus.”9

From the Department of Neurology, University of Ulsan, Asan Medical Center, Seoul, Korea.
Supported by research fund from the Korean Ministry of Health and Welfare (03-PJ1-PG1-CH06-0001).
Received October 6, 2003. Accepted in final form January 28, 2004.
Address correspondence and reprint requests to Dr. J.S. Kim, Department of Neurology, Asan Medical Center, Song Pa, PO Box 145, Seoul 138-600, Korea;
e-mail: jongskim@amc.seoul.kr

Copyright © 2004 by AAN Enterprises, Inc. 1491

Table Characteristics of the patients

Patient no./sex/age, y Risk factors Symptoms Ocular findings Exotropia‡ Convergence

1/M/54 HT, SM VT, DIP Lt 11/2-INO ⫹ ⫹

2/M/52 DM, PHS DIP Rt 11/2-INO ⫹ ⫹
3/F/57 HT, DM VT, DIP Lt INO ⫹ ⫹
4/F/70 HT, DM, SM DIP, HA Lt 11/2-INO ⫹ ⫹
5/F/53 HT DZ, NA Lt INO ⫺ ⫹
6/M/57 SM DZ, DIP Lt 11/2-INO ⫹ ⫺
7/F/67 DM VT, DIP Lt INO ⫹ ⫹
8/M/59 HT, DM, Af, PHS DIP Lt INO ⫹ ⫺
9/M/54 HT, DM DIP Lt 11/2-INO ⫹ ⫹
10/M/74 HT, DM DZ, DIP Rt 11/2-INO ⫺ ⫹
11/M/67 SM DIP Rt INO ⫹ ⫹
12/M/45 SM DIP Rt INO ⫺ ⫹
13/M/41 HT DIP Lt INO ⫺ ⫺
14/F/75 DM VT, DIP, NA Lt INO ⫹ ⫺
15/M/65 HT, DM, SM, CHD DIP Lt INO ⫺ ⫹
16/F/69 DM VT, DIP, NA Rt INO ⫹ ⫺
17/F/69 HT DIP, NA, HA Rt INO ⫹ ⫹
18/M/78 HT, DM DZ, DIP Rt INO ⫺ ⫹
19/F/66 HT, DM DZ, DIP, NA, HA Bilateral INO ⫹ ⫹
20/M/82 DM VT, DIP Bilateral INO ⫹ ⫺
21/M/70 HT, DM, SM VT, DIP, NA Lt INO ⫹ ⫹
22/F/50 None VT, DIP Rt INO ⫹ ⫹
23/F/60 HT DIP Rt INO ⫹ ⫺
24/F/65 HT VT, DIP Lt INO ⫹ ⫹
25/M/59 HT, SM, PHS DZ Rt INO ⫺ ⫺
26/M/49 SM DZ, DIP Lt INO ⫹ ⫹
27/F/61 HT DIP Lt DIP ⫹ ⫺
28/F/57 HT, DM VT, DIP, NA Rt INO ⫺ ⫺
29/M/47 HT DIP Lt INO ⫺ ⫺
30/F/70 HT, DM DZ, DIP Rt INO ⫹ ⫺

* Side of hypertropia.
† Lesions responsible for INO.
‡ Of contralateral eye.

INO ⫽ internuclear ophthalmoplegia; HT ⫽ hypertension; DM ⫽ diabetes mellitus; SM ⫽ cigarette smoking; PHS ⫽ past history of
stroke; Af ⫽ atrial fibillation; CHD ⫽ coronary heart disease; DZ ⫽ dizziness; VT ⫽ vertigo; DIP ⫽ diplopia; NA ⫽ nausea; HA ⫽
headache; DA ⫽ dysarthria; LA ⫽ limb ataxia; GA ⫽ gait ataxia; SS ⫽ sensory symptom; C-pons ⫽ caudal pons; R-pons ⫽ rostral
pons; VA ⫽ vertebral artery; BA ⫽ basilar artery; PCA ⫽ posterior cerebral artery; SCA ⫽ superior cerebellar artery; steno ⫽ stenosis;
occl ⫽ occlusion.

For statistics, we used ␹2 tests for comparison of the character-
istics between the patients with isolated INO and those with
concomitant neurologic signs. The t-test was used when the dura-
tion of INO was assessed. All statistical tests were performed with
the use of the SPSS program (version 10.0; SPSS, Chicago, IL); p
values of ⬍0.05 were regarded as indicating significance.
Results. The table summarizes the characteristics of the
patients.
Demography and risk factors of the patients. There
were 17 men and 13 women with age ranging from 41 to 82
years (mean 62 years). Vascular risk factors included hy-
1492 NEUROLOGY 62 May (1 of 2) 2004
pertension in 20, diabetes mellitus in 16, current cigarette
smoking in 9, and atrial fibrillation in 1. Four had a past
history of stroke, and one had a history of coronary heart
disease. Follow-up periods ranged from 1 to 72 (mean 15)
months.
Clinical findings. The most common initial symptom
was diplopia, which was present in 28 patients. It was
usually horizontal and was maximally detected when the
patients were made to see objects on the side opposite to
the impaired adduction. However, vertical or diagonal dip-
lopia was also present in eight patients (Patients 1, 4, 5,
Table Continued

Skew* Other neurologic signs MRI lesions† Other MRI lesion Angiogram findings INO duration, d

⫹, Lt Lt FP, perioral SS t C-Pons, medulla Medulla Both VA distal occl 30

Rt FP, DA Rt C-pons WNL 60
⫹, Lt t FP, DA, GA, Rt hand–foot S Lt C-pons VA-BA junction steno 20
⫹, Lt DA, GA, Rt perioral–hand SS Lt C-pons WNL 360
⫹, Lt Rt hand–foot SS, DA, GA Lt R-pons Medulla BA proximal occl 7
Lt R-pons R-pons WNL 7
⫹, Lt Lt R-pons WNL 30
DA Lt R-pons WNL 30
Lt R-pons R-pons, isthmus WNL 7
⫹, Lt DA, GA, hand–foot SS Rt R-pons R-pons Rt SCA steno 30
Rt R-pons WNL 7
Rt R-pons WNL 5
Lt R-pons, isthmus WNL 60
GA Lt R-pons, isthmus Lt SCA steno 5
DA, GA, perioral SS Lt isthmus Isthmus Both PCA, BA steno 30
GA Rt isthmus Midbrain Rt PCA sten 7
Rt isthmus Rt PCA, SCA steno 30
DA, GA, Lt LA Rt isthmus Midbrain Rt PCA steno 7
Both isthmus WNL 30
DA, GA Both isthmus WNL 150
Rt hand–foot SS, GA Lt isthmus WNL 30
⫹, Rt GA Rt isthmus Midbrain WNL 150
⫹, Rt Lt isthmus Lt SCA steno 30
⫹, Rt Rt face–hand SS Lt isthmus Isthmus Lt SCA steno 7
Rt isthmus Midbrain Rt PCA steno 30
Perioral SS, GA Lt Isthmus Isthmus WNL 2
Lt isthmus 1
⫹, Rt Rt isthmus WNL 2
⫹, Lt DA, Rt LA, GA Midbrain, isthmus WNL 150
⫹, Rt DA, Lt LA, GA Rt midbrain Isthmus WNL 60

22, 24, 26, 28, and 29). Vertigo or dizziness was present in
18 patients, which was usually more severe with head or
body movements. Nausea was present in seven and gener-
alized headache in three patients.
Ocular findings. Twenty-four patients presented with
INO, whereas six initially showed one-and-a-half syn-
drome, which soon changed into INO within several days.
The INO was unilateral in 28 and bilateral in 2 (Patients
19 and 20). Convergence eye movements were preserved in
18 patients (60%). Skew deviation was present in 12 pa-
tients (40%), with the ipsilateral eyes deviated upward in
10 of them. On forward gaze, exotropia of the contralateral
eye was noted in 21 patients (70%). Patient 19 showed
transient downbeat nystagmus, whereas Patient 28 had
upbeat nystagmus. Patient 29 showed transient upgaze
limitation.
Other neurologic dysfunction. In 12 patients (Patients
6, 7, 9, 11 to 13, 17, 19, 23, 25, 27, and 28), INO was the
only neurologic sign, whereas in the other patients, other
minor neurologic signs were present. Aside from INO, nine
patients had sensory symptoms (paresthesia). The symp-
toms were restricted to the perioral area only in three
(Patients 1, 15, and 26), perioral area and hand in two
(Patients 4 and 24), and hand and foot in four patients
(Patients 3, 5, 10, and 21). In these patients, objective
sensory deficit was either absent (Patients 1, 4, 15, and 26)
or mild. Mild dysarthria was present in 11 patients. Gait
ataxia was present in 15, but limb ataxia was present only
in 3 patients (Patients 18, 29, and 30). Lower motor neu-
ron–type facial palsy was present in three patients (Pa-
tients 1, 2, and 3) on the side ipsilateral to the lesions.
Outcome. In all patients, the INO resolved in 1 day to
12 months, usually within 1 month. In Patient 5, the INO
resolved in 1 week but transiently reappeared 10 days
later. During the follow-up period, none had recurrent
strokes. All survived except for Patient 20, who died from
congestive heart failure.
Imaging findings. MRI findings. All patients had in-
farcts, and none had a hemorrhagic lesion. The paramed-
ian dorsal infarcts responsible for the INO were located in
May (1 of 2) 2004 NEUROLOGY 62 1493

Figure. (Top) Patient 1. T2-weighted MRI shows dorsal
infarcts at the lower pons and rostral medullary areas.
Angiogram shows occlusion of the left distal vertebral
artery (white arrow) and retrograde filling of the basilar
artery from the posterior communicating arteries (black
arrow). There was also an occlusion of the proximal por-
tion of the right vertebral artery (not shown here). (Mid-
dle) Patient 10. Diffusion-weighted MRI shows a dorsal
pontine infarction that extends ventrally. MR angiogram
shows focal stenosis of the right superior cerebellar artery
(arrow). (Bottom) Patient 16. T2-weighted MRI shows dor-
sal isthmus lesion that extends to the midbrain area. MR
angiogram shows focal stenosis of the right posterior cere-
bral artery (arrow). Numbers indicate patients’ numbers.
Lesions responsible for internuclear ophthalmoplegia are
marked as R.
the caudal pons in 4 (Patients 1 to 4), rostral pons in 8
(Patients 5 to 12), rostral pons and isthmus in 2 (Patients
13 and 14), isthmus area in 14, isthmus and midbrain in 1
(Patient 29), and midbrain in 1 (Patient 30). The dorsal
pontine lesion of Patient 1 extended to the dorsal medulla.
Aside from the dorsal lesions involving the MLF, addi-
tional ischemic lesions were shown ventral to the reference
lesions in 12 patients (40%), at the same level as (Patients
6, 10, 15, 24, and 26), one level below (Patients 5 and 30),
or one level above (Patients 9, 16, 18, 22, and 25) the
reference lesions (table). The lesions were at either the
paramedian or the lateral territory.
Angiographic findings. In 14 patients (47%), there
were abnormalities in the posterior circulation that were
considered to be responsible for the infarction. Four pa-
tients had basilar artery (BA) diseases; two (Patients 1
and 5) had occlusion of the distal vertebral artery/proximal
BA, and the upper two-thirds of the BA was supplied by
flows from the posterior communicating arteries (figure,
top). Patient 3 had severe proximal BA stenosis, and Pa-
tient 15 showed irregularity of the BA. Atherosclerotic nar-
rowing of the proximal portion of the superior cerebellar
artery (SCA) (Patients 10, 14, 23, and 24) (see the figure,
1494 NEUROLOGY 62 May (1 of 2) 2004
middle), posterior cerebral artery (PCA) (Patients 15, 16,
18, and 25) (see the figure, bottom), or both (Patient 17)
was found in nine patients. In addition, atherosclerotic
changes in vessels unrelated to the patients’ symptoms
were observed in eight patients (Patients 9, 16, 18, 19, 22,
24, 27, and 30).
Presumed pathogenesis. According to the MRI and an-
giographic findings, the presumed pathogenesis was con-
sidered as follows: branch atheromatous disease from the
PCA in three (Patients 16, 18, and 25), from the SCA in
four (Patients 10, 14, 23, and 24), from the BA in four (1, 3,
5, and 15), and from either the PCA or the SCA in one
(Patient 17). Distal occlusion of small penetrating vessels
arising from the BA was a presumed pathogenesis in 11
patients with dorsal pontine infarction without accompa-
nying ventral lesions (Patients 2, 4, 7, 11 to 13, 19 to 21,
27, and 28). The pathogenesis of the patients with accom-
panying ventral lesions with normal angiogram findings
could be an occlusion of the proximal portion of the pene-
trating arteries (Patients 6, 9, 22, 26, 29, and 30). In Pa-
tient 8, who had an atrial fibrillation, whether the small
dorsal pontine infarct was caused by cardiogenic embolism
remains unclear.
Comparison between patients with isolated INO and
those with other minor symptoms. We compared the pa-
tients with isolated INO (n ⫽ 12) and those with other
minor neurologic deficits (n ⫽ 18) to elucidate possible
differences. There were no differences between these two
groups as regards gender, age, number of risk factors, fre-
quencies of vascular abnormalities, and MRI lesions not
related to the INO. However, the duration of INO tended
to be longer (p ⫽ 0.05) in patients with other minor neuro-
logic signs (mean 63.6 ⫾ 89.8 days) as compared with
those with isolated INO (mean 18 ⫾ 17.9 days).
Discussion. This is the largest collection of pa-
tients presenting with INO as the sole or predomi-
nant sign of dorsal brainstem stroke. Previous
reports were small in number, and the pathogenesis
has rarely been discussed.
The most common cause of isolated INO has been
shown to be either stroke or MS.2,10 During the study
period, the author found only four admitted patients
with isolated INO due to MS. Therefore, cerebral
infarction is likely to be the most frequent cause of
isolated INO in this country where MS is rare com-
pared with Europe and North America. Neverthe-
less, our patients occupied only 0.47% of total
admitted patients with ischemic stroke. If we con-
sider purely isolated INO only (n ⫽ 12), the preva-
lence would be reduced to 0.2%.
The incidence of stroke-induced isolated INO may
have been underestimated, considering a previous
study result that only 52% of the ischemic lesions
producing INO were identified by MRI.11 However,
during the study period, only four patients with pure
INO with presumed ischemic cause were found. If
these cases were included, the incidence of stroke-
induced isolated INO would increase to 0.3%. Never-
theless, without documentation of the lesions, the
ischemic etiology cannot be certain. For this reason,
patients with negative MRI findings were excluded
in this study. In Patients 22 and 23, the lesions
detected by diffusion-weighted MRI were not identi-
fied by T2-weighted MRI, whereas a lesion shown
with T2-weighted MRI was not identified in
diffusion-weighted MRI in Patient 28. It seems likely
that repeated MRI would increase the sensitivity of
the diagnosis.
In the majority of our patients, the INO was uni-
lateral. Although bilateral INO has been considered
to be usually caused by MS,1,10 strokes causing bilat-
eral INO have also been described,12 as in our Pa-
tients 19 and 20. Skew deviation was present in 40%
of the patients with hyperdeviation on the side of the
INO. This phenomenon was probably related to the
disruption of otolith– ocular connections.13 Although
the majority of the patients initially presented with
INO due to a selective involvement of the MLF, addi-
tional gaze paresis (one-and-a-half syndrome) was
present at the initial stage in six patients, suggest-
ing that the lesions also involved the paramedian
pontine reticular formation. However, the one-and-a-
half syndrome soon changed into INO, illustrating
that the primarily involved area was MLF in these
patients as well.
Although diplopia and dizziness were the usual
complaints, sensory symptoms in the restricted body
parts were reported in occasional patients. This is
probably caused by partial involvement of the medial
lemniscus located just ventral to the MLF.14 Gait
ataxia and dysarthria may be due to involvement of
the adjacent cerebellar tracts. In addition, a lower
motor neuron–type facial palsy was seen in three
patients with caudal pontine lesions, suggesting that
facial nerve fascicles were involved at this level.
It was noteworthy that aside from dorsal brain-
stem infarcts responsible for INO, MRI showed addi-
tional accompanying ischemic lesions ventral to the
reference lesions in 40% of our patients. The lesions
were located at the level either identical to or differ-
ent from those responsible for INO (see the table and
figure). Although these lesions were not more fre-
quent in patients with other neurologic signs than in
those without, they could be responsible for some of
the nonocular signs such as sensory symptoms.
It has been shown that the tegmental area at the
midbrain level is supplied by circumferential
branches from the PCA or SCA as well as the long
penetrating branches from the BA. At the level of the
rostral pons, the tegmentum is supplied by branches
from the SCA or BA, whereas branches from the BA
or anterior inferior cerebellar artery (AICA) supply
the caudal pontine tegmentum.15,16 It also has been
shown that these branches course obliquely in a cau-
dal direction in the rostral tegmentum, whereas they
run rostrally in the caudal area.15,16 This nonhorizon-
tal direction of the penetrating vessels appears to
explain the different level of accompanying lesions
from that of dorsal infarcts responsible for INO.
In patients with dorsal isthmus lesions, the ven-
tral extension was in the rostral direction in four
patients so as to involve the midbrain (Patients 16,
18, 22, and 25). The ventral lesions were adjacent to
the PCA (Patients 16, 18, and 25), and PCA stenosis
was indeed found in these patients (see the figure,
bottom). On the other hand, focal SCA stenosis was
commonly observed in patients with dorsal rostral or
isthmus lesions that were not accompanied by ven-
tral midbrain lesions (Patients 10, 14, 17, 23, and 24)
(see the figure, middle). Therefore, atheromatous oc-
clusion of branches arising from either the PCA or
the SCA that supply the dorsal isthmus seems to be
responsible for the infarction in these patients.
On the other hand, among the four patients with
caudal pontine lesions, two (Patients 1 and 3) had
significant proximal BA occlusion or stenosis (see the
figure, top). Another patient (Patient 5) with rostral
pontine lesion plus pontomedullary junction infarc-
tion showed a nearly identical vascular pattern as
Patient 1: occlusion of proximal BA with reverse flow
of upper BA through the posterior communicating
arteries. Thus, lower to middorsal pontine infarction
producing isolated INO may be caused by severe
proximal BA disease with relatively good collateral
circulation. Dorsal lesions without accompanying
ventral lesions at the caudal (Patients 2 and 4), ros-
tral (Patients 7, 8, and 11), or isthmus–midbrain
(Patients 19, 20, 21, 27 to 30) level were associated
with normal MR angiogram findings. These lesions
were probably caused by an occlusion of small pene-
trating branches.
Thus, unlike previous reports that isolated INO
was caused by small penetrating artery occlusion,17
our findings illustrate that the pathogenesis of INO
syndrome varies greatly, ranging from small-vessel
occlusion, branch atheromatous disease of PCA,
SCA, or BA, or severe proximal BA stenosis or occlu-
sion. Generally, INO associated with rostrally lo-
cated lesions is related to PCA or SCA diseases,
whereas that with caudally located lesions is associ-
ated with either small-vessel or BA diseases. How-
ever, the interpretation of our data should be made
with caution because evaluation of relatively small
arteries such as the SCA may not be sufficiently
accurate with MR angiography. Moreover, although
we did not find problems in the AICA in any of the
patients, proper evaluation of AICA with MR angiog-
raphy remains difficult. Therefore, the possible con-
tribution of AICA disease cannot be strictly ruled
out.
Finally, the prognosis of INO in our series was
excellent in that it disappeared in all the patients,
usually within 1 month. This result is more favor-
able than that of a previous study, which showed
that the resolution rate of INO was 79%.11 This is
probably related to the fact that we considered only
patients without other major neurologic deficits such
as hemiparesis. Thus, the size of the infarct was
probably smaller than previously reported ones.
Moreover, we found that the duration of INO tended
to be longer in patients with other neurologic signs
as compared with those without, an observation con-
sistent with a previous report.11
May (1 of 2) 2004 NEUROLOGY 62 1495
References
1. Smith JL, Cogan DG. Internuclear ophthalmoplegia: a review of fifty-
eight cases. Arch Ophthalmol 1959;61:687– 694.
2. Leigh RJ, Zee DS. The neurology of eye movements. 3rd ed. New York:
Oxford University Press, 1999:497–511.
3. Bassetti C, Bogousslavsky J, Barth A, Regli F. Isolated infarcts of the
pons. Neurology 1996;46:165–175.
4. Kataoka S, Hori A, Shirakawa T, Hirose G. Paramedian pontine infarc-
tion: neurological–topographical correlation. Stroke 1997;28:809 – 815.
5. Kumral E, Bayülkem G, Evyapan D. Clinical spectrum of pontine in-
farction: clinical–MRI correlations. J Neurol 2002;249:1659 –1670.
6. Atlas SW, Grossman RI, Savino PJ, et al. Internuclear ophthalmoplegia:
MR–anatomic correlation. AJNR Am J Neuroradiol 1987;8:243–247.
7. Awerbuch G, Brown M, Levin JR. Magnetic resonance imaging corre-
lates of internuclear ophthalmoplegia. Int J Neurosci 1990;52:39 – 43.
8. Alexander JA, Castillo M, Hoffman JC Jr. Magnetic resonance findings
in a patient with internuclear ophthalmoplegia. J Clin Neuroophthal-
mol 1991;11:58 – 61.
9. Parent A. Carpenter’s human neuroanatomy. 9th ed. Baltimore: Wil-
liams & Wilkins, 1996:937–982.
10. Leigh RJ, Daroff RB, Troost T. Supranuclear disorders of eye move-
ments. In: Glaser JS, ed. Neuro-ophthalmology. 3rd ed. Philadelphia:
Lippincott Williams & Wilkins, 1999:345–368.
11. Eggenberger E, Golnik K, Lee A, et al. Prognosis of ischemic internu-
clear ophthalmoplegia. Ophthalmology 2002;109:1676 –1678.
12. Takamatsu K, Ohta T. A case of infarction in the pontine tegmentum
involving the bilateral MLF and unilateral paramedian pontine reticu-
lar formation. Clin Neurol 1995;35:277–281.
13. Brandt T, Dieterich M. Skew deviation with ocular torsion: a vestibular
brainstem sign of topographic diagnostic value. Ann Neurol 1993;33:
528 –534.
14. Kim JS, Bae YH. Pure or predominant sensory stroke due to brainstem
lesion. Stroke 1997;28:1761–1764.
15. Parent A. Carpenter’s human neuroanatomy. 9th ed. Baltimore: Wil-
liams & Wilkins, 1996:93–128.
16. Tatu L, Moulin T, Bogousslavksy J, Duvernoy H. Arterial territories of
human brain: brainstem and cerebellum. Neurology 1996;47:1125–
1135.
17. Caplan LR. Posterior circulation disease: clinical findings, diagnosis,
and management. Cambridge, MA: Blackwell Science, 1996:381– 443.

ACCESS www.neurology.org NOW

FOR FULL-TEXT ARTICLES
Neurology online is now available to all subscribers. Our online version features extensive search capability by title key
words, article key words, and author names. Subscribers can search full-text article Neurology archives to 1999 and
can access link references with PubMed. The one-time activation requires only your subscriber number, which
appears on your mailing label. If this label is not available to you, call 1-800-638-3030 (United States) or 1-301-714-
2300 (outside United States) to obtain this number.

1496 NEUROLOGY 62 May (1 of 2) 2004