Usefulness of MR Imaging in Idiopathic Oculomotor Nerve Palsy Cases: T2-weighted IDEAL.

Introduction

Patients with oculomotor nerve palsy (ONP) have clinical symptoms of diplopia and/or blepharoptosis. Herein, we identified damaged oculomotor nerves with T2-weighted (T2W) interactive decomposition of water/fat using echo asymmetry and least-squares estimation (IDEAL) imaging,[1] a kind of fat-suppression imaging, and a variant of 3-point Dixon method,[2] in idiopathic ONP patients.

Case Reports

We encountered eight patients with idiopathic ONP and used T2W IDEAL imaging (Signa HDxt 3.0T; GE Healthcare, Milwaukee, WI, USA) to visualize their oculomotor nerves. The repetition and echo times of the T2W images shown were 3400–4483.34 ms and 81.79–86.3, respectively, in IDEAL images. Radiologists were not given any information about which side of oculomotor nerve was impaired. We reviewed their clinical symptoms and results of their examinations (Table 1). Here, we present a normal control (Fig. 1a) and two representative cases.

Table 1

Clinical characteristics of patients with idiopathic oculomotor nerve palsy with abnormal MRI findings

	Case 1	Case 2	Case 3	Case 4	Case 5	Case 6	Case 7	Case 8
Age (M/F)	43 (F)	23 (M)	59 (F)	55 (F)	84 (F)	40 (F)	55 (M)	59 (M)
Prodromal infection	Yes	No	No	No	No	Yes	No	No
Initial symptoms	Diplopia	Diplopia	Blepharoptosis	Diplopia	Blepharoptosis, Diplopia	Diplopia	Diplopia	Blepharoptosis, Diplopia
Oculomotor abnormality
Affected side(s)1	L2	R	R	L	R < L3	L	L	R
Anisocoria	Yes	No	No	No	No	No	No	No
Light reflex	Impaired	Impaired	Impaired	No	No	No	No	Impaired
Eye movement	Impaired	Impaired	Impaired	Impaired	Impaired	Impaired	Impaired	Impaired
Blepharoptosis	No	No	Yes	No	Yes	No	No	Yes
Other neurological findings	No	No	No	No	No	No	No	No
Abnormal immunity
Autoantibodies	Yes	Yes	No	No	No	Yes	No	No
Diabetes mellitus	No	No	No	No	No	No	No	No
CSF findings (cells/μl/protein [mg/dl])	6/31	3/25	9/32	0/45	0/48	n.d.4	3/70	0/39
MRI findings of the oculomotor nerve
High signal on T2W IDEAL image (R/L)	−/L	R > L	R/−	−/L	R < L	−/L	−/L	R/−
Location of high signal	A–C5	A–C	A–C	A–C, CS6	A–C	A–C, CS	A–C, CS	A–C
Enhancement with Gd (R/L)	−/L	R > L	R/−	−/L	n.d.	n.d.	−/L	R/−
Abnormality of extraocular muscles	Yes	Yes	No	Yes	Yes	No	Yes	Yes
Response to steroid
Outcome	Good7	Complete8	Complete	Good	Good	Good	Good	Complete
Tx9 courses required	4	3	3	3	3	Oral PSL10	1	5
Recurrence	No	No	No	No	Yes (2 m)	No	No	Yes (3 y)
MRI findings after Tx	Improved	Unchanged	Improved	Unchanged	Unchanged	n.d.	n.d.	Unchanged

Only the side of the oculomotor nerve(s) showing abnormal findings are indicated.

L and R indicate the side of the oculomotor nerve affected.

R < L indicates the left oculomotor nerve affected more severely than the right, and R > L indicates vice versa.

n.d.: not done.

A–C: the marginal area between the orbital apex and cavernous sinus.

CS: the cavernous sinus area.

Good,

Complete: We defined the terms to evaluate the responsiveness to steroid pulse therapies in the acute phase; “complete” meant recovery to their original oculomotor function, “good” meant recovery without diplopia in daily life.

Tx: therapies.

PSL: prednisolone 20 mg/day. CSF, Cerebrospinal fluid; Gd, Gadolinium; IDEAL, Interactive Decomposition of water/fat using Echo Asymmetry and Least-squares estimation.

Fig. 1

The repetition and echo time of the T2-weighted (T2W) images shown were 3400–4483.34 ms and 81.79–86.3, respectively, in T2W IDEAL images. (a) MRI T2W IDEAL images (a serial of coronal section) in a normal control. Each arrowhead indicates oculomotor nerves and are demonstrated as iso-signal intensities. V1: ophthalmic nerve, V2: maxillary nerve, VI: abducensnerve. (b) MRI T2W IDEAL image (coronal section) of the head in Case 1—the left oculomotor nerve demonstrated high signal intensity. (c) MRI T2W IDEAL image (coronal section) of the head in Case 2—the left oculomotor nerve demonstrated high signal intensity. (d) MRI post-contrast T1W IDEAL image (coronal section) in Case 2. The right oculomotor nerve demonstrated contrast enhancement. The TR and TE were 583 and 10 ms, respectively, in post-contrast fat-saturated T1-weighted (T1W) image. (e) The right medial (MR) and inferior rectus (IR) muscles also demonstrated high signal intensities.

Case 1

A 43-year-old female was admitted because of double vision. A 1.5T brain MRI did not detect any abnormalities. The patient’s eye movement was impaired in all directions. In T2W IDEAL images obtained with 3T MRI, the left oculomotor nerve itself showed high signal intensity (Fig. 1b). We found gadolinium enhancement of the left oculomotor nerve.

Case 2

A 23-year-old male was admitted due to acute-onset double vision. We confirmed the right eye’s outward deviation. T2W IDEAL images revealed that 15 mm of the right oculomotor nerve was swollen with high signal intensity from the cavernous sinus to the orbital apex (Fig. 1c) and that the nerve showed gadolinium enhancement (Fig. 1d). The right medial (MR) and inferior rectus (IR) muscles demonstrated high signal intensity on T2W IDEAL imaging (Fig. 1e).

Discussion

Upon MRI, all eight idiopathic ONP patients showed high signal intensities on their oculomotor nerves. We administered gadolinium to six of the ONP patients and all of them showed enhancement. Steroid pulse therapies which we prescribed for seven of the idiopathic ONP patients were effective. However, after their therapies, only two ONP patients showed improvement in their oculomotor nerves upon MRI (Table 1).

Although the oculomotor nerve is less than 1 mm thick, we were able to detect the affected oculomotor nerves with higher signal intensity in idiopathic ONP patients by using T2W IDEAL images in thin slices (2.5–3.0 mm) of coronal sections. Compared to short-tau inversion recovery and fluid-attenuated inversion recovery imaging, damaged oculomotor nerves in ONP patients showed higher intensity than other orbital tissues, another healthy oculomotor nerve, and optic nerves on T2W IDEAL images. This protocol also showed increased signal intensity of extraocular muscles which might have reflected denervation of the oculomotor nerve (Fig. 1e). We also found that gadolinium-enhanced images were useful to detect damaged oculomotor nerves.[3] All of the damaged oculomotor nerves indicated high intensity from the orbital apex to the cavernous sinus. The narrow structure and venous plexus are anatomical reasons why the oculomotor nerve can be damaged by any type of inflammation and occlusion. Furthermore, we suspect that an aberrant immune response might have been involved in ONP patients because steroid pulse therapy was effective.