Optic, trigeminal, and facial neuropathy related to anti-neurofascin 155 antibody

Abstract

Objective: To characterize the frequency and patterns of optic, trigeminal, and facial nerve involvement by neuroimaging and electrophysiology in IgG4 anti-neurofascin 155 antibody-positive (NF155⁺ ) chronic inflammatory demyelinating polyneuropathy (CIDP).

Methods: Thirteen IgG4 NF155⁺ CIDP patients with mean onset age of 34 years (11 men) were subjected to neurological examination, blink reflex, and visual-evoked potential (VEP) testing, and axial and/or coronal T2-weighted head magnetic resonance imaging (MRI).

Results: Among 13 patients, facial sensory impairment, facial weakness, and apparent visual impairment were observed in three (23.1%), two (15.4%), and two (15.4%) patients, respectively. All 12 patients tested had blink reflex abnormalities: absent and/or delayed R1 in 11 (91.7%), and absent and/or delayed R2 in 10 (83.3%). R1 latencies had strong positive correlations with serum anti-NF155 antibody levels (r = 0.9, P ≤ 0.0001 on both sides) and distal and F wave latencies of the median and ulnar nerves. Absent and/or prolonged VEPs were observed in 10/13 (76.9%) patients and 17/26 (65.4%) eyes. On MRI, hypertrophy, and high signal intensity of trigeminal nerves were detected in 9/13 (69.2%) and 10/13 (76.9%) patients, respectively, whereas optic nerves were normal in all patients. The intra-orbital trigeminal nerve width on coronal sections showed a significant positive correlation with disease duration.

Interpretation: Subclinical demyelination frequently occurs in the optic, trigeminal, and facial nerves in IgG4 NF155⁺ CIDP, suggesting that both central and peripheral myelin structures of the cranial nerves are involved in this condition, whereas nerve hypertrophy only develops in myelinated peripheral nerve fibers.

Figure 1

Representative blink reflex traces in a patient with IgG4 NF155⁺ CIDP. Compared with a disease control 63‐year‐old woman with multifocal acquired demyelinating sensory and motor neuropathy (A), the latencies of R1 and ipsilateral and contralateral R2 are clearly prolonged in Case 4 with IgG4 NF155⁺ CIDP (B).

Figure 2

Representative traces of visual‐evoked potentials in a patient with IgG4 NF155⁺ CIDP. Compared with a disease control 40‐year‐old man with atopic myelitis (120.9 ms, upper raw in A), the right eye of Case 5 with IgG4 NF155⁺ CIDP shows delayed P100 15’ latency (136.8 ms, lower raw in A). Case 5 does not show any optic nerve abnormalities on MRI (B and C).

Figure 3

Hypertrophy and signal abnormality of trigeminal nerves in patients with IgG4 NF155⁺ CIDP. Head MRI of a 53‐year‐old female disease control patient with neuromyelitis optica spectrum disorder (NMOSD) (1–6) and of NF155⁺ CIDP patients (Case 8, 7–13; Case 9, 14–19; Case 12, 20–25, and Case 13, 26–31) are shown. Trigeminal nerve roots (the emitting portion of the trigeminal nerve from the brainstem) (1, 7, 14, 20, 26), intra‐orbital ophthalmic and maxillary branches (2–4, 8–10, 15–17, 21–23, and 27–29) and mandibular branches at the level of the foramen ovale are shown in axial (5, 11, 18, 24 and 30) and coronal (6, 12, 19, 25, and 31) views. The ophthalmic and maxillary branches (arrowheads) and mandibular branch (arrows) are identified in all NF155⁺ CIDP cases whereas these branches are hardly visible in the control. The mandibular branches (arrows) at the level of the foramen ovale have higher signal intensity and/or hypertrophy in Cases 8 (11 and 12), 9 (18 and 19), 12 (24 and 25), and 13 (30 and 31) compared with the control (5 and 6) (arrows). The coronal view of 3D SHINKEI also demonstrates hypertrophic mandibular branches in Case 8 (13). Abbreviations: CIDP = chronic inflammatory demyelinating polyneuropathy, NF 155⁺ = anti‐neurofascin 155 antibody positive, Rt = right.

Figure 4

Correlation between blink reflex results and ulnar NCS parameters in IgG4 NF155⁺ CIDP. Correlations between R1, iR2, and cR2 latencies and NCS parameters (DL, CMAP, MCV, and F wave latency) are shown. CIDP, chronic inflammatory demyelinating polyneuropathy; CMAP, compound muscle action potential; DL, distal latency; Lt, left; MCV, motor conduction velocity; NCS, nerve conduction study; NF 155⁺, anti‐neurofascin 155 antibody‐positive; Rt, right.

Figure 5

Correlations between blink reflex parameters and anti‐NF155 antibody levels and between trigeminal nerve hypertrophy and disease duration in IgG4 NF155⁺ CIDP. Correlations of anti‐NF155 antibody levels with R1, iR2, and cR2 latencies on stimulation of either side are shown in A, B, and C, respectively. Anti‐NF155 antibody levels are expressed as MFI ratios. ¹⁴ (D) shows a significant positive correlation between the maximum width of the intra‐orbital trigeminal nerve on each side and the period from onset to MRI. CIDP, chronic inflammatory demyelinating polyneuropathy; HC, healthy controls; Lt, left; MFI, mean fluorescence intensity; NF 155⁺, anti‐neurofascin 155 antibody positive; Rt = right.