Diffusion tensor magnetic resonance imaging of trigeminal nerves in relapsing herpetic keratouveitis

Abstract

Background: Corneal hypoesthesia is the landmark of HSV and VZV keratitis and can lead to neurotrophic keratitis. Diffusion tensor imaging (DTI) is a new magnetic resonance imaging (MRI) derived technique, which offers possibilities to study axonal architecture. We aimed at assessing the potential impact of recurrent HSV or VZV-related keratitis on the axonal architecture of trigeminal nerves using DTI.

Design: Prospective non-interventional study.

Participants: Twelve patients and 24 controls.

Methods: DTI using MRI of the trigeminal fibers and corneal esthesiometry using the Cochet-Bonnet esthesiometer were acquired for patients affected by unilateral and recurrent HSV or VZV-related keratitis (3 months after the last corneal inflammatory event), and control subjects with no history of ocular or neuronal disease affecting the trigeminal pathways.

Main outcome measures: Fractional anisotropy (FA) and apparent diffusion coefficient (ADC) were compared between the 2 eyes of both patients and controls, and correlated with corneal esthesiometry.

Results: FA was lower in the trigeminal fibers ipsilateral to the affected eye compared to the non-affected side (0.39±0.02 versus 0.46±0.04, P=0.03). This difference was more important than the intra-individual variability observed in controls. Concomitantly, the asymmetry in ADC results was significantly correlated with the loss of corneal sensitivity in the affected eye.

Conclusions: Corneal hypoesthesia related to HSV and VZV keratitis is associated with persistent modifications in the architecture and functionality of the trigeminal fibers. These results add further explanation to the pathogenesis of HSV and VZV-induced neurotrophic keratitis, which may occur despite an apparent quiescence of the disease.

Fig 1. Tractography and delineation of regions of interest (white circles) of trigeminal nerves in a control (control n° 1, image A) and a patient with a left herpetic stromal keratitis (patient n°2, image B).

These images illustrate the cisternal portion of the trigeminal nerves. The fibers are colour-coded for side: red fibers are on the left, green fibers are on the right). In the control, trigeminal nerves are symmetric. Note the reduction of the number of fibers in affected side (left) of the patient (arrow).

Fig 2. Comparison between the fractional anisotropy (FA) results in the affected eyes and the in non-affected eyes of the patients with a proven or strongly presumed keratitis or keratouveitis related to either HSV or VZV.

The bars indicate the standard error of the mean. Asterisk shows that the difference between values was significant (Kruskal-Wallis test, P = 0.03).

Fig 3. Comparison between the asymmetry of fractional anisotropy (FA) results in patients with a proven or strongly presumed keratitis or keratouveitis related to either HSV or VZV (non-affected eye minus affected eye) and in the control group (higher value minus lower value).

The bars indicate the standard error of the mean. Asterisk shows that the difference between values was significant (Kruskal-Wallis test, P = 0.02).

Fig 4. Asymmetry in the corneal sensitivity (affected eye minus non-affected eye) as a function of the asymmetry in apparent diffusion coefficient (ADC) results (higher value minus lower value) in patients with a proven or strongly presumed keratitis or keratouveitis related to either HSV or VZV.

The solid line indicates the significant correlation between results (Spearman test, P = 0.05).