Clinical Neuroimaging of Photophobia in Individuals With Chronic Ocular Surface Pain

Abstract

Purpose: To examine neural mechanisms underlying photophobia in individuals with chronic ocular surface pain by using functional magnetic resonance imaging (fMRI).

Design: Cross-sectional case/control analysis.

Methods: A total of 16 individuals from the Miami Veterans Affairs eye clinic underwent comprehensive ocular surface evaluations and were surveyed for ocular surface symptoms. Case patients included patients who reported chronic ocular surface pain symptoms and light sensitivity at least most of the time over 1 week. Controls included persons without chronic ocular surface pain who reported no or minimal light sensitivity. All patients viewed light stimuli during 2 fMRI scans, one before and one after topical anesthetic instillation, and rated their level of pain intensity to the stimulus at the end of each scan. Areas of brain activation in response to light stimuli presentation were correlated with pain responses and examined post- vs pre-anesthesia.

Results: Case patients (n = 8) reported higher pain intensity ratings than controls (n = 8) in response to light stimuli during fMRI. Case patient ratings correlated more with light-evoked activation in pain-related areas within the trigeminal brainstem, primary somatosensory cortex (S1), anterior mid-cingulate cortex (aMCC), and insula than in controls. Topical anesthesia led to varying responses in pain ratings among case patients as well as decreased light-evoked activation in S1 and aMCC.

Conclusions: The trigeminal nociceptive system may contribute to photophobia in individuals with chronic ocular surface pain. We demonstrate modulation of cortical structures in this pathway with topically applied anesthetic to the eyes. Further understanding of modulatory interactions that govern ocular surface pain and photophobia is critical for developing effective, precision-based therapies.

Keywords: Ocular pain; Pain processing; Photophobia; fMRI.

Figure 1.. Light-induced pain intensity ratings relative to rest in cases, pre- and post-anesthesia.

Cases had greater evoked pain to the light stimulus pre-anesthetic compared to post-anesthetic conditions. Cases reported variable effects in response to anesthetic on pain intensity ratings. Overall, 3 cases reported decreased, 3 increased, and 2 no change in evoked pain post- vs. pre-anesthetic.

Figure 2.

(A) Cases had greater light-induced activation in the brainstem compared with controls before anesthetic placement. Spinal trigeminal nucleus (spV) (yellow arrows) shows greater activation in cases than in controls. Brainstem activation findings are overlaid onto the SUIT atlas with corresponding MNI coordinates. Group average activation for cases and controls (red-to-yellow) and group contrast for cases>controls (dark-to-light green) are displayed. No areas were significantly decreased in cases relative to controls. Both activation and contrast maps had an individual voxel threshold of z>2.3 and cluster-threshold of P<0.05. (B) Cases had greater light-induced activation in the whole brain compared with controls before anesthetic placement. Group average activation for cases and controls (red-to-yellow) and group contrast for cases>controls (dark-to-light green) are displayed with MNI atlas underlay. Both activation and contrast maps had an individual voxel threshold of z>2.3 and a cluster-threshold of P<0.05. spV= spinal trigeminal nucleus; S1= primary somatosensory cortex; S2= secondary somatosensory cortex; SMC= supplementary motor cortex; aMCC= anterior mid-cingulate cortex; AI= anterior insula; CuC= cuneal cortex; LG= lingual gyrus.

Figure 3.. Correlations between parameter estimates of fMRI activation and light-induced pain intensity ratings in cases vs. controls before anesthetic placement.

Cases had positive correlations between light-induced pain ratings and activity in cortical areas related to pain processing. The WIKIBrainstem and Harvard-Oxford Subcortical and Cortical atlases were used to create anatomical masks of each region. Functional masks were created from group-level contrast maps to pull parameter estimates using FEAT. spV=spinal trigeminal nucleus; S1=primary somatosensory cortex.

Figure 4.. Light-induced activation in cases before vs. after anesthesia.

(A) Pre-anesthetic activation and (B) areas of significantly decreased activation in response to light following anesthesia at the whole-brain level. Group average activation for cases (red-to-yellow) and contrast post- vs. pre-anesthetic within cases (dark-to-light blue) are displayed with MNI atlas underlay. Both activation and contrast maps had an individual voxel threshold of z>2.3 and a cluster-threshold of P<0.05. S1= primary somatosensory cortex; SMC= supplementary motor cortex; aMCC= anterior mid-cingulate cortex.

Figure 5.. Schematic representation of shared trigeminal-related mechanisms in ocular pain, migraine, and photophobia.

RGC=retinal ganglion cells; OPN=olivary pretectal nucleus; SSN=superior salivatory nucleus; spV=spinal trigeminal nucleus. Adapted figure with permission.