Sjögren Syndrome Eyes Have Decreased Corneal Nerve Densities, Inflammatory Infiltrates

Mature doctor examining patient doing eyesight measurement in ophthalmology clinic
Researchers demonstrate the applicability of in vivo confocal microscopy in monitoring corneal health for patients with Sjögren syndrome.

In vivo confocal microscopy (IVCM) can detect corneal nerve changes in Sjögren syndrome (SS) and may be useful in diagnosis and monitoring of the disease’s progression, according to a study published in Ocular Surface.

Patients (n=71, mean age 50.9±12.0 years) with SS were enrolled at the CHNO des Quinze-Vingts center in France for this retrospective case-control study. The patients were matched with a group of 20 healthy controls and 20 patients with meibomian gland dysfunction (MGD). Nearly a third of the SS cohort (27%) had known small fiber neuropathy (SFN). All study participants were evaluated with the Ocular Surface Disease Index (OSDI) questionnaire and graded using the Oxford scheme. They also underwent tear film break-up time (TBUT), Schirmer testing, Cochet-Bonnet aesthesiometer, and IVCM assessments.

Compared with the controls, the SS group had higher OSDI (mean, 64.2 vs 2.4; P <.0001) and Oxford scores (mean, 1.8 vs 0.2; P <.0001), and lower TBUT (mean, 3.8 vs 8.4 s; P <.0001), Schirmer test (mean, 5.8 vs 24.1 mm; P <.0001), and esthesiometry (mean, 5.5 vs 6.0; P =.026). The SS and MGD groups had similar OSDI scores but differed significantly for the other 4 measures (all P ≤.037).

During IVCM, the SS group had a greater density of inflammatory cells (mean, 86.2 vs 36.5 cells/mm2; P <.001), tortuosity (mean, 2.3 vs 1.8 grade; P <.001), and neuromas (mean, 5.0 vs 1.5; P =.001) and lower nerve density (mean, 16.7 vs 20.5 mm/mm2; P <.005) and number of nerves (mean, 87.3 vs 99.3 nerves/mm2; P =.03) compared with controls. Compared with MGD, SS had lower nerve densities (P =.042) and greater tortuosity (P =.025).

Significant correlations were observed in the SS cohort between Oxford scores and the number of neuromas (r, -0.309; P =.009), Schirmer score (r, -0.303; P =.012), inflammatory cells (r, 0.298; P =.012), and TBUT (r, -0.236; P =.048). TBUT was correlated with nerve density (r, 0.311; P =.008) and inflammatory cells (r, -0.259; P =.029). Reflectivity correlated with nerve density (r, 0.539; P <.001), tortuosity (r, 0.518; P <.001), and inflammatory cells (r, -0.325; P =.006). Tortuosity correlated with nerve density (0.467; P <.001) and inflammatory cells (r, -0.280; P =.018).

Stratified by SFN status, patients with SFN had more neuromas (mean, 7.9 vs 4.0; P =.008).

This study may have been limited by not considering treatment with cyclosporine.

“IVCM has enabled us to demonstrate changes in corneal nerve density and morphology in SS. Our study, carried out on a large cohort of SS patients, confirms the decrease in density of the sub-basal plexus nerves in SS, the existence of inflammatory infiltrates and characteristic neuromas, allowing us to better understand the pathophysiology of this complex disease and to consider new therapeutic approaches aimed at neuroprotection,” study authors explain.


Luzu J, Antoine L, Annabelle R-LG, et al. In vivo confocal microscopic study of corneal innervation in Sjögren’s Syndrome with or without small fiber neuropathy. Ocul Surf. 2022;25:155-162. doi:10.1016/j.jtos.2022.07.003