Ocular Surface Pathology in Patients Suffering from Mercury Intoxication

Abstract

Purpose: To report the ocular surface pathology of patients suffering from acute/subacute mercury vapor intoxication.

Design: Cross-sectional study.

Participants: Male workers intoxicated with inorganic mercury referred for ophthalmic involvement and healthy control subjects.

Methods: The following tests were performed: dry eye (DE)-related symptoms indicated by the ocular surface disease (OSDI) index questionnaire; tear osmolarity; analysis of 23 tear cytokine concentrations and principal component and hierarchical agglomerative cluster analyses; tear break-up time (T-BUT); corneal fluorescein and conjunctival lissamine green staining; tear production by Schirmer and tear lysozyme tests; mechanical and thermal corneal sensitivity (non-contact esthesiometry); and corneal nerve analysis and dendritic cell density by in vivo confocal microscopy (IVCM).

Results: Twenty-two out of 29 evaluated patients entered the study. Most had DE-related symptoms (OSDI values > 12), that were severe in 63.6% of them. Tear osmolarity was elevated (>308 mOsms/L) in 83.4% of patients (mean 336.23 (28.71) mOsm/L). Corneal and conjunctival staining were unremarkable. T-BUT was low (<7 s) in 22.7% of patients. Schirmer test and tear lysozyme concentration were low in 13.6% and 27.3% of cases, respectively. Corneal esthesiometry showed patient mechanical (mean 147.81 (53.36) mL/min) and thermal thresholds to heat (+2.35 (+1.10) °C) and cold (-2.57 (-1.24) °C) to be significantly higher than controls. Corneal IVCM revealed lower values for nerve density (6.4 (2.94) n/mm²), nerve branching density (2 (2.50) n/mm²), and dendritic cell density (9.1 (8.84) n/mm²) in patients. Tear levels of IL-12p70, IL-6, RANTES, and VEGF were increased, whereas EGF and IP-10/CXCL10 were decreased compared to controls. Based on cytokine levels, two clusters of patients were identified. Compared to Cluster 1, Cluster 2 patients had significantly increased tear levels of 18 cytokines, decreased tear lysozyme, lower nerve branching density, fewer dendritic cells, and higher urine mercury levels.

Conclusions: Patients suffering from systemic mercury intoxication showed symptoms and signs of ocular surface pathology, mainly by targeting the trigeminal nerve, as shown by alterations in corneal sensitivity and sub-basal nerve morphology.

Keywords: cornea toxic effects; corneal esthesiometry; corneal innervation; corneal nerves; in vivo confocal microscopy; mercury poisoning; neurogenic dry eye; occupational exposure; tear biomarkers; tear cytokines.

Figure 1

In vivo confocal microscopy images of the central cornea in a healthy control and in a mercury-intoxicated patient. For both corneas, the images were taken at a depth of 50–80 μM. (Left) In the healthy control subject, the following measurements were made: nerve density, 15/mm²; tortuosity, 2; mean nerve length, 27.94 mm/mm²; nerve branch density, 7/mm²; dendritic cell density, 4/mm²; and reflectivity, 109.1 gray units. (Right) In the mercury-intoxicated subject, the following measurements were made: nerve density, 8/mm²; tortuosity, 3; mean nerve length, 13.77 mm/mm²; nerve branch density, 3/mm²; dendritic cell density, 0.5/mm²; and reflectivity, 76.29 gray units.

Figure 2

Fold change of tear cytokine levels in mercury-poisoned patients versus healthy controls. Tear cytokines are shown on the Y-axis. The fold change (FC) with the 95% confidence interval (CI, horizontal lines) for each cytokine is shown on the X-axis. The case: control FC was defined as the relative expression of the cytokine concentration in the patient group divided by the control group. Data on the X-axis are presented in a base 2 logarithmic (log2) scale. The vertical dashed line represents no change. The FCs were significant if the 95% CI did not cross the vertical dashed line. The farther the distance to the vertical dashed line, the greater the statistical significance. Positive values mean over-expression and negative values mean under-expression.