Loss of Corneal Nerves and Corneal Haze in Patients with Fuchs’ Endothelial Corneal Dystrophy with the Transcription Factor 4 Gene Trinucleotide Repeat Expansion

OBJECTIVE: Seventy percent of Fuchs’ endothelial corneal dystrophy (FECD) cases are caused by an intronic trinucleotide repeat expansion in the transcription factor 4 gene (TCF4). The objective of this study was to characterize the corneal subbasal nerve plexus and corneal haze in patients with FECD with (RE+) and without the trinucleotide repeat expansion (RE−) and to assess the correlation of these parameters with disease severity. DESIGN: Cross-sectional, single-center study. PARTICIPANTS: Fifty-two eyes of 29 subjects with a modified Krachmer grade of FECD severity from 1 to 6 were included in the study. Fifteen of the 29 subjects carried an expanded TCF4 allele length of ≥ 40 cytosine-thymine-guanine repeats (RE+). MAIN OUTCOMES MEASURES: In vivo confocal microscopy assessments of corneal nerve fiber length (CNFL), corneal nerve branch density, corneal nerve fiber density (CNFD), and anterior corneal stromal backscatter (haze); Scheimpflug tomography densitometry measurements of haze in anterior, central, and posterior corneal layers. RESULTS: Using confocal microscopy, we detected a negative correlation between FECD severity and both CNFL and CNFD in the eyes of RE+ subjects (Spearman ρ = −0.45, P = 0.029 and ρ = −0.62, P = 0.0015, respectively) but not in the eyes of RE− subjects. Additionally, CNFD negatively correlated with the repeat length of the expanded allele in the RE+ subjects (Spearman ρ = −0.42, P = 0.038). We found a positive correlation between anterior stromal backscatter and severity in both the RE+ and RE− groups (ρ = 0.60, P = 0.0023 and ρ = 0.44, P = 0.024, respectively). The anterior, central, and posterior Scheimpflug densitometry measurements also positively correlated with severity in both the RE+ and RE− groups (P = 5.5 × 10(−5), 2.5 × 10(−4), and 2.9 × 10(−4), respectively, after adjusting for the expansion status in a pooled analysis. However, for patients with severe FECD (Krachmer grades 5 and 6), the posterior densitometry measurements were higher in the RE+ group than in the RE− group (P < 0.05). CONCLUSIONS: Loss of corneal nerves in FECD supports the classification of the TCF4 trinucleotide repeat expansion disorder as a neurodegenerative disease. Haze in the anterior, central, and posterior cornea correlate with severity, irrespective of the genotype. Quantitative assessments of corneal nerves and corneal haze may be useful to gauge and monitor FECD disease severity in RE+ patients.