This study highlights the distribution and evolution of refractive errors in patients with IRDs from an early age. Our results demonstrate a strong association between different IRDs and significant ametropia during the first decade of life, with high refractive errors present in approximately one-third of patients. Moreover, we observed considerable heterogeneity within IRDs, with distinct refractive error patterns correlating with specific diagnoses and genes.

To the best of our knowledge, our cohort represents one of the largest refractive error database in early onset IRDs to date, and the first study to examine RE longitudinal progression in patients with different types of IRDs from a young age over time.

In healthy paediatric populations, refractive error development and changes have been well-documented by Flitcroft and colleagues [14]. Their research suggests that between 3 months and 3.5 years of age, a series of key processes occur, including a shift in mean refraction from +2.00 D to approximately +0.75 D, a reduction in variability, and the emergence of a positively skewed refractive error distribution. These processes, collectively referred to as emmetropization, are most rapid during early childhood, continuing at a slower pace until around 6 years of age. This phenomenon is observed globally, even in populations with high myopia rates among children and adults.

Our findings support that emmetropization does not seem to occur in patients with IRDs from an early age. At the initial visit (mean age of 3.6 years), the mean refractive error was 0.00 D, median of 1.9 D with a high standard deviation of 6.0, accompanied by a notable prevalence of high hyperopia and myopia. After a mean follow-up of 5.6 years (mean age of 9 years), a significant myopic shift was observed, with a mean SE of −1.7 D, a median of −0.4 D and a persistently high SD of 6.1.

Our results align with previous studies, such as Hendriks and associates’ findings on CSNB patients, who demonstrated high myopia, especially in those with mutations in the CACNA1F, NYX, and TRPM1 genes [12].

Similarly, other publications report that CSNB patients had a significant myopic shift over time [7]. Igelman and associates [15] focused on the progressive aspect of myopia among CSNB paediatric patients, analysing samples of 78 paediatric patients with CSNB, out of which 41 with CACNA1F, 22 with NYX and 15 with TRPM1 genes. They concluded that all 3 genes showed myopic progression during follow-up with a mean annual progression for TRPM1 patients of −0.326 D. Yassin later on [13], reported seven patients with NYX variants presenting with highly myopic values.

Poels recently published a study suggesting that the refractive error of children with CSNB changes minimally after the age of 4 years old [16]. In his study, only 12 patients with TRMP1-related CSNB, from whom only 9 had refractive follow up, limiting their conclusions regarding this gene specifically.

This study reinforced the association of TRPM1 with high myopia, showing a more pronounced mean myopic progression of −0.56D per year and median of -0.41 D per year (as observed in our cohort of 21 patients). This myopic increase is seen all along the first decade of life, contrary to Poels observations in 2024 [16]. highlighting the need for regular monitoring and consideration of myopia control treatments. The high prevalence of TRPM1-related CSNB patients in our cohort is explained because of two known founder mutations that account for vast majority of cases in our population: a nonsense mutation c.880 A > T (p.Lys294*) and a large genomic deletion (36,445 bp) encompassing exons 2–7 of TRPM1 [17].

In our cohort, patients with Blue Cone Monochromacy (BCM) also exhibited a strong association with high myopia. Our data, along with Semenov and colleagues’ study, supports the connection between BCM and high myopia [18]. Our cohort of 9 BCM patients had a mean SE of −8.6 D (median −7.9 D), and high myopia persisted over time with stable SE values.

Regarding CRB1 gene, previous publications report significant hypermetropia [19,20,21] in coincidence with our results showing that all CRB1 patients had high hypermetropia at the first visit, with a mean of +5.7 D (median 5.4 D). By the last visit, hypermetropia progressed to a mean of +6.6 D (median 6.9 D), with 87.5% of patients classified as highly hypermetropic. This trend was particularly prominent in patients with Leber Congenital Amaurosis (LCA) phenotype, emphasising the importance of early refractive correction.

Our study is one of the first to quantify the refractive progression in achromatopsia patients, showing a shift from mild hypermetropia toward emmetropization over time [12, 22].

While the findings regarding RPGR variants remain mixed, with some studies indicating a high myopia association, our cohort showed that RPGR patients were relatively emmetropic, supporting Yassin’s findings [12, 22]. Regarding CFAP410 and RPE65 genes, not much is known about their refractive natural history. In our cohort we saw a consistent lowering of their SE with time in both genes, but the number of patients was too small to get a reliable refractive pattern.

In conclusion, this study emphasises the critical need for early significant ametropia detection and tailored interventions to optimise visual outcomes in young patients with IRDs. Given the significant myopia progression in TRMP1-related CSNB, early treatment with of low-dose atropine or peripheral defocus glasses for slowing myopia progression, should be taken into consideration. In addition, in patients with CRB1-related LCA, early refractive monitoring is essential to avoid secondary amblyogenic effects of uncorrected high hypermetropia. These results underline the complexity of refractive outcomes in IRDs and highlight the need for further research to explore the interactions between genetic and environmental factors.