Abstract
Purpose :
Choroidal abnormalities (CAs) have recently been introduced as new clinical criteria for the diagnosis of neurofibromatosis type 1 (NF1).The purpose of this study was to assess the long term natural history of CAs in a large pediatric NF1 population, quantifying their progression in number and dimensions.
Methods :
Pediatric patients (<16 years old) affected by NF1 with a minimum follow-up of 3 years with at least one CA in one eye were consecutively recruited. Near-infrared (NIR) imaging was performed to identify CAs, which were quantified in number and size using the open-source ImageJ software (National Institutes of Health, Bethesda, Maryland, USA)(Fig 1). CAs area and perimeter were normalized for the optic disc dimensions, to avoid possible bias related to the growing process of the eye.
Results :
This study enrolled 99 eyes of 53 patients (31 males and 22 females), with a mean age of 7.1 ± 3.8 years old. All 99 eyes were examined at baseline and at least at 3-years follow-up (T3); 1-year follow-up (T1) was available for 80 eyes, 5-years follow-up (T5) for 59 eyes and 7-years follow-up (T7) for 36 eyes. Mean number of CAs increased from baseline (3.6 ± 3.2) to different follow-up visits (4.6 ± 3.5 at T1, 6.4 ± 4.1 at T3, 8.1 ± 4.8 at T5 and 9.6 ± 5.3 at T7), with a statistically significant trend (p<0.0001), and an estimated growth rate of 0.82 CAs per year, adjusted for age and for CAs number at baseline (p<0.0001). CAs area and perimeter also significantly increased during follow-up (p<0.0001 for each parameter). Patient age at baseline was inversely correlated with CAs number along time (coefficient=-0.1313, p=0.0068), while no correlation was found between patient age and CAs progression in size.
Conclusions :
In NF1 pediatric patients CAs increased both in number and dimensions, independently from the physiological growth of the eye. While the increase of CAs number occurs particularly at an earlier age, the increase in CAs dimensions is a slow process that remains constant during childhood.
This abstract was presented at the 2022 ARVO Annual Meeting, held in Denver, CO, May 1-4, 2022, and virtually.