Assessing fixation stability in prematurity

Abstract

Purpose : To assess fixation stability (FS) in individuals born before and after 37 weeks of gestation (WG) using two different devices: the Macular Integrity Assessment system (MAIA) and an adaptive optics scanning light ophthalmoscope (AOSLO).

Methods : Fourteen adult individuals born after 37 WG (4M, 10F; Age range: 21 to 38 yrs) and seven premature individuals born before 37 WG without any neurodegenerative or ocular pathology (2M, 5F; Age range: 12 to 30 yrs) underwent examination in one eye each using the MAIA and AOSLO. The MAIA used a 1 deg diameter red circle fixation target and a tracking frequency of 25 Hz. The AOSLO used a 0.153 deg x 0.153 deg black square fixation target and a tracking frequency of 16 Hz. Coordinates corresponding to the retinal loci subjects used to fixate on the target were used to calculate BCEA68 (bivariate contour ellipse area encompassing 68% of the highest density points) and ISOA68 (isoline contour area which utilizes kernel density estimation to estimate the probability density function of eye positions and encompasses the area that encloses 68% of the highest density estimate). Assessments were made on the same day and agreement between the two devices was also analyzed using a Bland-Altman plot.

Results : The mean (± stdev) BCEA68 was 0.0491 ± 0.0336 deg² for MAIA and 0.0264 ± 0.0123 deg² for AOSLO. From the Bland-Altman plot analysis, the logBCEA68 mean difference was 0.2321 (95% CI = 0.0919 to 0.3723). No significant difference in BCEA68 was observed between individuals born before or after 37 WG with either the MAIA (Mann-Whitney U test: p=0.6359) or the AOSLO (Welch’s unpaired t-test: p=0.72). The mean (± stdev) ISOA68 was 0.0434 ± 0.0301 deg² for the MAIA and 0.0177 ± 0.0072 deg² for AOSLO. The logISOA68 Bland-Altman analysis showed a mean difference of 0.3318 (95% CI = 0.2132 to 0.4505). Similar to the BCEA68 findings, there was no difference in FS between individuals born before or after 37 WG for the ISOA68 data for either the MAIA (Mann-Whitney U test: p=0.25) or for the AOSLO (Welch’s unpaired t-test: p=0.97).

Conclusions : Individuals with a history of premature birth had similar FS compared to those born full-term, though larger samples are needed to reconcile our observations with those from prior studies. Measurements of FS from the MAIA and AOSLO do not appear to be interchangeable, highlighting the importance of factoring in the method used when comparing results.

This abstract was presented at the 2023 ARVO Annual Meeting, held in New Orleans, LA, April 23-27, 2023.