A randomised AI-supported long-term evaluation of a layer-by-layer macular thickness profile after manual and femtosecond laser cataract surgery

Abstract

Objectives: To evaluate retinal thickening 18-months after low-pulse energy femtosecond laser-assisted (LCS) and manual cataract surgery (MCS) using artificial intelligence (AI)-based automated retinal layer segmentation. Methods: Design: Prospective, randomised, intraindividual-controlled study Setting: Department of Ophthalmology, Medical University of Vienna. Patient Population: 120 eyes of sixty patients with age-related cataract Intervention: Bilateral same-day LCS and MCS in a randomised sequence. Main Outcome Measure: Retinal thickness preoperative, after 1 week, 3 weeks, 6 weeks and 18 months in the central 1mm, 3mm, 6mm. Inner nuclear layer (INL), outer plexiform layer (OPL), outer nuclear layer (ONL), total retinal thickness (TRT) and photoreceptor (PR) thickness were segmented from Spectralis optical coherence tomography (OCT) using automated AI retinal layer segmentation. Results: Fifty-six patients completed the follow-up. LCS compared to MCS did not impact any of the investigated retinal layers at any follow-up visit (p>0.05). For the central 1mm, a significant increase in TRT was seen after 1 week followed by an elevated plateau thereafter. For the 3mm and 6mm, TRT increased only after 3 weeks and 6 weeks and decreased again until 18-months. TRT remained significantly increased compared to preoperative thickness (p<0.001). INL and ONL were the main causative layers for the total TRT increase. PR declined 1 week after surgery but regained preoperative values 18 months after surgery. Conclusions: Low-energy femtosecond laser pre-treatment did not influence thickness of the retinal layers in any topographic area compared to manual high fluidic phacoemulsification. TRT did not return to preoperative values 18 months after surgery.