To explore the damage of panretinal photocoagulation (PRP) to the subbasal nerve plexus (SNP) and its related mechanisms by comparing SNP changes in wide-field mosaic between before and after PRP treatment in diabetic patients.

A randomized controlled study was conducted.Fifty-seven patients (114 eyes) with type 2 diabetes mellitus and binocular diabetic retinopathy (DR) stage IV to receive PRP treatment in Shanxi Eye Hospital from April to November 2019 were enrolled.The subjects were randomly divided into horizontal-vertical laser group and vertical-horizontal laser group according to a random number table.Twenty-nine eyes from 29 patients were assigned to the horizontal-vertical laser group with the photocoagulation sequence of temporal-nasal-inferior-superior.Twenty-eight eyes from 28 patients were assigned to the vertical-horizontal laser group with the photocoagulation sequence of inferior-superior-temporal-nasal.The severer eyes of each subject were chosen as the treatment eye and the contralateral eyes were chosen as the control eye.Corneal confocal laser scanning microscopy (CCM) was performed before PRP treatment, 1 week after each photocoagulation, and 1 month after the completion of PRP treatment to collect images of the SNP over an area of 2-3 mm around the whorl-like pattern.Captured images at each time were merged into one image by using the Photoshop CC 2017 image processing software, and then the nerve fiber length (NFL) of whorl-like pattern was measured by Neuron J image analysis software.McGill pain questionnaire was used to investigate the pain of patients after each photocoagulation.The NFL changes of SNP at different time points were compared between different eyes and different photocoagulation sequence groups.The study protocol adhered to the Declaration of Helsinki and was approved by an Ethics Committee of Shanxi Eye Hospital (No.201804b). Written informed consent was obtained from each patient prior to entering the study cohort.

After PRP treatment, there were different degrees of neural structure loss of SNP nerve fibers in 11 treatment eyes, but there was no significant change in SNP nerve fibers in the control eyes.There were significant differences in NFL between the treatment eyes and the control eyes at various time points (F_eyes=2.020, P=0.039; F_time=4.062, P=0.001). In the horizontal-vertical laser group, different degrees of neural structure loss on the photocoagulation side were found in SNP nerve fibers after the first and second photocoagulation.In the vertical-horizontal laser group, different degrees of neural structure loss on the photocoagulation side were found in SNP nerve fibers after the third and fourth photocoagulation.There was no significant difference in NFL of treatment eyes between the two groups (F_group=0.099, P=0.754), but there was a significant difference in NFL at various time points before and after treatment (F_time=5.231, P<0.001). There were 9 (9/57) patients who complained of pain after PRP, which occurred at the first time of photocoagulation in 7 of them.

SNP damage may occur after PRP in patients with DR, and SNP is prone to be damaged on the photocoagulation side when performing horizontal photocoagulation.

Diabetic retinopathy; Cornea; Nerve damage; Panretinal photocoagulation; Corneal subbasal nerve plexus; Whorl-like pattern; Wide-field mosaic