The study population comprised children and adolescents aged between 6 and 17 years who were admitted to the Child Psychiatry Department of Kartal Dr Lütfi Kırdar Hospital with a diagnosis of ADHD. This prospective study received ethical approval from the Ethics Committee of Haydarpaşa Numune Training and Research Hospital (HNEAH-KAEK 2023/KK/167) and was conducted in accordance with the Declaration of Helsinki. The participants had been receiving methylphenidate treatment for at least 1 year; however, it was insufficient, therefore guanfacine treatment was added. Following a pediatric psychiatric evaluation, the children were subsequently evaluated by an ophthalmologist. Guanfacine treatment was initiated following the completion of the ophthalmologic evaluation. The children who continued regular follow-up visits were re-evaluated by the ophthalmologist after an evaluation by a child psychiatrist at the sixth-month follow-up. The procedures were fully explained to the participants, the goal of the study was elucidated, and written informed consent was collected from all subjects and their parents prior to enrollment. Patient admission to the research and follow-up processes proceeded between May 2023 and June 2024.

The same physicians from the outpatient clinic for child and adolescent psychiatry evaluated each patient before and 6 months after the start of ADHD therapy. The assessments were performed following the criteria established in the Current and Lifetime Version of the Schedule for Affective Disorders and Schizophrenia for School-Age Children, Diagnostic and Statistical Manual of Mental Disorders-5 (DSM-5) (K-SADS-PL-DSM-5-T) and were based on the DSM-5 diagnostic criteria [15]. The K-SADS-PL semi-structured interview program was employed to ascertain the clinical diagnosis of ADHD. The T-DSM-IV-S Rating Scale was completed by the parents prior to and 6 months following the commencement of treatment. The T-DSM-IV-S scale helped to assess clinical features and their severity.

Detailed ophthalmic examinations of all participants were recorded. The study included participants who had an intraocular pressure (IOP) of < 21 mm Hg, and a best corrected visual acuity (BCVA) of > 0.6 according to a Snellen chart examination. All participants were orthophoric according to the cover/uncover test, prism, and alternating cover test. Patients with a history of previous eye surgery or trauma, refractive error (RE) outside the + 4/-4D range, psychiatric comorbidity other than ADHD, coexisting glaucoma, corneal, retinal, pathology or systemic diseases such as diabetes mellitus, were excluded from the study.

Each participant underwent slit-lamp examinations, fundus examinations, IOP measurements using non-contact tonometry (Nidek NT-530P Noncontact tono/pachymeter, Hiroishi-Cho, Japan), corneal topography (Oculus Pentacam HR, Germany) for CCT, ACD and corneal aberration (total root mean square (RMS), RMS low-order aberration (LOA), and RMS high-order aberrations (HOA) (including trefoil, coma, and spherical aberration)) measurements, enhanced depth imaging (EDI)-OCT (Spectralis, Heidelberg Engineering, Heidelberg, Germany) for central macular thickness (CMT), retinal nerve fiber layer (RNFL), and choroidal thickness (CT) measurements. The REs of the participants were noted according to the autorefractometry (Topcon, Auto Kerato-Refractometer KR.8100 A, Tokyo, Japan) measurements performed 45 min after two drops of cyclopentolate 1% (Sikloplejin %1, Abdi İbrahim, Turkey) in 5 min intervals. The spherical equivalent refraction measurements of the eyes were recorded. The power of the sphere plus (cylinder power/2) was used to calculate the spherical equivalent. These procedures were performed twice before and 6 months after the onset of guanfacine treatment. Data from the right eyes of all participants were used.

To minimize the potential impact of diurnal fluctuations in CCT, corneal topography images of the subjects were acquired at 11:00 a.m. EDI-OCT scans were performed in a dark room from 10:00 to 11:00 a.m. The RNFL scanning mode automatically analyzed the thickness parameters of the peripapillary RNFL. The CMT was automatically measured and assessed using the OCT software. The CT was assessed by measuring the external surface of the hyper-reflective line, known as the “retinal pigment epithelium” layer, and calculating the distance from this outer surface to the hyper-reflective line of the inner scleral border via EDI-OCT. The caliper contained within the software of the device was used to measure the thickness of the choroid layer at the fovea, 500 μm nasal, 1500 μm nasal, 500 μm temporal, and 1500 μm temporal to the fovea. The CVI was assessed using the ImageJ software (Version 1.50a; National Institutes of Health, Bethesda, MD, USA) (Fig. 1). CVI measurements were calculated by two different experienced ophthalmologists who were blinded to the study and averaged measurements were used.

CVI calculation with EDI-OCT image binarization. A. EDI-OCT image of a patient. B. The image was binarized using the auto-local threshold from Niblack. C. The color threshold tool was used to select the dark pixels, representing the luminal area (yellow lines)

The statistical analysis of the study was conducted using the IBM SPSS version 26.0 software package. The demographic information of the subjects and all ocular characteristics were examined using descriptive statistics (frequency, percentage, median, mean, standard deviation, minimum-maximum values). The paired samples t-test was chosen for normally distributed variables, whereas the Wilcoxon signed-rank test was used for non-normally distributed variables in the comparison of baseline and 6-month data within the ADHD group. The Shapiro-Wilk test was employed to confirm the normality of distribution of the variables. All statistical analyses were conducted at a 95% confidence interval with a significance threshold of p < 0.05.