Longitudinal Evaluation of Chromatic Pupillometry and Retinal Thickness in Dogs With Sudden Acquired Retinal Degeneration Syndrome Authors: Soohyun Kim, Ann E Cooper, David J Maggs, Ariana D Marangakis, Tran Nguyen, Monica Motta, Sara M Thomasy Published: 2026 ABSTRACT ABSTRACTObjectiveTo determine longitudinal changes in chromatic pupillary light reflexes (cPLR) and retinal layers in dogs with sudden acquired retinal degeneration syndrome (SARDS).Animals StudiedClinical evaluation: Fifteen SARDS‐affected and nine control dogs. Histology: Achieved globes from five SARDS‐affected (1–7 years post‐diagnosis) and eight age‐matched controls.ProceduresDogs underwent ophthalmic examination and electroretinography to confirm SARDS. Chromatic pupillometry and Fourier‐domain optical coherence tomography (FD‐OCT) were performed at baseline (once for control; < 1‐month post‐blindness for SARDS‐affected dogs), and follow‐up (3, 6 and 12 months post‐diagnosis). The extent, velocity and latency of pupil constriction were evaluated. Retinal thickness was measured from the FD‐OCT images. Ganglion cell counts were determined from histologic sections stained with hematoxylin and eosin.ResultsAt baseline, SARDS‐affected dogs showed significantly reduced extent and velocity of red light PLR, and reduced extent with prolonged latency of white light PLR, relative to controls; white light velocity and all blue light parameters did not differ. White light PLR velocity significantly decreased over time in SARDS‐affected dogs. FD‐OCT in SARDS‐affected dogs revealed progressive total retinal thinning but stable inner retinal thickness. Histology of SARDS‐affected eyes identified diffuse outer retinal atrophy with relative inner retinal preservation; ganglion cell counts were not different between groups.ConclusionsChromatic pupillometry reflects the selective preservation of melanopsin‐containing retinal ganglion cell function despite progressive photoreceptor degeneration in SARDS. The combined longitudinal use of chromatic pupillometry and FD‐OCT provides a comprehensive assessment of the evolving functional and structural changes in SARDS‐affected dogs. Introduction Sudden acquired retinal degeneration syndrome (SARDS) is one of the leading causes of irreversible vision loss in dogs, characterized by acute‐onset bilateral blindness with initially normal ophthalmic examination findings [1, 2]. Despite proposed hormonal [3], autoimmune [4, 5], and heritable [2, 6] etiologies, the pathogenesis of SARDS remains unclear, hindering the development of effective treatments. Because of the numerous different breeds affected, a simple monogenic genetic trait is considered unlikely [7]. Though no proven treatment exists for SARDS, early diagnosis and differentiation from other potentially treatable causes of acute blindness, such as meningitis, encephalitis, optic neuritis, and retinal detachment is crucial. Currently, electroretinography (ERG) is the diagnostic standard, but requires specialized equipment and expertise, limiting its accessibility. The pupillary light reflex (PLR), a rapid noninvasive evaluation of retinal and optic nerve function [8], offers a potential adjunctive assessment of retinal function. However, traditional PLR using white‐light cannot distinguish retinal from optic nerve or central pathway dysfunction. Prior research has identified distinct rod–cone and intrinsically photosensitive retinal ganglion cell (ipRGC)‐mediated PLR pathways [9, 10]. Melanopsin‐expressing ipRGCs, projecting to midbrain structures, mediate PLR and regulate circadian rhythms, even in the presence of severe photoreceptor damage [9, 11]. Chromatic pupillometry, using blue (480 nm, targeting melanopsin‐containing ipRGCs) and red (630 nm, targeting cones) light, allows for selective assessment of these pathways [12, 13, 14]. The distinct absorption spectra of melanopsin‐containing ipRGCs and cone photoreceptors enable separate measurement of their responses [15]. Chromatic pupillometry has been investigated in various canine retinal diseases, including SARDS [16, 17], progressive retinal atrophy [17, 18], retinal degeneration [4, 18], retinal detachment [4], and optic neuritis [17]. While Grozdanic et al. [8] reported that PLRs in SARDS‐affected dogs were normal in response to blue light but absent in response to red light, Terakado et al. [17] found variable results, including decreased PLRs in response to red or blue light in some cases. This variability is unsurprising given that the aforementioned studies were cross‐sectional, assessing subjects at a single time point without considering disease stage. Despite the apparently abrupt onset and non‐progressive clinical course, anatomic evaluation of SARDS‐affected eyes suggests that the structural changes are complex, progressive, and dependent on disease stage. Histopathologic evaluation reveals photoreceptor outer segment loss within 3 weeks of disease onset followed by outer nuclear layer thinning [19, 20, 21]. However, in early stages of the disease (i.e., within 6 weeks of vision loss), some dogs exhibit significantly thickened inner retina and photoreceptor layer measurements compared to controls, suggesting that retinal thickening might be one of the earliest detectable changes [22, 23]. Additionally, increased vitreous haze has been observed in SARDS patients using OCT, potentially indicating an underlying inflammatory component [23]. As the disease progresses, however, this initial response gives way to rapid retinal degeneration. Within weeks, there is a complete loss of definition of the photoreceptor bands visible on OCT, with them appearing instead as a single homogenous layer [24]. This is accompanied by a significant thinning of the outer nuclear layer and the entire outer retina across all quadrants [24]. Advanced cases often show multifocal attenuation of the retinal vasculature, and in some instances, small multifocal retinal detachments associated with fluorescein leakage at the level of the retinal pigment epithelium [24]. Ultrastructural analysis has also revealed that SARDS‐affected eyes have a significantly thicker Bruch's membrane compared to eyes of control dogs and dogs with progressive retinal atrophy [7]. Over months to years, the inner retina also degenerates, leading to the loss of bipolar, amacrine, and ganglion cells [19, 21]. Given the marked evidence of progressive retinal pathology, we hypothesized that the cPLR may vary notably and in a clinically important way depending on the SARDS disease stage. If so, chromatic pupillometry may be useful in tracking progression, estimating disease stage, and monitoring the efficacy of future therapies. Therefore, this study was designed to longitudinally monitor cPLRs in SARDS‐affected dogs. We correlated pupillary responses with clinical findings and in vivo histological correlates using Fourier‐domain optical coherence tomography (FD‐OCT) across different disease stages. Methods Animals and Inclusion/Exclusion Criteria This study was approved by the Institutional Animal Care and Use Committee (IACUC) of the University of California, Davis (#19355) and conducted in accordance with the Association for Research in Vision and Ophthalmology (ARVO) Statement for the Use of Animals in Ophthalmic and Vision Research. Owners gave informed consent for enrollment and completed a standardized questionnaire to provide detailed patient history. All dogs underwent a complete ocular examination performed by a board‐certified veterinary ophthalmologist or an ophthalmology resident under direct supervision in an approved training program. Dogs were assigned to one of two groups: SARDS‐affected or normal (“control”). Inclusion criteria for the SARDS group were as follows: (1) acute onset of bilateral complete blindness within a month prior to the initial evaluation; (2) ophthalmic examination findings insufficient to explain the blindness; and (3) a completely extinguished ERG. Inclusion criteria for the control group were as follows: (1) owner‐reported normal vision; (2) normal ophthalmic examination findings; (3) normal ERG responses; and (4) absence of neurologic dysfunction. Dogs were excluded from both groups if they exhibited any of the following: abnormalities suggestive of diseases likely to influence pupillary constriction (e.g., marked iris atrophy, uveitis, lens instability); or intraocular pressure (IOP) > 20 mmHg in either eye, as measured by rebound tonometry. SARDS‐affected dogs were evaluated at up to four time points: within 1 month of owner‐reported blindness (baseline; n = 15), and at approximately 3 (n = 15), 6 (n = 14), and 12 months (n = 12) following vision loss. Control dogs (n = 9) were evaluated at a single time point. For each dog, at each evaluation, bilateral chromatic pupillometry, maze testing, ophthalmic examination, and FD‐OCT were conducted in the sequence outlined below. ERG was performed only once at the initial visit. Video Chromatic Pupillometry Chromatic PLRs were evaluated using the Melan‐100 (BioMed Vision Technologies, Ames, IA) following the manufacturer's instructions. Following confinement for at least 15 min in a quiet, dark room, dogs were gently restrained under dim red‐light conditions and cPLRs were elicited with red (630 nm), blue (480 nm), and white light‐emitting diode (LED) light sources, in that order, each set to an illumination intensity of 200 kcd/m2 and held approximately 1 in. from the corneal surface for 10 s. The right eye was assessed first; the left eye was then assessed following 30 s of dark adaptation. Thus, at least 60 s elapsed between direct stimulation of each eye with each colored light stimulus. For each light color, the direct PLR was recorded once in each eye using a video recording device consisting of a 60‐mm Nikkor lens (F‐2.8; Nikon Inc., Melville, NY) fitted with a coaxial infrared videoretinoscopic attachment (Raspberry pi, Pencoed, Wales, UK) attached to a video monitor. In order to provide illumination for video capture of the resting pupil size without inadvertent retinal stimulation, infrared light‐emitting diodes were mounted on either side of a 3D printed sleeve through which the Melan‐100 device was inserted. For each light source, still frames were extracted from the pupillometry video at baseline (before illumination) and when minimum pupil diameter was achieved following stimulation. For each image, total pupil area was measured in mm2 using image‐processing software (ImageJ version 1.421; NIH free‐domain). Subtle discrepancies in the distance between the camera and the animal's eye which could influence measurements were normalized by expressing the pupil area as a proportion of the limbus‐to‐limbus corneal diameter measured horizontally from the same still frames. The following parameters of pupillary constriction were recorded: (1) extent of pupil constriction calculated as the change in pupil area from baseline (i.e., (baseline pupil area − minimum pupil area following 10 s of light stimulation)/(baseline pupil area) × 100), and reported as a percentage; (2) latency of pupil constriction calculated as time from stimulus onset until start of pupil constriction, and reported in s; (3) time to maximum pupil constriction calculated as the time from beginning of pupil constriction to minimum pupil size, and reported in s; and (4) velocity of pupil constriction calculated as extent of pupil constriction (i.e., (baseline pupil diameter (mm) − minimum pupil diameter (mm))/time to maximal pupil constriction (s)), and reported in mm/s. Maze Testing and Ophthalmic Examination A maze with plastic obstacles was used to evaluate the dogs' ability to navigate under different light intensities, without patching their eyes, thus assessing their binocular vision before ophthalmic examination so as to avoid the effects of pupillary dilation or recent bright illumination. For a standardized maze setup, six orange caution cones, each with a 12″ base width and 36″ in height, served as obstacles. These were placed within a fenced area measuring 105 cm in width and 525 cm in length. Though the specific obstacle placement was varied for each trial to prevent the dogs from learning their locations, the same general layout was maintained for each individual patient. Each dog was tested in mesopic conditions at 2 lx (indoors with doors closed and red overhead lighting), and then in photopic conditions at 10 lx (indoors with doors closed and white overhead lighting). Ambient luminance was assessed using a photometer (Sekonic Studio Deluxe II model L‐398M photometer; Sekonic Co. Ltd., Tokyo, Japan), and each dog was allowed to adapt to the dim lighting for 10 min prior to testing. Following dark adaptation, an assistant released the dog, who was then allowed to navigate freely through the maze while a second assistant stood at the opposite end. If the dog was not moving, the second assistant repeatedly called its name to encourage movement. For all trials, total time to complete the maze as well as number of times the dog contacted an obstacle within the maze were recorded by an observer. Dogs completed the maze three times for each lighting condition, and results were averaged. After maze testing, eyes underwent complete ophthalmic examination including assessment of menace response, dazzle and pupillary light reflexes (white light), rebound tonometry (TonoVet, iCare, Finland), slit‐lamp biomicroscopy (SL‐15 or SL‐17, Kowa Co. Ltd., Tokyo, Japan), and binocular indirect ophthalmoscopy (Keeler Instruments, PA, USA) with a pan‐retinal 2.2 lens (Volk, OH, USA). Pharmacologic mydriasis was achieved with topical 1% tropicamide (Tropicamide; Falcon Pharmaceuticals, Fort Worth, TX) for the fundic examination, ERG, FD‐OCT, and fundic photography. Electroretinography and Advanced Imaging To minimize confounding effects of previous light exposure, electroretinography was performed prior to fundic photography and following 1 h of dark adaptation after ophthalmic examination. The ERG was conducted without sedation using a handheld multispecies ERG unit (HMsERG Model‐1000; OcuScience, Henderson, NV) that includes a built‐in mini Ganzfield stimulator which was used to deliver both the background illumination and the flash stimuli. After applying one drop of 0.5% proparacaine hydrochloride ophthalmic solution (Alcon Laboratories Inc., Fort Worth, TX), a contact lens electrode (ERG‐jet, Fabrinal SA, Switzerland) was applied to the cornea and electrical conduction enhanced by topical application of hydroxypropyl methylcellulose gel (Gonak; Akorn Inc., Lake Forest, IL). Subdermal reference and ground electrodes (F‐E2; OcuScience, Henderson, NV) were placed 3 cm from the lateral canthus over the zygomatic arch and at the occiput, respectively. Electrode impedance was maintained at < 5 kΩ, and bandpass was 0.3–300 Hz. Outer retinal function was then recorded using the preprogrammed HMsERG “Quick Ret Check Plus” protocol. Rod function was tested using a dim stimulus (average of 10 flashes at 0.5 Hz and 10 mcd*s/m2). Subsequently, the mixed rod–cone response to standard intensity (average of four flashes at 0.1 Hz and 3000 mcd*s/m2) and high‐intensity (average of four flashes at 0.05 Hz and 10 000 mcd*s/m2) stimuli were recorded. Retinas of all eyes were imaged using FD‐OCT (RTVue 100, software version 6.1; Optovue Inc., Fremont, CA). If necessary, dogs were sedated with intravenous injection of butorphanol tartrate (Torbugesic; Fort Dodge Animal Health, Overland Park, KS, 0.2–0.3 mg/kg) then placed in sternal recumbency and the following scans were obtained from both eyes of all dogs: (1) 6 × 4 mm raster scans over the peripapillary region and dorsolateral to the optic disc, (2) 6‐mm horizontal line scans along the dorsal aspect of the optic disc extending temporally through the approximate region of the area centralis (i.e., within the visual streak dorsolaterally from the optic nerve head), and (3) bisecting the optic nerve head. Total retinal thickness (from the nerve fiber layer to the retinal pigment epithelium) and inner retinal thickness (from the inner nuclear layer to the inner limiting membrane) were manually measured at sites 1‐mm temporal to the optic nerve head using the RTVue software (version 6.1) measuring tool. Outer retinal thickness was then calculated at each site by subtracting inner retinal thickness from total retinal thickness. Immediately following FD‐OCT, representative color images of the fundus of both eyes of all dogs were captured using a fundus camera (CF‐1 Digital Retinal Camera; Canon, Lake Success, NY, USA). Histological Evaluation To investigate histologic changes within the retina, we examined eight globes from five dogs previously diagnosed with SARDS by a veterinary ophthalmologist based on clinical findings and ERG and subsequently enucleated for unrelated reasons 1–7 years after diagnosis. Eight age‐matched control globes were obtained from eight dogs enucleated for various reasons unrelated to retinal disease, glaucoma, or other conditions that could affect retinal morphology. All globes were selected from the Comparative Ocular Pathology Laboratory of Wisconsin (COPLOW) database and were not included in the aforementioned in vivo portion of this study. Five‐micrometer thick, hematoxylin‐ and eosin‐stained parasagittal sections of whole globes were digitally imaged and evaluated using pathology slide viewing software (Aperio ImageScope 12.4.6, Leica Biosystems, Deer Park, IL). Images were captured at 40× magnification from regions approximately 1 mm superior (tapetal) and inferior (non‐tapetal) to the optic nerve head. Ganglion cells were quantified by hand‐counting using three images per region (6 images per globe); these counts were performed by S.K., masked to disease status. Statistical Analysis All data are presented as median values along with the interquartile range (IQR). Statistical analysis was performed using GraphPad Prism 9.04 (GraphPad Software Inc., San Diego, CA). Mann–Whitney rank sum tests were used to compare maze test results, baseline cPLR parameters, and total and inner retinal thickness between the SARDS‐affected and control groups. Changes in all measured values over time within the SARDS‐affected group were assessed using the Kruskal–Wallis test followed by Dunn's multiple comparisons test for post hoc analysis. A p‐value of < 0.05 was considered statistically significant for all analyses. Data were analyzed separately for right and left eyes, and with both eyes combined; however, no statistically significant differences were found between right and left eyes. Therefore, data from both eyes of each dog were averaged to generate animal‐level means for all subsequent analyses. This approach accounts for the lack of independence between eyes of the same animal. The results presented herein therefore represent animal‐level data. Results Demographic and Clinical Characteristics A total of 47 dogs were screened for the study (30 suspected SARDS and 17 normal controls). Of those, 24 dogs (15 SARDS‐affected dogs and nine age‐matched control dogs) fulfilled all inclusion criteria and were successfully enrolled. The remaining 23 dogs (15 suspected SARDS and 8 controls) were excluded prior to enrollment due to concurrent ocular disease, owner declining follow‐up, or incomplete diagnostic workup. Ultimately, 15 SARDS‐affected dogs (7 spayed females and 8 castrated males) and nine control dogs (4 spayed females and 5 castrated males) were included in this study. Median (range) duration of vision loss in SARDS‐affected dogs, from onset until the first evaluation, was 22 (10–30) days. Median (IQR) age of SARDS‐affected dogs was 10.0 (6.0–11.0) years, and of control dogs was 10.0 (6.5–11.0) years. Median (IQR) body weight was 10.3 (7.0–19.7) kg in the SARDS group and 12.1 (6.0–18.95) kg in the control group (Table S1). A statistical difference in age (p = 0.8232) or body weight (p = 0.8153) was not detected between the two study populations. Neuro‐ophthalmic examination findings of all control dogs were within normal limits. By contrast, all SARDS‐affected dogs exhibited normal palpebral and dazzle reflexes but absent menace responses and a range of alterations in direct and consensual PLRs with regular intensity white light. At baseline, ophthalmic examination findings (including fundic appearance) were unremarkable in all eyes from both groups. In the SARDS‐affected dogs, progressive degenerative fundic changes were observed at later time points, including varying degrees of generalized tapetal hyperreflectivity and retinal vascular attenuation; otherwise, ophthalmic examination findings remained unremarkable. Median (IQR) IOP was 11.0 (9.0–12.2) mm Hg in control dogs and 11.0 (9.0–12.3) mm Hg in SARDS‐affected dogs (p = 0.7521). Electroretinography revealed absent waveforms in all SARDS‐affected dogs, while control dogs showed normal waveforms. Collectively, these findings confirmed the SARDS diagnosis in all affected dogs. Maze test results corroborated the clinical diagnosis of blindness in the SARDS group. At baseline, the median (IQR) maze transit times in mesopic (p < 0.0001) and photopic (p < 0.0001) conditions were significantly faster in control dogs (6.17 (5.22–8.30) s and 4.50 (3.46–5.83) s, respectively) than in SARDS‐affected dogs (19.51 (13.35–27.76) s and 18.59 (11.85–34.56) s, respectively). While no obstacle collisions were observed in the control group under either lighting condition, SARDS‐affected dogs exhibited significantly more obstacle collisions under both mesopic (4.3 (3.3–5.7) incidents; p < 0.0001) and photopic (5.0 (3.0–6.0) incidents; p < 0.0001) conditions. Relative to baseline, no significant change was detected over time for maze transit times or obstacle collisions in SARDS‐affected dogs performing under mesopic or photopic conditions (data not presented). Chromatic Pupillometry At baseline, median (IQR) extent of pupil constriction was significantly less in SARDS‐affected (n = 15) than in control dogs (n = 9) when they were stimulated with red (17.1 (11.5–20.7)% vs. 66.7 (63.8–70.8)%, respectively; p = 0.0001) or white (50.0 (43.7–57.3)% vs. 63.4 (58.4–70.3)%, respectively; p = 0.0464) light. Conversely, no significant difference in extent of pupil constriction was detected between SARDS‐affected and control dogs when stimulated with blue light (79.0 (75.0–87.7)% vs. 84.9 (82.8–88.1)%, respectively; p = 0.6058) (Table 1). No significant changes in extent of pupil constriction over time were detected in the SARDS‐affected dogs with any light stimulus color (p = 0.2786, 0.6479, and 0.2747 for red, blue, and white lights, respectively) (Figure 1A). TABLE 1Chromatic pupillometry parameters in 15 SARDS‐affected and nine normal control dogs at baseline.Light colorNormal control dogsSARDS‐affected dogsExtent of pupil constriction (%)Red66.717.1*(63.8–70.8)(11.5–20.7)Blue84.979.0(82.8–88.1)(75.0–87.7)White63.450.0*(58.4–70.3)(43.7–57.3)Latency of pupil constriction (s)Red0.520.54(0.45–0.63)(0.53–0.57)Blue0.500.46(0.46–0.55)(0.41–0.55)White0.370.44*(0.33–0.43)(0.41–0.51)Velocity of pupil constriction (mm/s)Red2.060.35*(1.92–2.36)(0.10–0.67)Blue3.654.61(2.55–6.91)(2.03–5.59)White2.842.99(2.58–3.54)(2.29–3.88) Note: Data are analyzed as per‐animal means of both eyes and presented as median (interquartile range). A Mann–Whitney rank sum test revealed a statistically significant difference for the indicated parameters (*p < 0.0001). Chromatic pupillometry parameters in 15 SARDS‐affected and nine normal control dogs at baseline. Note: Data are analyzed as per‐animal means of both eyes and presented as median (interquartile range). A Mann–Whitney rank sum test revealed a statistically significant difference for the indicated parameters (*p < 0.0001). Box and whisker plots showing longitudinal chromatic pupillometry parameters in SARDS‐affected dogs. Data are presented as per‐animal means of both eyes. Control dogs (n = 9) were evaluated once; SARDS‐affected dogs were evaluated at baseline (n = 15), 3 months (n = 15), 6 months (n = 14), and 12 months (n = 12). Red light (pink boxes) targets photoreceptors; blue light (blue boxes) predominantly targets melanopsin‐containing ipRGCs; white light (white boxes) stimulates both pathways. Brackets with p values indicate significant differences (Mann–Whitney U test for Control vs. Baseline comparisons; Kruskal–Wallis test followed by Dunn's multiple comparisons with Bonferroni correction for longitudinal comparisons among SARDS‐affected dogs). (A) Extent of pupil constriction (%) was significantly reduced in SARDS‐affected dogs compared to controls for red (***p = 0.0001) and white (*p = 0.0464) light at baseline; no significant longitudinal changes were detected for any light stimulus. (B) Latency of pupil constriction (s) was significantly longer in SARDS‐affected dogs compared to controls for white light at baseline (*p = 0.0274); no significant longitudinal changes were detected for any light stimulus. (C) Velocity of pupil constriction (mm/s) was significantly reduced in SARDS‐affected dogs compared to controls for red light at baseline (**p = 0.0079). The velocity of pupil constriction in response to white light significantly decreased over time in SARDS‐affected dogs (p = 0.0112), with significant reductions at 6 months (*p = 0.0408) and 12 months (*p = 0.0480) relative to baseline; no significant longitudinal changes were detected for red or blue light stimulation. Latency of pupillary constriction in SARDS‐affected dogs was not statistically different from control dogs at baseline in response to red (p = 0.6058) or blue (p = 0.7430) lights. However, pupil latency in response to white light was significantly longer in SARDS‐affected than in control dogs at baseline (p = 0.0274) (Table 1). In the SARDS‐affected group, no significant longitudinal changes in latency of pupil constriction were detected (p = 0.2815) in response to red light stimulation at 3 months (0.55 (0.44–0.71) s), at 6 months (0.55 (0.47–0.75) s), or at 12 months (0.81 (0.61–1.88) s). Although latency in response to red light increased and became more variable at 12 months, it did not reach statistical significance (Figure 1B). Latency of pupillary constriction in response to white light also showed no significant longitudinal changes (0.49 (0.46–0.66) s at 3 months, 0.60 (0.47–0.78) s at 6 months, 0.62 (0.58–0.72) s at 12 months; p = 0.1708), nor were there significant changes in the response to blue light stimulation over time (0.39 (0.33–0.62) s at 3 months, 0.48 (0.41–0.67) s at 6 months, 0.55 (0.44–0.63) s at 12 months; p = 0.4550). At baseline, velocity of pupillary constriction following red light stimulation was significantly reduced in SARDS‐affected dogs (0.35 (0.10–0.67) mm/s) compared to controls (2.06 (1.92–2.36) mm/s; p = 0.0079) (Table 1); however, no significant changes were observed over time within SARDS‐affected dogs (0.99 (0.45–1.20) mm/s at 3 months, 0.58 (0.29–1.16) mm/s at 6 months, 0.13 (0.05–0.64) mm/s at 12 months; p = 0.7265). When stimulated with blue light, significant differences in pupil constriction velocity were not detected in SARDS‐affected versus control dogs at baseline (p = 0.6058; Table 1) nor over time in SARDS‐affected dogs (3.72 (2.62–5.23) mm/s at 3 months, 2.17 (1.71–2.95) mm/s at 6 months, 2.09 (1.77–3.92) mm/s at 12 months; p = 0.1512; Figure 1C). Velocity of pupil constriction with white light stimulation also showed no significant difference between SARDS‐affected and control dogs at baseline (p = 1.0000; Table 1). However, it significantly decreased over time (p = 0.0112) with lower velocities at 6 and 12 months (1.80 (1.36–2.02) mm/s and 1.19 (1.12–1.47) mm/s, respectively) compared with baseline (p = 0.0408 and 0.0480, respectively), but no differences at 3 months (2.40 (2.15–3.23) mm/s; p = 1.0000) (Figure 1C). Retinal Thickness Median (IQR) total and outer retinal thickness measured 1 mm temporal to the optic nerve head on FD‐OCT did not differ significantly between SARDS‐affected (198.0 (193.6–204.4) μm and 104.6 (95.4–146.8) μm, respectively) and control dogs (202.5 (191.8–215.5) μm and 113.3 (60.0–160.0) μm, respectively) at baseline (p = 0.3044 and 0.5451, respectively; Figure 2). However, total retinal thickness in SARDS‐affected dogs declined significantly and progressively at each follow‐up examination (164.0 (157.0–181.5) μm at 3 months, 151.5 (143.5–157.5) μm at 6 months, 121.0 (109.5–131.5) μm at 12 months; p < 0.0001), as did outer retinal thickness (84.1 (30.7–123.4) μm at 3 months, 65.0 (11.0–97.6) μm at 6 months, and 41.0 (15.6–79.5) μm at 12 months; p < 0.0001; Figure 2C). By contrast, inner retinal thickness in SARDS‐affected dogs did not change significantly over time (90.12 (80.99–93.74) μm at baseline, 85.84 (80.70–93.45) μm at 3 months, 87.68 (81.39–92.38) μm at 6 months, and 80.16 (76.57–87.45) μm at 12 months; p = 0.2183), nor did it differ significantly from control dogs (92.27 (81.65–95.88) μm; p = 0.5475; Figure 2). Box and whisker plots showing total (A), inner (B), and outer (C) retinal thickness measured 1 mm temporal to the optic nerve head using Fourier‐domain optical coherence tomography in 15 SARDS‐affected and nine normal control dogs at baseline (both populations) and over time (SARDS‐affected dogs only). (A) No statistically significant difference in total retinal thickness was detected between SARDS‐affected and normal control dogs at baseline (p = 0.3044; Mann–Whitney test). However, at each follow‐up examination, total retinal thickness was progressively and significantly reduced compared to baseline (p < 0.0001; Kruskal–Wallis test followed by Dunn's multiple comparisons test). Statistical significance levels for pairwise comparisons are indicated on the graph (****p < 0.0001, **p = 0.0022, *p = 0.0196). (B) No statistically significant difference in inner retinal thickness was detected between SARDS‐affected and normal control dogs at baseline (p = 0.5475; Mann–Whitney test). Furthermore, in SARDS‐affected dogs, inner retinal thickness at all follow‐up examinations was not significantly different from that measured at baseline (p = 0.2183; Kruskal–Wallis test followed by Dunn's multiple comparisons test). (C) No statistically significant difference in outer retinal thickness was detected between SARDS‐affected and normal control dogs at baseline (p = 0.5451; Mann–Whitney test). However, at each follow‐up examination, outer retinal thickness was progressively and significantly reduced compared to baseline (p < 0.0001; Kruskal–Wallis test followed by Dunn's multiple comparisons test). Statistical significance levels for pairwise comparisons are indicated on the graph (****p < 0.0001, *p < 0.05). Retinal Abnormalities in Histologic Observation Detailed patient signalment, including the indications for enucleation, for all eyes included in the histological evaluation is presented in Table S2. All SARDS‐affected globes exhibited histologic evidence of diffuse outer retinal atrophy, characterized by photoreceptor loss. Although mild, multifocal thinning of the inner nuclear layer was sometimes observed, the inner retina was largely preserved even 7 years after SARDS diagnosis (Figure 3). Median (IQR) ganglion cell density did not differ significantly between SARDS‐affected and age‐matched control globes in the non‐tapetal (6.0 (5.3–6.8) vs. 6.0 (6.0–7.8) cells/field of view; p = 0.3259) or tapetal (12.5 (11.3–16.0) vs. 15.0 (13.5–16.8) cells/field of view; p = 0.1817) regions of the fundus. In both populations, ganglion cell counts were significantly lower in the inferior (non‐tapetal) retina compared to the superior (tapetal) retina (control eyes: p = 0.0002; SARDS‐affected eyes: p = 0.0002). The optic nerve head appeared normal in all globes examined from both populations. Representative photomicrographs of the retina and choroid from control and SARDS‐affected eyes. (A, B) Control eye from a 13‐year‐old female spayed Australian Cattle Dog without SARDS (Dog #1; Table S2). Note the normal retinal architecture in tapetal (A) and non‐tapetal (B) regions as evident by the distinct ganglion cells (black arrows) and inner nuclear layer (orange bracket). (C–H) Representative images from tapetal (C, E, G) or non‐tapetal (D, F, H) regions of SARDS‐affected eyes at various times following diagnosis. (C, D) Right eye of a 13‐year‐old female spayed Brittany Spaniel 1 year following SARDS diagnosis; (E, F) Right eye of a 13‐year‐old male Dachshund 3 years following SARDS diagnosis. (G, H) Right eye of a 14‐year‐old female spayed Miniature Schnauzer 7 years following SARDS diagnosis. Note that all SARDS‐affected eyes exhibit diffuse outer retinal atrophy with preservation of the inner retina including ganglion cells (black arrows) and (even at 7 years following diagnosis) only mild progressive multifocal thinning (black arrowheads) of the inner nuclear layer (orange bracket) in some cases. Artifactual retinal detachments (in C, D, F, G, H) present due to tissue processing. Discussion This study addresses a notable gap in the existing literature by providing a longitudinal evaluation of dynamic changes in cPLRs and retinal thickness in dogs with SARDS. While previous cross‐sectional studies have offered valuable insights [8, 17], the more recently noted progressive nature of some of the features in SARDS‐affected dogs necessitates long‐term assessments to truly understand the disease's impact on retinal structure and function. Our findings contribute significantly to this understanding, reinforcing the utility of chromatic pupillometry as a valuable adjunctive diagnostic and prognostic tool, and consistently demonstrating that SARDS is predominantly a photoreceptor degeneration with remarkable preservation of inner retinal anatomy and function. Our results corroborate previous observations regarding the distinct pupillary responses in SARDS‐affected eyes. At baseline, the marked reduction in red‐light induced pupillary constriction was evident, indicative of photoreceptor dysfunction. These findings align with those of Grozdanic et al. [8], who have reported impaired photoreceptor‐mediated responses in SARDS. Crucially, in our longitudinal assessment, the blue light responses, which reflect melanopsin‐expressing ipRGC activity, remained largely preserved and comparable to controls for up to 12 months, with no significant longitudinal change in extent, latency, or velocity of constriction. White light responses, which reflect a mixed contribution from both photoreceptor‐mediated and ipRGC‐mediated pathways, showed extent and latency of constriction did not change significantly over time, but velocity of constriction to white light significantly declined by 6 and 12 months compared with baseline. This divergence, a stable blue‐light response alongside a declining white‐light velocity, is consistent with continued deterioration of the photoreceptor‐driven component of the mixed white‐light signal, superimposed on a preserved ipRGC‐driven component, and further supports the selective resilience of ipRGCs relative to the photoreceptor‐mediated pathway in SARDS. This resilience of ipRGCs in the face of significant photoreceptor damage is consistent with previous research [9, 11] and is further supported by the stable extent, latency, and velocity of blue light pupil constriction throughout our study, suggesting the relative preservation of melanopsin‐expressing ipRGCs despite significant photoreceptor pathway dysfunction. Terakado and coauthors [17] reported that 75% of 24 SARDS‐affected eyes tested using the Melan100 exhibited cPLR responses consistent with those proposed as diagnostic of SARDS in the device's manual. However, there were some variations including weak miosis in response to red light in four eyes and to blue light in another four eyes [17]. These authors did not specify the time of assessment relative to onset of clinical signs, assessed only whether pupils responded (not extent, latency, or velocity of response), and evaluated eyes at only a single time point. Based on data we present here, the subjective variations in pupil responsiveness reported by Terakado et al. [17] could be explained by variations in the time from SARDS onset and the chronic multifocal inner retinal thinning observed histologically in our study. Indeed, while early cases might show specific patterns, structural assessments indicate that retinal changes are highly dynamic. For instance, some SARDS suspects or very early‐stage cases may exhibit a significantly thicker photoreceptor layer and inner retina compared to controls, representing perhaps the earliest detectable structural alterations [22, 23]. Our longitudinal evaluation revealed a detectable reduction in the extent of red‐light induced pupillary constriction within 1 month of diagnosis, with no statistically significant subsequent change over the ensuing 11 months. This early functional decline is mirrored by OCT findings showing a complete loss of definition of the photoreceptor bands and significant thinning of the outer nuclear layer across all quadrants shortly after onset of clinical signs [24]. Furthermore, the observation of increased vitreous haze in SARDS patients via OCT suggests an early inflammatory component that may influence these initial physiological responses [23]. Although the extent of red light PLR constriction did not change significantly over the follow‐up period, a larger cohort may reveal additional longitudinal change that this study was not sufficiently powered to detect. Further longitudinal evaluation of cPLR and retinal changes in larger cohorts of SARDS‐affected dogs is warranted. While the extent and velocity of pupillary constriction in response to red‐light stimulation were significantly reduced immediately following SARDS diagnosis, the latency of constriction was not significantly different from controls at baseline and showed no significant longitudinal changes. Reduced pupillary constriction velocity is not uncommon in patients with retinal diseases such as retinitis pigmentosa or diabetic retinopathy, and could correlate with functional vision impairment [25, 26]. In SARDS‐affected dogs, constriction velocity in response to red light was significantly reduced at baseline and remained consistently low throughout the study, indicating an early and substantial impairment of photoreceptor‐mediated pupillary responses. This contrasts with blue light responses, where velocity of pupil constriction remained unaffected over the entire study period, further supporting the greater resilience of the ipRGC pathway compared to photoreceptor‐mediated PLR pathways in SARDS‐affected dogs. Although red‐light latency did not change significantly over the study period, a post hoc power estimate based on the observed variability suggests that detecting a difference of this magnitude with adequate power would require a substantially larger cohort than was available in this study. A larger, adequately powered longitudinal study is therefore warranted to determine whether this represents a true latency trend. Using FD‐OCT, we gained crucial in vivo insights into the ongoing structural changes occurring in SARDS‐affected eyes, which help explain the sustained responses to blue light. Over a 12‐month period, we observed progressive outer retinal thinning in these eyes. This aligns with the known pathophysiology of primary photoreceptor degeneration in SARDS [19, 20, 21], where rapid photoreceptor outer segment loss occurs due to apoptosis within 3 weeks of blindness onset, followed by a slower outer retinal degeneration, ultimately leading to end‐stage retinal degeneration [21, 27]. OCT and fluorescein angiography have further documented that this progression can include multifocal strictures (or “boxcar” segmentation) of the retinal vasculature and even small retinal detachments with associated leakage of the retinal pigment epithelial barrier during metaplastic changes in these cells [24]. A unique finding of our study is the long‐term preservation of inner retinal architecture, including ganglion cells, for up to 7 years post‐diagnosis, as confirmed by histology. This corroborates our FD‐OCT findings and underscores SARDS' localized and permanent impact on the outer retina. The preservation of the inner retina, despite significant outer retinal damage, likely accounts for the sustained blue‐light responses even with overall retinal thickness loss. Collectively, our functional (pupillometry) and structural (FD‐OCT and histology) data strongly support the hypothesis that SARDS is primarily a photoreceptor degeneration with a relative preservation of the inner retina and ipRGC function. While FD‐OCT provides valuable in vivo structural information, future studies incorporating functional assessments, such as multifocal ERG, are warranted for a more complete understanding. The sustained preservation of inner retinal structure and function, particularly the melanopsin‐mediated pupillary responses, opens the door for the potential use of optogenetic therapies to restore vision in SARDS‐affected dogs. Optogenetic therapy is an innovative approach that combines genetic and optical techniques to control neuronal activity through the use of light‐sensitive proteins known as opsins [28], and has shown promise for treatment of degenerative retinal diseases [28, 29]. By delivering opsin‐encoding genes to surviving bipolar and ganglion cells, this technique aims to restore light sensitivity to these cells [30]. The fact that blue‐light responses are maintained even in the face of progressive outer retinal thinning suggests that the ipRGCs remain functional. This observation combined with maintenance of a normal inner retinal thickness suggests that other RGCs and bipolar cells also remain intact and functional and are ideal candidates for optogenetic reprogramming. Therefore, our data suggest significant potential for optogenetic therapy in SARDS‐affected dogs by directly inducing light sensitivity in surviving RGCs and bipolar cells. In summary, this longitudinal study supports the utility of chromatic pupillometry, particularly using the Melan100, as a valuable adjunctive diagnostic tool for SARDS in dogs. The consistent dissociation observed between red and blue light pupillary responses, combined with the outer retinal thinning seen with FD‐OCT and histology, clearly reflects selective photoreceptor degeneration alongside resilience of ipRGCs in this disease. These insights not only deepen our understanding of SARDS but also highlight a significant potential for vision restoration through optogenetic therapies in affected dogs. Future research involving larger cohorts and the integration of more advanced functional assessments will be vital for more comprehensive understanding and to further refine diagnosis, monitoring, and management of this debilitating condition. Author Contributions David J. Maggs: writing – review and editing, validation, supervision. Ann E. Cooper: writing – original draft, writing – review and editing, conceptualization, investigation, methodology, funding acquisition. Soohyun Kim: investigation, methodology, writing – original draft, writing – review and editing, visualization, formal analysis, data curation. Tran Nguyen: software, project administration, resources, data curation. Ariana D. Marangakis: project administration, resources, data curation. Sara M. Thomasy: methodology, conceptualization, funding acquisition, investigation, writing – review and editing, validation, supervision. Monica Motta: project administration, validation. Funding This work was financially supported by the Vision for Animals Foundation (VAF2016‐01), the University of California Davis School of Veterinary Medicine Center for Companion Animal Health (2015‐14F), the National Institute of Health K08 (EY021142) and a generous donation from Pat and Patty Stoffers in memory of their dog Sophie. Ethics Statement Ethical approval was obtained from the University of California's Animal Care and Use Committee, protocol #19355. Conflicts of Interest The authors declare no conflicts of interest. Supporting information Table S1: Breed, age, sex, and body weight of 15 SARDS‐affected and nine normal control dogs included in the in vivo aspects of this study (FS, female spayed; MC, male castrated). Table S2: Breed, age, sex, and reason for enucleation for dogs contributing globes for histological analysis in this study. FS, female spayed; KCS, keratoconjunctivitis sicca; M, male; MC, male castrated; OD, oculus dexter (right eye); OS, oculus sinister (left eye); OU, oculus uterque (both eyes). Acknowledgments The authors gratefully acknowledge Gillian C. Shaw and the Comparative Ocular Pathology Laboratory of Wisconsin (COPLOW) for providing eight globes from five dogs with SARDS and eight age‐matched control globes from their database. The authors recognize Tanya Garcia‐Nolen for electrical engineering assistance with pupillometer construction and Nadya Dooley for technical support with Raspberry Pi programming. They also thank the following undergraduate student workers in the laboratory: Alyssa Hohen, Connor Chang, Josefina Nordenstahl, Carina Pasqualino, Danielle Madrigal, Madison Mukai, Meghan Gann, Megan Gragg, Sadie Serdhal, and Stephanie Stromberg. Artificial Intelligence Statement: The authors have used AI only for language editing (grammar, clarity, and style). The authors reviewed all edits and take full responsibility for the final content. All intellectual content, analysis, and conclusions are the original work of the authors. Data Availability Statement The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions. References Associated Data Supplementary Materials Table S1: Breed, age, sex, and body weight of 15 SARDS‐affected and nine normal control dogs included in the in vivo aspects of this study (FS, female spayed; MC, male castrated). Table S2: Breed, age, sex, and reason for enucleation for dogs contributing globes for histological analysis in this study. FS, female spayed; KCS, keratoconjunctivitis sicca; M, male; MC, male castrated; OD, oculus dexter (right eye); OS, oculus sinister (left eye); OU, oculus uterque (both eyes). Data Availability Statement The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.