Representative images obtained using spectral-domain optical coherence tomography (SD-OCT; iVue, Optovue, Fremont, CA, USA) from a French Bulldog. A and B) 6 × 6 mm retinal maps displayed with the ETDRS grid and associated values, overlaid on the En face image of the mapped region, alongside the corresponding segmentation map highlighting the superior (S), inferior (I), nasal (N), and temporal (T) quadrants. C) Central raster scans (seven in total), each representing the average of five acquisitions, with selection of the evaluated retinal region. D) SD-OCT image selected raster scan used for morphological analysis.

Regarding the outer retinal layers, the highest mean thickness was observed in the dorsal outer region (135.58 ± 23.37 µm), while lower values were found in the central outer and ventral outer regions, both approximately 133 µm. The temporal outer region showed intermediate values with a narrow confidence interval (138.41–147.00 µm), indicating reduced data dispersion. For the inner retina, the highest mean thickness was found in the ventral inner region (55.36 ± 10.66 µm), whereas lower values were recorded in the central inner (50.67 ± 8.32 µm) and dorsal inner (50.24 ± 11.47 µm) regions.

When the ganglion cell complex (GCC) was assessed (Fig. 3), all evaluated sectors exhibited a normal distribution; therefore, mean ± SD values were used, as presented in Table 2.

Representative images obtained using spectral-domain optical coherence tomography (SD-OCT; iVue, Optovue, Fremont, CA, USA) from a French Bulldog. A) En face image of the area centralis selected for evaluation. B) Pachymetry map of the ganglion cell complex (GCC) layer. C) Cross-sectional SD-OCT image showing the ganglion cell complex, including the retinal nerve fiber layer (RNFL), ganglion cell layer (GCL), and inner plexiform layer (IPL).

Linear regression analysis of ganglion cell complex (GCC) thickness as a function of age (in months) in French Bulldogs. A significant negative correlation was observed between age and GCC thickness (red line; p < 0.05), with the 95% confidence interval shown in pink. Each yellow dot represents an individual measurement.

4. Discussion

Spectral-domain optical coherence tomography (SD-OCT) has emerged as a noninvasive and highly accurate diagnostic tool for acquiring high-resolution cross-sectional images of the retina (Cubuk et al. 2016). Technological advancements have also enabled the acquisition of detailed images of other ocular structures, including both the anterior and posterior segments. In our study, all SD-OCT images were of high quality regardless of the specific dog’s cornea, angle, and retina evaluated, and the procedure was well tolerated by the dogs.

OCT assessment can be performed both quantitatively and qualitatively. For quantitative analysis, normative databases (NDBs) are used to compare the acquired images with reference values. In human ophthalmology, various ethnic groups have been studied to provide more accurate data for interpreting OCT results (Cubuk et al. 2016; Antar et al. 2019). In veterinary ophthalmology, although the use of OCT has advanced in both research (Alario and Pirie 2013) and clinical practice (Famose 2014), additional data are needed to establish NDBs for different species and breeds.

When considering breed-specific NDBs in veterinary ophthalmology, available data are scarce. Central corneal thickness (CCT) and CET have been evaluated in healthy dogs by comparing ultrasonographic pachymetry with OCT measurements (Alario and Pirie 2014; Strom et al. 2016). The mean CCT for all eyes examined via SD-OCT was 587.7 ± 32.44 µm, and no significant difference in CCT was found based on the age or sex of the animal (Alario and Pirie 2014). In another study (Strom et al. 2016), CCT was assessed in Beagle dogs using time-domain (TD-OCT) and Fourier-domain OCT (FD-OCT). The mean CCT measured by TD-OCT in 152 eyes of Beagle dogs was 594.81 ± 33.02 µm. No significant effect was observed for the eye evaluated (p = 0.820); however, a significant sex-related effect was identified (p = 0.034), with intact males (600.16 ± 32.84 µm) exhibiting significantly greater CCT compared to intact females (580.32 ± 29.24 µm). Across all TD-OCT data, CCT significantly increased with age (p < 0.001). When comparing the two OCT modalities, TD-OCT yielded significantly higher CCT values than FD-OCT (p < 0.001) (Strom et al. 2016). In a study group of eight dogs, the average corneal thickness was 535 µm (range 500–620 µm) (Famose 2014). In our study, the mean CCT was 561.97 ± 45.26 µm, which is slightly lower than values previously reported in the literature (Famose 2014). Moreover, although no statistically significant difference was observed between sexes (p = 0.513), contrasting with findings reported in Beagles (Strom et al. 2016), there was a trend toward higher total corneal thickness values in males. These differences are clinically relevant, particularly in the context of corneal ulcerative lesions in brachycephalic breeds, which are predisposed to both ulcerative and non-ulcerative keratopathies.

Regarding corneal epithelium thickness (CET), data available in the literature are scarce. One study evaluated different regions of the corneal epithelium in Beagle dogs using FD-OCT; values obtained for all locations differed significantly, with the central corneal, superior paraxial, and superior perilimbal epithelial thicknesses measured at 52.38 ± 7.27, 56.96 ± 6.47, and 69.06 ± 8.84 µm, respectively (Strom et al. 2016). In another study evaluating healthy dogs and cats, the average measured epithelial thickness was 55 µm in dogs (range 50–60 µm) and 60 µm in cats (range 55–65 µm) (Famose 2014). An in vitro study conducted in rabbits using digitally recorded OCT images obtained limbal and CET measurements for 12 corneas. The limbal epithelial thickness (37.6 ± 1.4 µm) was significantly less than the central thickness (45.8 ± 2.2 µm; p < 0.001) (Reiser et al. 2005). In contrast, the mean value of CET in the brachycephalic group was 58.7 ± 6.3 µm, while in the non-brachycephalic group it was 55.4 ± 5.2 µm; there were no significant differences between groups (Jeong et al. 2023). In our study, the mean CCT was 561.97 ± 45.26 µm, slightly lower than previously reported values (Reiser et al. 2005; Strom et al. 2016; Famose 2014). Although no statistically significant difference was observed between genders (p = 0.513), in contrast to data for Beagles (Strom et al. 2016), there was a tendency toward slightly higher total corneal thickness in males. These findings are particularly relevant when assessing corneal ulcerative lesions in brachycephalic breeds, given their higher predisposition to both ulcerative and non-ulcerative keratopathies.

Regarding CET, published data remain limited. In our study, CET was higher (78.88 ± 8.77 µm) than previously reported (Reiser et al. 2005; Strom et al. 2016; Famose 2014). Although CCT appears thinner in French Bulldogs compared to other breeds, the corneal epithelium is thicker, which may represent a breed-specific trait. Previous studies reported CET values of 58.7 ± 6.3 µm in brachycephalic dogs, lower than those observed in the present study (Jeong et al. 2023). This discrepancy may be attributed to the lack of breed standardization among brachycephalic dogs or the manual measurement of epithelial thickness by different observers. Recently, a study in brachycephalic dogs with dry eye disease (DED) found significantly higher CET values in DED patients (122.24 ± 53.96 µm) compared to healthy dogs (75.46 ± 11.27 µm; p < 0.001) (Brito et al. 2025). These results highlight the importance of evaluating CET in both healthy dogs and those with DED. Detailed morphological analysis of the normal corneal epithelium using OCT provides essential insights into the pathophysiological processes affecting the ocular surface and allows more accurate interpretation of epithelial changes in diseased corneas.

For iridocorneal angle (ICA) assessment, additional diagnostic tools such as high-resolution ultrasound or ultrasound biomicroscopy (UBM) are often required to visualize the entire ciliary cleft (Kim et al. 2023). In one study using SD-OCT in dogs of various breeds, the mean ICA was 31.4 ± 6.4° (range: 20.8–52.3°) at the first measurement and 31.0 ± 5.7° (range: 16.2–47.4°) at the second, with different observers (Shim et al. 2022). In Beagle dogs, ICA values obtained with the Visante® OCT were reported as 42 ± 4° (Almazan et al. 2013). In our study, ICA values were higher (55.68 ± 9.88°), suggesting either a breed-specific anatomical characteristic or differences related to the imaging equipment used. Although gonioscopy is widely recognized as the gold standard for ICA assessment in both humans (Kato et al. 2006) and animals (Ekesten and Narfström 1991), OCT-based values may serve as normative database parameters for French Bulldogs and aid in the future evaluation of glaucoma. Regarding ICA in dogs using OCT, it is important to note that the canine anterior chamber is anatomically deeper than that of humans, which limits the visualization of the true angle recess. This anatomical feature often prevents complete delineation of the ICA structures, particularly the scleral spur and trabecular meshwork, in adult animals. Consequently, accurate measurement of the ICA by OCT is feasible mainly in young dogs, in which reduced ocular dimensions and less pronounced iris curvature allow improved optical access to the angle region.

OCT has increasingly become a valuable tool for assessing the retina and optic nerve in veterinary ophthalmology, offering high-resolution, non-invasive imaging of these structures. Recent efforts have focused on establishing normative data for retinal evaluation in animals (Hernandez-Merino et al. 2011) as well as identifying retinal abnormalities using OCT (Braga-Sá et al. 2018). In one study, whole retinal thickness in the superior temporal region (corresponding to the area centralis) was 198.7 ± 9.6 µm, while the inferior temporal region measured 164.4 ± 6.4 µm; the difference was statistically significant (p < 0.0001, paired t-test) (Hernandez-Merino et al. 2011). In another study evaluating diabetic dogs, total retinal thickness in the non-diabetic control group ranged from 203 µm to 238 µm (Braga-Sá et al. 2018).

Retinal thickness was also assessed in female Beagles at different life stages—puppies, adults, and elderly—with measurements taken in the dorsal, medial, ventral, and lateral quadrants at concentric distances from 1 mm to 6 mm from the area centralis. In adult dogs, average total retinal thickness values ranged from 234 µm to 175 µm (dorsal), 203 µm to 166 µm (medial), 170 µm to 132 µm (ventral), and 202 µm to 191 µm (lateral) (Ofri and Ekesten 2020). In the same study, outer retinal thickness revealed regional variation: dorsal (78 ± 8 µm to 70 ± 9 µm), medial (77 ± 7 µm to 67 ± 8 µm), ventral (69 ± 6 µm to 59 ± 5 µm), and lateral (81 ± 6 µm to 75 ± 7 µm). In our study, the mean total retinal thickness in the central region was 187.70 ± 23.90 µm (95% CI: 179.22–196.17 µm). In the nasal, temporal, and inferior regions, median values were 192.0 µm, 195.0 µm, and 193.0 µm, respectively. The dorsal region had a mean of 186.09 ± 29.99 µm, consistent with previously reported data (Hernandez-Merino et al. 2011; Braga-Sá et al. 2018; Ofri and Ekesten 2020). Such concordance reinforces that the measurements obtained reflect normal anatomical variability among canine breeds rather than methodological or instrumental bias. Therefore, the data presented here contribute to expanding normative reference values for retinal thickness in dogs, particularly brachycephalic breeds, within the expected physiological range previously described for healthy canines.

For outer retinal thickness, the highest mean value was observed in the superior external region (135.58 ± 23.37 µm), while the lowest mean values were recorded in the central and inferior external regions (both approximately 133 µm). Our results were higher than those reported in Beagle dogs (Ofri and Ekesten 2020), likely due to methodological differences. In our study, measurements were performed automatically by the OCT software, reducing measurement error and reinforcing the need for standardized NDBs by species and device. These findings support the existence of regional retinal thickness variations in dogs, which must be considered when using OCT in clinical evaluations, particularly in comparative studies and retinal disease diagnostics.

Data on inner retinal thickness (IR) in dogs are scarce. A canine study evaluating different retinal regions and age groups showed progressive thinning with increasing age (Occelli et al. 2020). In that study, IR thickness was highest in the area centralis (97.3 ± 10.5 µm to 123.8 ± 7.7 µm) and thinnest in the ventral region (78.7 ± 4.8 µm to 51.7 ± 5.5 µm). In our study, IR thickness values were lower, ranging from 50.24 ± 11.47 µm to 55.36 ± 10.66 µm. This discrepancy may be due to differences in anatomical region assessed, age of the dogs, or automated analysis used.

A study evaluating GCC thickness in dogs showed that GCC thickness varied by region and decreased with age in most retinal areas (Occelli et al. 2020). In the dorsal and temporal regions, values remained relatively stable, ranging from 55.3 ± 1.6 µm to 58.0 ± 6.3 µm (dorsal) and from 67.2 ± 1.4 µm to 77.1 ± 2.5 µm (temporal). The area centralis consistently showed the highest GCC thickness, peaking at 96.3 ± 8.5 µm at four weeks and gradually declining to 72.0–76.7 µm in older dogs. In our study, the mean total GCC thickness was 72.38 ± 9.52 µm (95% CI: 68.94–75.81 µm). These findings corroborate previous reports and support the observation of age-related GCC thinning. We observed an approximate decrease of 0.42 µm in GCC thickness per month of age. Therefore, age should be considered when interpreting GCC measurements in dogs.

A potential limitation of our study was the sample size, as a larger sample would likely provide a more normal data distribution and reduce the influence of outliers. Another limitation was the absence of gonioscopic ICA evaluation and correlation with OCT-derived values. Finally, only a single retinal sector was evaluated; assessing additional regions may yield more precise data regarding retinal thickness. However, the results obtained were statistically robust, indicating that these limitations likely did not affect the validity of the study’s results. Future studies should aim to establish NDBs with larger sample sizes, include age stratification, perform gonioscopy, and evaluate multiple retinal sectors.

In conclusion, this study generated novel normative OCT data for French Bulldogs, which may be valuable for future clinical and experimental evaluation of ocular disorders in this breed. Notably, we observed that French Bulldogs present with thinner CCT and thicker CET compared to other canine breeds. These differences should be considered when evaluating French Bulldogs with corneal diseases. SD-OCT proved to be a useful diagnostic tool for investigating ICA narrowing in normotensive French Bulldogs. Furthermore, our findings highlight regional retinal thickness variation in dogs, which must be taken into account in clinical and comparative retinal assessments. Finally, age should be considered an important factor influencing GCC thickness.

Representative images obtained using spectral-domain optical coherence tomography (SD-OCT; iVue, Optovue, Fremont, CA, USA) from a French Bulldog. A and B) 6 × 6 mm retinal maps displayed with the ETDRS grid and associated values, overlaid on the En face image of the mapped region, alongside the corresponding segmentation map highlighting the superior (S), inferior (I), nasal (N), and temporal (T) quadrants. C) Central raster scans (seven in total), each representing the average of five acquisitions, with selection of the evaluated retinal region. D) SD-OCT image selected raster scan used for morphological analysis.

Linear regression analysis of ganglion cell complex (GCC) thickness as a function of age (in months) in French Bulldogs. A significant negative correlation was observed between age and GCC thickness (red line; p < 0.05), with the 95% confidence interval shown in pink. Each yellow dot represents an individual measurement.