Sulci of the canine brain: a review of terminology

Over the last decades there have been several publications of anatomical and neurological textbooks, which include descriptions about the dogs’ brain. However, the terminology used is inconsistent, partly due to individual differences in neocortical gyration and partly due to the common practice of adapting terms from human and murine anatomy. In order to identify such incongruences, in Study 1, we reviewed the existing literature and identified the common terms used as well as any discrepancies between textbooks. Three main forms of inconsistencies were found; a) the use of terms that are not included in the Nomina Anatomica Veterinaria (NAV), b) the inclusion of structures that are listed as not canine-specific, and c) the use of similar names to identify potentially different anatomical structures. To address these issues, in Study 2 we investigated the consistency in appearance of the cerebral sulci, performing a macroscopical examination on 79 canine brains obtained through the Canine Brain and Tissue Bank (CBTB). We then evaluated whether sulci on the frontal regions of brachycephalic breeds differed from those of mesocephalic and dolichocephalic groups, as frontal and olfactory regions are subjected to the most extreme modifications following the shortening of the skull. The statistical analysis showed no difference across the skull length types regarding the occurrence of these sulci, although furrows on the lateral side of the brain proved to be more stable than those on the medial side. In Study 3, we summarized the findings in accordance with the NAV to produce a definitive index of the terms that we recommend be used for each identified sulci. Such an index is beneficial for educational, clinical use, and research (e.g. neuroscience) purposes. The dog is emerging as a pioneering and exceptional model in comparative neuroscience, and therefore the implications for canine neuroscience research should not be underestimated.

Brain atlases are therefore fundamental tools for several clinical and research purposes [1][2][3][4][5]. 61 One essential aspect of such research activities is the need to compare, and thus identify, brain 62 structures and areas across different modalities (from macroscopic modalities to histological 63 studies). For this to be done reliably, there is a need for tools such as universal atlases (i.e. an 64 atlas that allows viewing the same structure as depicted through various indirect imaging (CT, 65 MR) or through a more direct approach (surgical intervention and endoscopy)). Consequently, 66 the comparison of brain structures under different imaging modalities requires several atlases, 67 and those currently available in the literature are rarely standardized in their terminology and 68 present a large variability in terminology [6-9]. Thus, the first step to creating a comparative 69 brain atlas for dogs requires the specification of a common descriptive language and 70 clarification of the present terminology in order to make it comparable for every user. named Gyrus ectomarginalis caudalis [11][12][13][14], or Gyrus ectosagittalis caudalis [11,14], or 85 Gyrus suprasylvius posterior [6,15]). One reason for these incongruences is that in dogs there 86 are no clearly defined borders for individual brain lobes as there are in humans -where, for 87 example, the border between frontal and temporal lobes is defined by a well-recognized 88 anatomical element, the Sulcus centralis [16]. Furthermore, the surface morphology of dogs' 89 brains can vary slightly, even between individuals of the same breed type, due to the different 90 patterns of localization and/or length of their sulci [17,18].

91
One of the main issues that arises from such nomenclature incongruences is that it 92 makes it difficult to perform comparisons between research findings that use different terms.

93
Furthermore, the assessment of intracranial structures (including anatomical variations and 94 potential abnormalities) requires a common descriptive language between clinicians and 95 researchers. Similar issues have been resolved in the case of another species, the domestic cat, 96 through the production of a tissue probability feline cortical brain map. This map consists of a We aimed to review the current situation in the naming of the basic structures of the 100 canine brain, highlighting the discrepancies found in different literature sources, and then, 101 based on a study where we evaluated canine brains, we summarize the sulci that are highly 102 prevalent in dogs. We then present a summarizing In order to produce a comprehensive summary of the terms used in canine 117 neuroanatomy, we reviewed the literature searching specifically for canine brain cortical 118 features' terminology and investigated their consistency in a sample of unprocessed dog brains, 119 in three studies as follows: in Study 1, we aimed to assess the level of consistency among the 120 terms used in the veterinary clinical, educational, and research nomenclature for dog brains. In   The Sulcus corporis callosi was named Sulcus callosus in three textbooks [8,42,43]. 156 Additionally, in some textbooks, the suffix "lateralis" was used instead of "marginalis": this as Fissura rhinalis [6,44], and the Sulcus suprasplenialis as Sulcus ectosplenialis [39]. Finally, 161 some textbooks used the more human-related nomenclature based on anterior-posterior terms 162 in place of rostralis-caudalis [6,15]. 163 We identified terms that were commonly used in the textbooks and were not present in 164 the NAV; these were the Sulcus ectogenualis [8,13,15

175
A different type of incongruence emerged regarding the identification of certain sulci. 176 We identified inconsistencies in relation to the localization of the Sulcus ansatus [46]. The 177 following alternatives were found: some described it as a medial continuation from the Sulcus 178 coronalis to the Sulcus marginalis [14,15,43], or as a rostro-medial branch derived from the The brains were obtained from dogs after they had been euthanized for medical reasons 224 (that had no effect on the central nervous system gross morphology), and their bodies donated purposes prior to the current study; therefore, information about the sex, exact age, and breed 234 of the dogs could not always be obtained at the time of the brain anatomical assessment. 235 However, the dogs' head morphology group could be extrapolated from craniometrical data.  The sulci were grouped in two ways for further analysis. Initially, we identified the 281 sulci located in the medial area of the brain and those located in the lateral area of the brain.

282
Scores of the sulci in the two groups were averaged and compared to assess the consistency in were not normally distributed, therefore the data were analysed using non-parametric tests. All 292 tests were two-tailed and the level of significance was set at 0.05.

295
The scoring revealed that 31% of the sulci could be identified in more than 90% of the 296 hemispheres and 33% of the sulci were identifiable in less than 10% of the hemispheres (Fig   13   297 2). As expected, the Fissura longitudinalis cerebri was identified in all hemispheres (as it 298 should separate the two hemispheres, and only in pathological conditions disappears, where 299 the two halves of the brain fuses), also the Sulcus rhinalis lateralis pars rostralis could be 300 identified in all dogs, while the Sulcus rostralis was missing in the majority of the dogs. In 301 order to avoid pseudoreplication, for each sulcus of each dog, a mean score was calculated for 302 the sulci from the left and right hemispheric scores, which was then used for further analysis.   in Table 2 we provide the terms that primarily according to the NAV's guidelines but also 345 broadening its spectrum we recommend be used for each identified sulci and compared them  (Fig4/19). For the terms that had more than one alternative name based on the 363 literature, we recommend the use of the following terms (see also  [14].

411
In Study 2, we assessed the presence of the main sulci in a large sample of dogs and we furrows on the lateral side of the brain proved to be more stable than those on the medial side.

419
One limitation of the evaluation performed in Study 2, is that fixation in formalin ex situ can 420 cause flattening on the surface if the brain is not placed properly in the container, and surface 421 blood vessels can make impression on the cortex that appear similar to the original grooves.

422
While knowledge of basic topography may aid in differentiating in these cases, it was not 423 possible to exclude the chance that these types of events occurred. In future studies, the true 424 depth of the sulci would be assessed through T2-weighted MR imaging, where the Liquor 425 cerebrospinalis could clearly highlight the grooves. By making 3-dimensional models from the 426 MR images one could also interactively visualise not only the grooves but the gyri as well. 427 We have provided a thorough review of the current state of the terminology used to 428 describe the canine brain. Our summary and recommendations for appropriate terminology can 429 serve as a base for future canine brain atlases and medical terminology.