Evaluation of radiographic, computed tomographic, and cadaveric anatomy of the head of boa constrictors (2024)

Materials and Methods

Animals—Four Boa constrictor imperator cadavers were obtained for this study; 3 were male and 1 was female, and snakes weighed from 3.4 to 5.6 kg and had a body length ranging from 189 to 221 cm.

All of the animals were referred to the Department of Veterinary Clinical Sciences at the University of Padua for specialty examination and were euthanized, with their owners' consent, because of advanced clinical conditions. A complete postmortem examination was performed on each snake, which revealed pneumonia in 3 cases and egg retention in 1 case. Gross examination ruled out any lesion in the head of each snake.

Imaging procedures—To minimize postmortem changes, the heads were dissected immediately after death and radiographic and CT studies were performed on each specimen. Right and left laterolateral and dorsoventral and ventrodorsal x-ray views were obtained by use of conventional x-ray equipment^a (140 kVp; 600 mA) and a high-detail screen-film combination.^b The CT images were obtained in transverse and sagittal planes by means of a single-slice third-generation CT scanner^c by use of the following scanning protocol to ensure optimal image quality: axial acquisition mode, rotation time of 7.6 seconds, voltage of 75 kV, amperage of 80 mA, and slice thickness of 1.5 mm; the images were then displayed in a bone tissue window (window length, 500; window width, 3,000). Some tests that used both higher peak kilovolts and higher milliamperes (with lower rotation time) on the same specimens were performed, but the images were of lower quality.

Anatomic dissection—All specimens were immediately dissected after the imaging procedures. Two heads were dissected following a strati graphic approach (Figures 1 and 2), whereas the remaining 2 specimens were designated for cross-sectional anatomic studies following the same planes of the CT studies: transverse (Figures 3–8) and sagittal (Figures 9 and 10). These latter specimens were placed on a plastic support, moved to a freezer (−20°C) immediately upon CT scan completion, and carefully maintained in the same position as in the CT study for 24 hours.

Cross-sectional anatomic dissection was performed strictly following the imaging protocol by means of an electric band saw. However, each section was 3 mm thick and therefore included 2 contiguous CT slices. The difference between the CT and anatomic slice thickness was meant to curtail the likelihood of cut errors (which would be greater with a finer slice thickness) and therefore to maintain the best correlation between the cuts and imaging sections. Slices were cleaned with water, numbered, and photographed on the cranial and caudal surfaces.

Individual anatomic structures were first identified in the anatomically dissected and cross-sectioned heads on the basis of anatomic references and then matched with the corresponding structures in the radiographs and CT scans. Because of the absence of a single reference for the anatomic nomenclature, individual anatomic structures were named following the only international publications on ophidian anatomy presently available.^28–33 Not all structures identified in the cadavers were identified on the radiographs and CT scans and vice versa.

To overcome the superimposition of structures, radiographic images and dissection photographs were matched. Dorsoventral (Figure 1) and right lateral (Figure 2) radiographic images of the head were matched with corresponding photographic images obtained during superficial and deep plane stratigraphic dissection. Lines were superimposed on a photograph of a boa constrictor head (Figure 3) to indicate the approximate levels of the sections of CT studies (Figures 4–10). Matched transverse and sagittal sections obtained from the cross-sectional studies were selected.

Results

The limited quality of the CT images was the consequence of an intrinsic lack of resolution in the CT technique applied to small-sized specimens and the inability to reduce the field of view of the device to less than the limit of 16 cm.

All clinically relevant structures of the head were indicated in the cross-sectional and anatomic dissections. A small amount of mucus could be seen in the oral cavity of the cross-sectioned head (Figures 4–8) because the animal was affected by pneumonia.

Bony structures of the head were evident on radiographs and CT scans. The bones composing the lower jaw had less than the minimum resolution level for radiographs and CT scans, so they appeared as a unique bone. Soft tissues were poorly defined on radiographs; only the masticatory muscles were evident on the dorsoventral and ventrodorsal radiographs as a soft tissue opacity behind and around the supratemporal and quadrate bone. On the CT scans, most of the soft tissues were not distinctly recognizable.

Discussion

Reptiles, like mammals, have great anatomic variability among species, but unlike mammals, they also have extreme individual variability. It is important not to use results of the present study for interpretation of imaging of other snake species unless specific differences among the species under analysis and boa constrictors are well-known.

Snakes are the only vertebrates known to ingest whole prey larger in mass and diameter than themselves. This ability derives from a unique head anatomy not found among other animals.

The snake skull is composed of a snout and a braincase that represent the fixed regions of the skull. Loosely attached to these are the palatomaxillary apparatus (divided into maxilla and pterigoideus), which is hom*ologous to the maxilla, and a series of bones (supratemporal, quadrate, articular, surangular, dentary, coronoid, and splenial). Together, these bones are hom*ologous to the jaw in mammalians (Figures 1–10). The 2 sides of the lower jaw are divided and independent of each another and attached by a loose fold of skin and mucosa.

The joints between the supratemporal and quadrate bones and the lower jaw are extremely loose; this makes positioning of the snake head during imaging studies difficult because it is hard to obtain perfect symmetry between the 2 sides of the head. For this reason, evaluation of symmetry of a radiograph is better performed by positioning the snout and braincase because of their fixed position.

The bony structures of the skull were evident with CT and radiographic techniques. Nonetheless, some bones were not detectable as single elements when CT was used because of their limited size (eg, bones of lower jaw), which was smaller than the minimum resolution in conventional CT techniques. A more accurate evaluation is possible only through high-resolution x-ray CT devices such as mini-CT and micro-CT, which are normally used for laboratory animals with a body size similar to that of the specimens of the present study. However, such technology is not presently available in veterinary clinics but only in some specialized research centers.³⁴

Snakes have an acrodontal tooth attachment: a dental formation whereby the teeth are consolidated with the summit of the alveolar ridge of the bone without sockets. The teeth are arranged in 1 row on each jaw and 2 rows on each palatomaxillary apparatus. Teeth are periodically replaced throughout a snake's lifetime; dental buds develop on the base and inside the previous tooth while it is still functional.³⁰ Tooth and dental cavity anatomy were well visualized on radiographs and CT images (Figures 1, 2, 4–6, and 9).

The musculature of the head is well developed in boa constrictors; the ability of a boa constrictor to catch live prey requires enormous strength in relation to its small head size. The major musculature components are located on the caudolateral side of the head, where the adductor muscles are situated. They are distributed in superficial and deep planes. The musculature of the head was not well visualized on radiographs (Figures 1 and 2) but was evident on CT images (Figures 6–10).

The snake ear is composed of only inner and middle compartments and completely lacks an external portion. The middle ear is composed of the tympanic cavity (lacking a tympanic membrane), which is a narrow fissure, and the ossicular chain (columella and extracolumella [cartilaginous expansion of the distal portion of the columella]), which passes through an air space and inserts medially on the oval window of the cochlea. The inner ear is quite complex; it is composed of the bony and membranous labyrinths. The last part of the cochlear channel is the lagena and is covered by a coating of otoliths (limestone concretions),³⁵ which were clearly visible on radiographs (Figure 1) and CT images (Figures 9 and 10).

The oral glands in snakes differ greatly depending on the species; boa constrictors have many well-developed oral glands distributed all around the lips (premaxillary, supralabial, and infralabial glands) and in the oral cavity (sublingual and palatine)²⁹; boa constrictors have, like all other boids, no venomous gland. The oral glands were not visualized on radiographs and were difficult to identify on CT images (Figures 1, 2, 4–7, 9, and 10); we hypothesize that the use of contrast medium could help in distinguishing oral glands from other structures in living animals.

Snakes possess a well-developed Harderian gland, which is a lachrymal gland extending from the rostral part of the orbit along its medial side and ventral to (and sometimes around) the optic nerve and back into the temporal region.³⁶ All of our specimens had a well-developed Harderian gland surrounding the optic nerve (Figures 1, 5, 6, and 9). The Harderian gland was well visualized on CT scans; because of the superimposition of other structures, it was not visible in the radiographs.

The small size of the head and presence of a high number of superimposed structures make correct positioning during imaging studies mandatory. Particular attention must be paid to correctly position movable structures such as jaws.

Radiographs provided a high level of detail regarding the bony structures and could be useful in evaluating pathological changes such as fractures, bone neo-plasia, and bone demineralization. Soft tissue definition was poor, and detailed evaluation was difficult.

Most of the main boa constrictor head structures were well visualized on CT scans. Nevertheless, as in mammal CT studies, soft tissues with similar structure and density, such as glands and muscles, are difficult to distinguish, particularly in cadavers, where the lack of vascularization reduces contrast. We believe that in live animals, injection of contrast medium could provide better differentiation. Moreover, multislice helical scanners (enabling reduction of slice thickness to sub-millimetric values) with a dynamic field of view could provide higher-quality images and better resolution of bone and soft tissues.