Magnetic resonance and computed tomographic features of 4 cases of canine congenital thoracic vertebral anomalies

Abstract

Magnetic resonance and computed tomography features of 4 cases of canine congenital vertebral anomalies (CVAs) are discussed. Two of the cases represent unusual presentations for such anomalies that commonly affect screw-tail or toy breeds. Moreover, the combination of CVAs and a congenital peritoneo-pericardial diaphragmatic hernia has never before been imaged.

Congenital vertebral anomalies (CVAs) may occur because of disturbances in embryonic development of the vertebrae including abnormal distribution of intersegmental arteries during the stage of resegmentation (1–3). These CVAs of the spinal column are not uncommon in dogs, but are seen less frequently in cats (4).

A variety of malformations is reported within the condition of CVAs considering both alterations in the shape and number of the vertebrae (hemivertebrae, butterfly vertebra, atlanto-axial malformations, cervical spondylomyelopathy, spina bifida, transitional vertebrae, and block vertebrae) (1,5,6). The term complex congenital vertebral anomalies has been used to denote the presence of several malformations occurring in an animal (7).

Most wedge-shaped vertebral deformities or hemivertebrae do not produce any clinical signs and are usually considered as incidental radiographic findings in dogs; they are most often diagnosed in the screw-tail breeds, affecting the thoraco-lumbar spine (1,6,8). However, they may be associated with moderate to severe angulation of the spine (scoliosis, kyphosis, or lordosis) and occasionally with narrowing of the spinal canal (particularly dorsal hemivertebrae) and instability of the involved segments producing spinal cord compression and intermittent trauma (2,9). The degree of spinal curvature depends on the number of vertebrae involved and the degree of each vertebral deformity (8).

Clinically affected animals are usually less than 1 y of age and clinical signs are suggestive of transverse myelopathy affecting T3 through L3 segments of the spinal cord. However, since the diagnosis of a clinical-CVA-related disorder in a young subject is challenging, a complete diagnostic work-up (for example, myelography and cerebrospinal fluid analysis) should be performed in any case (10). Hemivertebrae may also be associated with malformations in neural tissues such as spinal dysraphism or arachnoid cyst and with malformations of other organ systems (2).

Myelography is an accurate diagnostic technique to detect the precise cause and location of spinal cord diseases (9). However, the widespread availability of cross-sectional imaging techniques, such as magnetic resonance imaging (MRI) and computed tomography (CT), has enabled a significant improvement in diagnostic accuracy. Nevertheless, although MRI, CT, and CT-assisted myelography have been widely used in diagnosis of spinal cord compressions, only a few cross-sectional imaging features are available for domestic animal spinal cord CVA-related modifications in pets (4,11) and large animals (7).

The aim of this report was to describe the characteristic MRI (Cases # 1, 2, 3) and CT (case # 4) features of 4 cases of canine Complex CVAs that resulted in different clinical conditions.

Case description

Clinical cases

Four subjects (Case 1: a 6-month-old intact male pug dog; Case 2: a 5-month-old intact male English bulldog; Case 3: a 3-month-old female German shepherd dog; Case 4: a 1-year-old intact male flat-coated retriever) were referred to the authors’ institutions for neurological evaluation. Cases 1–3 had a history of weakness in the hind limbs, progressing to severe gait abnormalities and finally worsening to a non-ambulatory paraparesis. In Cases 1 and 2 lack of conscious proprioception was evident in both hind limbs; reflexes were normal in all 4 limbs in all dogs. In Cases 1 and 2 cutaneous trunci panniculus reflex was absent at any level; it disappeared caudal to the 9th thoracic vertebra in Case 3. There was no pain at spinal palpation in any of these cases. The complete blood (cell) count (CBC) and urinalysis were unremarkable for each of the first 3 subjects.

Case 4 was diagnosed as affected by thoracic CVA and a concurrent peritoneo-pericardial diaphragmatic hernia (PPDH) at 3 mo of age. The subject was referred at 12 mo for neurological assessment and possible PPDH pre-surgical planning. On physical examination, the dog had a lighter body weight and smaller stature than the average for his gender and breed, along with a dorsal deviation of the normal contour of the thoracic vertebral column and sternum. Furthermore, dyspnea was evident at rest and there was slight congestion of the mucous membranes. Muffled heart sounds and borborygmi were reported upon thoracic auscultation. The results of neurological examination were normal except for a slight, questionable, ataxia of the hind limbs. The CBC indicated mild mature neutrophilia [12.1 × 10⁹/L, reference interval (RI): 3.9 to 8.0 × 10⁹/L] and slight monocytosis (1.22 × 10⁹/L, RI: 0.22 to 1.1 × 10⁹/L). Serum biochemical abnormalities included mild hypoalbuminemia (22.8 g/L; RI: 26.0 to 30.0 g/L), elevated creatine kinase (CK, 172 IU/L; RI: 30 to 120 IU/L), and alkaline phosphatase (ALP, 192 IU/L; RI: 20 to 156 IU/L) concentrations. The urinalysis was unremarkable.

All subjects (Cases 1–4) had survey radiographs of the thorax, which revealed the presence of thoracic CVAs. To further evaluate for spinal cord conditions, Cases 2 and 3 were submitted to a C6-L3 spinal MRI, and Case 1 underwent a C5-S1 spinal MRI examination. Conventional thoracic CT examination was performed on Case 4 to evaluate both the PPDH and the thoracic spinal CVA.

Radiographic findings

Cases 1 and 2 showed a marked ventral deviation of the thoracic spine in a laterolateral radiographic projection centered on the T4–T5 vertebrae, followed by an abrupt dorsal angulation of the spine at the level of T8 and T7, respectively. By contrast, Cases 3 and 4 showed a moderate (#4) to marked (#3) kyphosis of the thoracic spine centered on the T10 and T5–T7 vertebrae, respectively. In Case 3, several vertebral bodies appeared to be incompletely developed, with lack of intervertebral spaces (Figure 1, A-C). Ventrodorsal radiographic views of the thoracic region revealed an S-shaped deformity (scoliosis) of the vertebral column in Cases 1–3. A detailed description of the thoracic spinal radiographic features of each dog is reported in Table 1. The cardiac silhouette of Case 4 was grossly enlarged and merged with the diaphragmatic outline caudo-ventrally (laterolateral projection) and centrally (ventrodorsal projection). In addition, gas bubbles were detectable within the soft-tissue density of the cardiac silhouette (Figure 1D).

Figure 1.

A,B,C — Laterolateral radiographic projections of the thoracic spine of Cases 1, 2, and 3, respectively; D — Laterolateral projection of the thorax of Case 4, showing the thoracic spinal CVA and the associated peritoneo-pericardial diaphragmatic hernia.

Table 1.

Detailed radiographic features of the thoracic congenital vertebral anomalies in each subject

Case number	Deviated spinal tract	Lordosis	Kyphosis	Scoliosis	Hemivertebra/misshapen vertebra	Butterfly vertebra	Spina bifida	Absent or reduced intervertebral space	Altered shape of spinous process
1	T1–T8	Yes	Yes	Yes	T8		T1		T5–T8
2	T1–T7	Yes	Yes	Yes	T5; T6; T7; T10	T10			T4–T6
3	T2–T11		Yes	Yes	T3; T4; T5; T6; T7; T8; T9; T10			T3/T4–T9/T10	T3–T10; fused T8–T9
4	T8–T11		Yes		T10; T11			T10/T11	T10

Magnetic resonance imaging and computed tomography findings

In patients 1,2,3, sagittal, transverse, and dorsal MR images were obtained by use of a 0.22-T permanent magnet (MrV; Paramed, Genoa, Italy). A 3rd generation conventional CT device (Tomoscan LX; Philips, Amsterdan, The Netherlands) was used to image the thorax in Case 4. Each MRI and CT scan confirmed the spinal deviations as described by individual X-ray examinations.

In Case 1, the intervertebral discs (IVDs) of C5-S1 had not degenerated. Nevertheless, misshapen T7–T8 and T8–T9 IVDs, with dorsal displacement of the nucleus pulposus, were found. The dorsoventral diameter of the vertebral canal was markedly reduced at the level of T8 but dilated along the cranial spinal segment T1–T7; in this latter area, the spinal cord was dorsally displaced inside the subarachnoidal space. The stenosis at the level of T8 caused a severe compression of the spinal cord (Figure 2, A-B).

Figure 2.

Case 1 — Midsagittal (A) T2-weighted MR image of the C6-T12 spinal tract showing the spinal canal stenosis at the level of T8 and the misshapen T7–T8, T8–T9 intervertebral discs. The spinal cord is severely compressed at the level of the stenosis (arrow). Transverse (B) T2-weighted MR image at the level of T8. The severe compression of the spinal cord is evident (arrow). Slice thickness was 3.5 mm.

In Case 2, the T5–T6, T6–T7, and T7–T8 IVDs were degenerate. The vertebral canal at the level of T7 was mildly reduced in its dorsoventral diameter but it was mildly enlarged along the cranial segment T1–T7; in this latter region, the spinal cord was dorsally displaced inside the subarachnoidal space. By contrast, the spinal cord appeared ventrally displaced at the level of T7, where a slight reduction in its dorsoventral plane could be observed (Figure 3, A-B).

Figure 3.

Case 2 — Midsagittal (A) and left parasagittal (B) T2-weighted MR images of the C7-L1 spinal tract showing the low degree of stenosis of the vertebral canal in spite of the severe lateral and ventro-dorsal deviation of the spine. The stenosis was confirmed on transverse views. The T5–T6, T6–T7 and T7–T8 intervertebral discs are degenerate. Slice thickness was 3.5 mm.

In Case 3, the IVDs were absent or completely degenerate from T3–T4 through T9–T10. Varying degrees of stenosis of the spinal canal due to severe malformation and malalignment of contiguous vertebral bodies T4–T9 were related to a variable level of spinal cord compression along the same spinal segment. Moreover, a left-sided deviation of the spinal cord was evident. A hydro-syringomyelia was present at the level of T2 (Figure 4, A-D).

Figure 4.

Case 3 — Midsagittal (A) and right parasagittal (B) T2-weighted MR images of the C6-T13 spinal tract showing a series of misshapen and misaligned vertebral bodies (T4–T9) and the resulting different degrees of spinal cord compression along the affected vertebral segment. Transverse T1-weighted MR images at the levels of T7 (C) and T1 (D) show the severe spinal cord compression (arrow) and the dilated ependymal canal (arrow head), respectively. Slice thickness was 3.5 mm.

In Case 4, the CT scan revealed that the liver, gallbladder, spleen, body of the stomach, pylorus, omentum, and proximal duodenum had herniated into the pericardial sac (Figure 5A). The CT scan also confirmed the presence of both T10 and T11 misshapen vertebral bodies (Figure 5B). Neither compressions nor anomalies affecting the spinal cord were observed along the thoracic spinal tract except for a slight right-sided deviation of the spinal cord at the level of T9 (Figure 5C) along with a ventral dislocation at the level of T10 (Figure 5B).

Figure 5.

Case 4 — A) Axial CT image of the thorax illustrating the cardiac silhouette (H) surrounded by the spleen (S), the liver (L), and the gastric fundus (GF) within the pericardial sac (slice-thickness was 10 mm). B) Axial CT image at the level of T10 showing the superimposition of both the misshapen T10 (*) on the T11 (**) vertebral body and the T10 thin spinous process (arrow). Ventral dislocation of the spinal cord is evident. (The CT-gantry tilt-angle was parallel to the T9–T10 intervertebral disc; slice-thickness was 3 mm). C) Axial CT image at the level of T9, showing the right-sided deviation (arrow) of the spinal cord. (The CT-gantry tilt-angle was parallel to the T8–T9 intervertebral disc; the slice thickness was 3 mm.)

Discussion

Congenital vertebral anomalies or hemivertebrae are commonly detected in brachicephalic breeds of dog, most notably in those that also have screw tails (such as English bulldogs, French bulldogs, Boston terriers, and pugs) or more generally in toy breeds, without any evidence of gender predisposition (8,9,11). Both the German shepherd dog and the flat-coated retriever reported here as Cases 3 and 4, respectively, are uncommon presentations for such spinal deformities.

Furthermore, with reference to Case 4, in canine medicine the PPDH has been associated with other congenital abnormalities such as umbilical hernias, pectus excavatum, sternebral abnormalities, and cardiac malformations (12,13). In humans, PPDH is often acquired secondary to trauma; however, in animals, it is almost always congenital. The cause of the anomaly may include prenatal injury, genetic defects, or a teratogen (14). The history of this patient revealed that during the 3rd week of pregnancy his dam was assaulted by another dog and consequently treated with amoxicillin for a week. It is not known whether the cause of such a lesion could be traumatic, genetic, or related to the antimicrobial drug. However, to the best of our knowledge, this is the first report of a canine PPDH associated with CVAs.

A further purpose of this paper was to show, through tomographic imaging techniques, the high heterogeneity of conditions that are gathered within the definition of congenital vertebral anomalies.

The most common CVA-affected site has been reported to be the mid-thoracic region, particularly in association with the T8 vertebra (11); the CVAs affecting our study subjects were all within the T3–T11 spinal tract. The number of misshapen vertebrae for each subject varied from 2 (Cases 1, 4) to 8 (Case 3), thus conforming with the definition of Complex-CVAs. Different degrees of abnormal spinal conformation were seen in all subjects, causing lateral deviation (scoliosis, Cases 1–3) with dorsal angulation of the spine (kyphosis, Cases 3–4) or a more complex deformity characterized by ventral deviation continuing with dorsal angulation (kypholordosis, Cases 1–2).

Radiographic findings of vertebral malformation and malalignment are not always associated with spinal canal stenosis and, conversely, a spinal cord compression or lesion can exist when survey radiographic findings are unremarkable. Thus vertebral column and spinal cord imaging are essential for a precise diagnosis (9).

The spinal cord Complex CVA-related tomographic imaging findings varied from: a) the absence of spinal canal stenosis/spinal cord compression (Case 4); b) through a moderate (Case 2) to severe (Case 1) localized stenosis/spinal compression; c) to multiple points of severe stenosis/spinal compressions with cranial hydrosyringomyelia, presumably secondary to the caudal severe stenotic condition (Case 3).

CVA-related clinical signs (when present) are suggestive of transverse myelopathy affecting the T3 through L3 segments of the spinal cord (11); the clinical presentations of all our study subjects match this condition. However, it should be stressed that, in our limited case series, the severity of the neurological deficits is not strictly related to the degree and severity of vertebral and spinal cord malformations. Consequently, a good neurological examination is mandatory and no prognostic considerations should be based on image analysis alone. CVJ