Upper Thoracic Disc Disease (T1–T9) in Large-Breed Dogs

Introduction

Degenerative intervertebral disc disease (IVDD) is one of the most common neurologic diseases in dogs. Although degeneration can occur at any area, the cervical (C), caudal thoracic (i.e., the tenth to thirteenth thoracic [T]), and lumbar (L) discs are the most commonly affected. Lesions located in the T1–T9 region are rarely seen. When lesions do occur in the upper thoracic region (i.e., T1–T9), it seems to happen predominantly in large-breed dogs.

Although there is an abundance of literature on IVDD in small-breed chondrodystrophic dogs, there is little data specifically focused on the occurrence of IVDD in large-breed dogs, especially large case reports. There is even less literature regarding upper thoracic IVDD (i.e., T1–T9) in dogs and it consists primarily of case reports or small case series.^1–5 According to Macias et al. (2002) T12–T13, T13–L1, L1–L2, and L2–L3 were the most common single sites affected in large-breed dogs. Forty-seven percent of those cases had multiple sites affected. There was no mention, however, as to whether any of the lesions were within the upper thoracic region (i.e., T1–T9).⁶ Of the 99 large-breed dogs in that study, German shepherd dogs were the most common large-breed dog affected with IVDD (33%).⁶ In Cudia and Duval’s study (1997), L1–L2 was the most commonly affected disc space in the large-breed dog, with L2–L3 being the second most commonly affected disc space. That study did not report any cases with lesions caudal to T10–T11.⁷ Only severely affected dogs that were surgical candidates were included in that study. Therefore, the study authors proposed that their study population may not represent the incidence of IVDD in the general population of large-breed dogs because many of the dogs with Type II disc protrusions were either subclinical or managed medically.⁷

The exact reason for the decreased clinical cases with IVDD in the upper thoracic region is unknown. The intercapital ligament connects the rib heads and provides an additional dorsal constraint under the dorsal longitudinal ligament, which is believed to reduce the rate of disc herniations from T1–T10.^8–11 Hansen (1952) reported that the intercapital ligament is either thin or nonexistent at T10, and Wilkens et al. (1996) proposed that the intercapital ligament may be thinner at T9–T10 in some dogs.^8–10,12 Movement around the cranial thoracic region is also thought to be limited by the large amount of muscle mass around the vertebrae, as well as the ligaments and musculature associated with the ribs.¹³

Since MRI became the standard for imaging in the study authors’ practices, several cases of dogs with upper thoracic (T1–T9) disc lesions have been identified. This retrospective study was prompted by a clinical impression that a significant number of large-breed dogs have intervertebral disc lesions between T1 and T9, where the intercapital ligament is proposed to be protective. During the period between 2002 and 2006, 17 dogs were diagnosed with upper thoracic discs. Ten of those were German shepherd dogs.¹⁴

The purpose of this study was to identify clinically affected large-breed dogs with IVDD in the upper thoracic area (i.e., T1–T9). The hypothesis was that there would be significantly more upper thoracic IVDD in large-breed dogs than small-breed dogs.

Materials and Methods

The medical records of all dogs at The Animal Specialty Center, Yonkers, New York, that had a neuroanatomic diagnosis of myelopathy on physical exam and IVDD confirmed via MRI between February 2008 and September 2011 were reviewed. Information retrieved from the medical records included age, breed, body weight, neuroanatomic localization, and MRI reports with specific locations noted to have intervertebral disc compression. All MRI reports were evaluated by either J.M.B. or R.J. Inclusion criteria were any dog that had myelopathy and IVDD with displacement of the spinal cord parenchyma noted on MRI. Cases were excluded if there was any other spinal pathology that was considered to be the primary problem. Such diseases included caudal cervical spondylomyelopathy (including disc-associated wobbler disease), discospondylitis, kyphosis, hemivertebrae, tumors, and atlantoaxial instability. If cases had IVDD at multiple regions, one region was chosen as the “primary” lesion region site based on neurologic examination and amount of spinal cord compression on MRI. The other regions of compression were noted as “multiple regions affected.”

Standard MRI^a sequences for a patient presenting with myelopathy include T2-weighted sagittal and transverse imaging. Additional sequences include T1-weighted pre- and postcontrast images and short T1 inversion recovery on a case-by-case basis (based on the T2-weighted images). Cases of cervical myelopathy were imaged from the brainstem to T2 and cases of thoracolumbar myelopathy were imaged from C7 to the third sacral vertebra (S3).

Dog breeds were separated into the small-breed and large-breed categories. To account for obesity and cachexia, dogs that were of breeds that are typically considered small-breed dogs (i.e., dachshunds) were kept in the small-breed category and dogs that were of breeds that are typically considered large-breed dogs (i.e., German shepherd dogs) were kept in the large-breed category regardless of weight. Dogs that were of breeds that range significantly in size (e.g., American pit bull terriers, basset hounds, mixed-breed dogs, and wheaten terriers) were allocated to the small-breed category if they were < 22.7 kg and to the large-breed category if they were > 22.7 kg. Lesions were recorded if there was IVDD noted with compression and then categorized into the following regions: cervical (C1–T1), upper thoracic (T1–T9), thoracolumbar (T9–L4), and lumbosacral (L4–S1). The T9–T10 region was grouped with the T10–L4 region as it had not been an uncommon site for IVDD in chondrodystrophic dogs at the authors’ hospitals.¹¹ The location of regions with lesions was categorized by size and breed.

Results

There were 723 dogs and 97 different breeds diagnosed with IVDD based on MRI findings from February 2008 to September 2011 that met the inclusion criteria of this study. There were 527 small-breed dogs (72.9%) and 196 large-breed dogs (27.1%). The most common small-breed dogs were dachshunds (n = 138), small mixed-breed dogs (n = 43), shih tzu (n = 29), beagles (n = 25), and pugs (n = 24). The most common large-breed dogs were Labrador retrievers (n = 41), German shepherd dogs (n = 37), mixed-breed dogs (n = 29), American pit bull terriers (n = 15), and doberman pinschers (n = 9). The mean age of all dogs was 7.4 yr (range, 6 mo–19 yr; median, 7 yr). The mean age of the small-breed dogs was 7 yr (range, 1 yr–19 yr; median, 6.5 yr), which was significantly lower than the mean age of large-breed dogs (8.6 yr; range, 6 mo–14 yr; median, 9 yr; P < 0.0001). The mean weight for all dogs was 16.3 kg (range, 1.4–72.7 kg; median, 10 kg). The mean weight for all small-breed dogs was 8.8 kg (range, 1.4–25.5 kg; median, 7.7 kg). The mean weight for all large-breed dogs was 36.4 kg (range, 20.9–72.7 kg; median, 35 kg). The least common lesion region for disc disease overall was T1–T9 (2.9%). There were no lesions at T1–T9 found in small-breed dogs, whereas 21 out of 196 large-breed dogs (10.7%) had T1–T9 lesions. Although lesions in the T1–T9 region were the least common site in dogs overall, it was the second most common lesion region in German shepherd dogs (29.7%) and golden retrievers (28.6%) as shown in Table 1. The most common primary upper thoracic lesion region site was T2–T3 (33.3%), followed by T4–T5 (25.9%), T3–T4 (18.5%), T5–T6 (11.1%), T6–T7 (7.4%), and T8–T9 (3.7%).

TABLE 1 Region of Primary Lesion Location Among Dogs

C, cervical vertebra(e); L, lumbar vertebra(e); S, sacral vertebra(e); T, thoracic vertebra(e).

View Fullscreen

There was a significant association between lesion region versus large- or small-breed dogs (P < 0.0001). Small-breed dogs were the majority in the C1–T1 and T9–L4 groups. There were 21 large-breed dogs (100%) in the T1–T9 group, with no small-breed dogs in that group, And breed was significantly associated with lesion region (P < 0.0001) as shown in Figures 1 and 2. The majority of dogs in the T1–T9 group were German shepherd dogs (n = 11; 52.4%; Table 2).

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

TABLE 2 Distribution of Dogs Within Regions

C, cervical vertebra(e); L, lumbar vertebra(e); S, sacral vertebra(e); T, thoracic vertebra(e).

View Fullscreen

Weight and age were also significantly associated with lesion region (P < 0.0001 and P < 0.0001, respectively). The mean weight of the C1–T1, T9–L4, and L4–S1 groups were 15.5 kg (range, 1.4–54.5 kg; median, 10 kg), 14.4 kg (range, 1.8–72.7 kg; median, 8.6 kg), and 24 kg (range, 3.6–54.5 kg; median, 21 kg), respectively. In contrast, the mean weight of the dogs in the T1–T9 group was 36.1 kg (range, 22.7–56 kg; median, 34.1 kg). The mean age of the C1–T1, T9–L4, and L4–S1 groups were 8 yr (range, 0.5–15 yr; median, 8 yr), 6.9 yr (range, 1–19 yr; median, 6 yr), and 7.8 yr (range, 2–14 yr; median, 7 yr), respectively. The mean age of dogs in the T1–T9 group was 9.3 yr (range, 5–12 yr; median, 9.5 yr).

Multiple regions with IVDD were significantly associated with increasing weight and age. The mean weight of dogs with lesions in multiple regions was 19.5 kg (range, 1.8–72.7 kg; median, 20.5 kg), whereas dogs with single regions had a mean weight of 15.7 kg (range, 1.4–59 kg; median, 9.5 kg; P = 0.007). The mean age of dogs with multiple regions affected was 9 yr (range, 2–14 yrs; median, 9 yr) versus dogs with a single region affected, which had a mean age of 7 yr (range, 6 mo–19 yrs; median, 7 yr; P < 0.0001).

Large-breed dogs were significantly more likely to have multiple regions affected than small-breed dogs (P = 0.0050; Table 3). Breed was also significantly associated with multiple regions affected (P = 0.0051) There were 13 German shepherd dogs with multiple regions affected, which was 35.1% of all German shepherd dogs with IVDD.

TABLE 3 Distribution of Dogs with Multiple Regions Affected with IVDD

C, cervical vertebra(e); IVDD, intervertebral disc disease; L, lumbar vertebra(e); S, sacral vertebra(e); T, thoracic vertebra(e).

View Fullscreen

The most common localization region to have multiple other regions affected was the T1–T9 group (P < 0.0001). There were 14 of 21 dogs (66.7%) in the T1–T9 group with multiple regions affected (Table 3), and 3 of 21 dogs had multiple sites affected within the T1–T9 region. All three of those dogs were German shepherd dogs.

Discussion

To the authors’ knowledge, this is the first study to report the prevalence of T1–T9 lesions in a large population of dogs. Notably, there were no T1–T9 lesions found in small-breed dogs, but 10.7% of all large-breed dogs that were diagnosed with IVDD based upon MRI had T1–T9 lesions. In other words, 100% of the included dogs with lesions from T1–T9 were large-breed dogs. Over half of the dogs with IVDD in the T1–T9 region were German shepherd dogs. It is interesting that this was also the second most common region affected in German shepherd dogs. The most common primary upper thoracic lesion sites within the T1–T9 region were T2–T3 (33.3%) and T4–T5 (25.9%). Dogs with IVDD in the T1–T9 region were older, heavier, and more likely to have multiple regions affected. German shepherd dogs were the only breed to have multiple sites affected within the upper thoracic region (n = 3).

The only literature on upper thoracic disc lesions in dogs that the authors are aware of consists of case reports and an unpublished retrospective study performed in 2011.⁵ The first two reports were single cases in dachshunds: The first was described by Gilmore et al. in 1983 and the second by Liptak et al. in 1999.^1,2 The most recent reports have focused specifically on the occurrence of upper thoracic disc disease in German shepherd dogs with the first documented case in 2002.³

The next report came as a short communication about three German shepherd dogs. All three dogs had disc protrusions at T2–T3 and hemidorsal laminectomies were performed.⁴ Gaitero et al. went on to present the first large case series in 2011, with 22 German shepherd dogs and 46 nonGerman shepherd dogs. They found that German shepherd dogs had a higher risk than nonGerman shepherd dogs for spinal cord compression. Among upper thoracic discs, risk was higher in the T2–T3, T3–T4, and T4–T5 regions, particularly T3–T4.⁵ The findings described in the current study support that data, with German shepherd dogs representing 52.4% of dogs with lesions in the upper thoracic region. This study also found that T2–T3 (33.3%), T3–T4 (18.5%), and T4–T5 (25.9%) had a higher risk of IVDD than the rest of the cranial thoracic region, but T2–T3 was the most common in our cohort.

Macias et al. (2002) found that German shepherd dogs were the most commonly affected large-breed dogs with IVDD overall.⁶ In Cudia and Duval's study (1997), mixed-breed dogs were the most common breed, with German shepherd dogs and Labrador retrievers next, respectively.⁷ The results of the current study support previous studies that found Labrador retrievers (20.9%) and German shepherd dogs (18.9%) as the most common large-breed dogs with IVDD.^6,7,15 It has been proposed that differences in biomechanical loading of the vertebral column may predispose certain dogs to either disc extrusion or protrusion, but there are no studies confirming that hypothesis. Differences in vertebral column torsional, compressive, and shear loads have been linked experimentally to disc degeneration, annular tearing, disc protrusion, and disc extrusion.^16–21 German shepherd dogs tend to have a sloping stance, which hypothetically may put more strain on the intervertebral discs, including those in the upper thoracic region. Further studies are needed to investigate that theory.

Historically, it was thought that the T1–T10 region was refractory to IVDD due to extra stability provided by the intercapital ligament, the sternum, and the rib cage. The intercapital ligament runs between the heads of the ribs, articulating with the thoracic vertebra. It runs dorsal to the annulus fibrosis of the intervertebral disc but ventral to the dorsal longitudinal ligament.¹² It is interesting that the caudal most thoracic vertebra with an intercapital ligament is not clearly defined in the literature. According to Miller, the intercapital ligament connects the right and left rib heads from the second through tenth ribs²²; Thacher describes it as running from T2–T11.²³ According to Slatter (2003), the intercapital ligament is absent on the first, twelfth, and thirteenth rib and only marginally developed at the eleventh rib pair.²⁴ Wilkens et al. (1996) reported three cases of T9–T10 intervertebral disc herniation and speculated that the intercapital ligament in those dogs may have been abnormal.¹² In the current study, the authors categorized T9–T10 sites with T10–L4 because IVDD is not uncommon at T9–T10 for both small and large dogs. Further investigation of the intercapital ligament is warranted in dogs affected with T1–T9 disc lesions

The T1–T9 lesions also present a challenge with how best to manage them. The best surgical approach to the upper thoracic region is controversial given the anatomy and proximity to the chest. Exposure is more difficult at those disc spaces due to the proximity of the adjacent rib heads, and care must be taken to avoid creating a pneumothorax.¹² The C7–T9 region is an especially difficult area to access surgically. A ventral slot is one option, but that involves a sternotomy and retraction of the aorta and vena cava. A modified dorsolateral hemilaminectomy is the more common approach in veterinary medicine, but the dorsal tendinous raphe and the large amount of musculature make that approach and visualization difficult. In humans, various surgical approaches have been described, including laminectomy, pediculectomy, and thoracoscopy.²⁵ Posterior laminectomy has been associated with poor results and has been largely abandoned; although, a posterior transdural approach has been reported with better direct visualization of the spinal cord with less manipulation.²⁶ The development of posterolateral and anterior approaches has shown a significant reduction in mortality.²⁵ More recently, minimally invasive techniques have come into favor, including fluoroscopic-assisted percutaneous intervertebral disc nucleoplasty, percutaneous laser disc decompression, and thoracoscopy.^27–30

Two of the 21 dogs with upper thoracic disc disease in the current study had paraplegia and moderate to severe spinal compression on MRI. Surgery was not recommended due to concurrent medical problems and/or prognosis. The other 19 of the 21 upper thoracic dogs presented with ataxia and conscious proprioceptive loss to their hind limbs. Sixteen of those 19 dogs had only mild to moderate compression and ≥ 4 sites with compression. Medical management was recommended given the complexity of surgery, moderate clinical signs, and/or number of sites. Surgery was recommended to and pursued by the owners of three dogs in the T1–T9 group. The surgical approach for the first case was a modified dorsolateral laminectomy at T4–T5, including removal of the dorsal spinous processes and dorsal interspinous ligament. That case had a subluxation at the surgery site, required a second procedure to stabilize the area, and was ultimately euthanized for deterioration of his neurologic status. It was decided for future cases to only remove the ventral half of the spinous process and lamina and leave the dorsal half of the spinous process and dorsal interspinous ligaments intact. The other two cases treated surgically in the current study continued to have ataxia, but their neurologic signs were ultimately considered to have stabilized and/or improved slightly. Those results were consistent with previous reports of surgical outcome in older, large-breed dogs with a chronic history of IVDD. In Cudia and Duval’s study (1997), 92% of large-breed dogs with Type II IVDD that were ambulatory prior to surgery had successful outcomes, and 90% of dogs that were nonambulatory with deep pain perception regained the ability to walk; however, almost 40% did not walk normally. They concluded that Type II disc disease had similar success rates as Type I disc disease, but that many dogs will have residual deficits and take longer to recover.⁷ Macias et al. (2002) found that the outcome of dogs with annular extrusions was poor compared with those with nuclear extrusions, regardless of management (22% success rate). However, nearly half of the dogs treated nonsurgically deteriorated within 1 yr and were ultimately euthanized. Although neurologic deficits often remained, no long-term deterioration in neurologic dysfunction was observed in dogs treated surgically. They felt that a “different measure of success” may be required for Type II dogs because the likely irreversible spinal cord atrophy may limit the degree of neurologic recovery.⁶

More studies need to be performed on upper thoracic disc disease to recommend surgical versus medical management. Some might argue that some of those patients should be managed medically, rather than surgically. As shown in the current study, German shepherd dogs with upper thoracic disc disease often have multiple regions affected (66.7%) and/or multiple sites affected within the T1–T9 region, which likely has a more guarded prognosis with surgery. This is further complicated because German shepherd dogs are one of the most common breeds to be diagnosed with degenerative myelopathy, which has a similar clinical presentation to upper thoracic IVDD. As degenerative myelopathy is a histopathologic diagnosis, it can be difficult to differentiate whether some of those cases are clinically affected by IVDD of the upper thoracic region, thoracolumbar region, or degenerative myelopathy.

MRI has become a more widely used diagnostic tool in veterinary medicine over the past decade and is reported to be the best method for early recognition of disc degeneration.^31,32 With the increased use of MRI at referral institutions, it is possible veterinarians will be more likely to find upper thoracic disc lesions. A similar pattern was observed in humans where thoracic disc herniations were thought to be 0.25–0.75% of all disc ruptures, but recent studies with MRI have shown that rate to be 11.1–14.5%.²⁵

Conclusion

Greater than 10% of all large-breed dogs diagnosed with IVDD on MRI had upper thoracic (T1–T9) lesions. Most of those were German shepherd dogs and Labrador retrievers. It is important to note that almost one-third of all German shepherd dogs diagnosed with IVDD were affected at the upper thoracic region. Because both upper thoracic IVDD (T1–T9), lower thoracic IVDD (T9–L4), and degenerative myelopathy may have chronic and progressive signs with similar neuroanatomical localizations, this study demonstrates that diagnostic imaging of all large-breed dogs presenting with thoracolumbar myelopathy should include their upper thoracic spine to rule out IVDD. This is particularly important in shepherd and retriever breeds.