Overriding Spinous Processes ("Kissing Spines") in Horses: Diagnosis, Treatment, and Outcome in 212 Cases

“Kissing spines” is a radiographic diagnosis. It occurs in 39% of the horse population but does not cause problems in all horses. Kissing spines make a horse 3 times more likely to have back pain. Kissing spines are more likely associated with clinical problems in Thoroughbreds, dressage horses, horses 5 years of age or less, and horses with 5 or more vertebrae involved. Thermography is very helpful in suggesting the pathology in the evaluation of horses with back pain. Lateral radiographs of the spinous processes of the thoracolumbar vertebrae are used to confirm and characterize the condition. A therapeutic regimen combining shockwave, mesotherapy, and exercise had the most successful outcome in clinical cases. Author’s address: Anoka Equine Veterinary Services, 16445 70th St. NE, Elk River, MN 55330; e-mail: [email protected].

1. Introduction

Equine back problems are considered a significant cause of alterations of gait and performance in the elite athlete. “Kissing” spinous processes are considered one of the causes of back pain.^1–3  However, there have been no large case studies that characterize, locate, and identify the pain associated with kissing spines. The prevalence of back problems in lameness practice has been reported as 2.2%, and the clinical signs are reported as highly variable.¹  The purposes of this study were to further investigate kissing spines as a cause of poor performance in horses, to characterize any breed or occupational predisposition, determine associated clinical signs, assess which diagnostic tests are effective, and describe treatments and outcomes of the cases.

2. Materials and Methods

The horses in this study were presented to Anoka Equine Veterinary Services from February 1, 2004, to January 31, 2011, for assessment of vague lameness or poor performance, and whose work-up led to a diagnosis of overriding spinous processes (kissing spines) (Group 1). For each horse, signalment, history, presenting complaint, and clinical findings, including thermography (when performed), were recorded. Back pain was determined to be present if the horse showed either marked reaction to digital pressure over the thoracic spine or thoracic epaxial muscles, or alternatively, resisted pressure (stiff, failure to ventroflex or dorsoflex under any stimulus). Thermographic evaluation was performed after the basic clinical examination and before radiography. Radiography of the horse’s back was performed in every case with either standard radiographic equipment or with direct digital radiography. Four serial lateral radiographic projections of the thoracolumbar spine were made from the withers through the lumbar vertebrae. The individual thoracic vertebrae were identified by their position relative to the anticlinal thoracic vertebrae (T15). The diagnosis of kissing spines was made when two or more vertebrae were either touching or overlapping (Fig. 1). The locations and number of kissing vertebrae involved were identified.
 

A protocol of shockwave, mesotherapy, and exercise (SME) was recommended for all cases; however, some clients elected more limited therapy. Extra-corporeal shockwave was performed using an 80-mm focused device^a delivering 2000 impulses at the highest energy setting. The treated area included the affected spinous processes plus the associated epaxial muscles. Mesotherapy is a technique of multiple (720) intradermal injections starting at the withers, then over the back and croup using a solution of anti-inflammatories and lactated Ringers. The exercise program was daily mounted walk and trot in a “long and low” frame to stretch the horse’s back, progressing to trotting over ground poles and cavellettis to encourage further stretching and flexion of the back. Assessment of saddle fit was recommended. Treatment outcome was considered successful if the clinical signs disappeared and the horse was able to resume its previous level of competition.

Horses with kissing spines (Group 1) were compared with two other groups of horses; horses that were diagnosed with back pain but did not have kissing spines (Group 2) and control horses (without back pain) (Group 3). Group 2 horses were used to determine if there were any historical or clinical findings that are pathognomonic for kissing spines. Group 2 horses were seen during the same period as the Group 1 horses. Group 3 were horses presented for prepurchase or pretraining soundness examination that did not have a history of or any clinical signs of back problems. These horses were seen between February 1, 2005, and January 31, 2011, and, after permission from their owners, had a radiographic evaluation of their backs as controls for this study. This latter group was used to determine the prevalence of overriding spinous processes in a non-clinical group. Groups were compared using Chi-square analysis, with significance set at p < 0.05.

Owners and veterinarians evaluated the effectiveness of treatment in eliminating the clinical problem. Treatment response was evaluated as excellent when the horse returned to its regular work, the owner’s complaint was eliminated, and veterinary examination showed elimination of clinical signs. A very good response was achieved when the horse returned to work and veterinary examination showed elimination of signs but the owner still had minor complaints. A good response was recorded when the horse returned to work, veterinary examination indicated relief of pain, but the horse was stiff and the owner had minor complaints. A poor response was recorded if the horse did not return to work despite some improvements in the owner’s complaint and the clinical signs. The therapy was considered of no value if no improvements were seen. Factors such as age, use, and number of vertebrae involved were compared as to their effects on outcome. Final resolution of the case was determined whenever possible by clinical re-evaluation or phone contact with the client.

3. Results

Four thousand four hundred seven horses were evaluated for lameness or poor performance between February 1, 2004, and January 31, 2011. Back pain was identified on clinical examination in 310 horses (7%). Two hundred twelve of the 310 horses (68%) were diagnosed with kissing spines. Seventy horses that had never shown any signs of back pain had their backs radiographed, of which 27 had overriding spinous processes (39%).

Signalment of Affected Horses

Group 1 consisted of a wide variety of breeds but most commonly in Thoroughbreds or Thoroughbred crosses, Quarter Horse types, or Warmbloods (Fig. 2). Ninety of the horses were geldings, 58 were mares, and 4 were stallions. The majority of Group 1 horses were between 6 and 10 years (Fig. 3). The horses were used in a wide variety of disciplines, but dressage was the most common (Fig. 4). In comparing Group 1 and Group 2 horses, a different distribution of breeds, age, and use was recorded (Figs. 2–4). These differences were statistically significant for overrepresentation in the kissing spines group of Thoroughbreds (p < 0.001), horses 5 years of age or less (p < 0.01), and dressage horses (p < 0.05).

History

The affected horses in this study presented for either back soreness or for lameness evaluation because of a change in the horse’s behavior. Wide varieties of abnormalities were reported in the horse’s behavior and could be divided into three categories: ground behavior, riding behavior, and discipline specific behavior (Table 1). A horse may exhibit more than one behavior. Group 1 and Group 2 horses had similar histories and behaviors. [...]