What Causes Kissing Spine and How Is It Treated?

Kissing spine, formally called overriding spinous processes (ORSP), happens when the tall bony projections along a horse’s vertebrae crowd together, narrow the gaps between them to less than four millimeters, and eventually touch or overlap. The condition most often shows up in the mid-to-caudal thoracic spine, roughly the area beneath where a saddle sits, and it is the single most common diagnosis behind primary equine back pain. What makes it tricky is that a huge proportion of horses show the radiographic signs without ever acting sore, so understanding what actually causes the pain and which treatments hold up long-term matters more than the X-ray alone.

Where It Happens and Why That Matters

A horse’s thoracic spine has eighteen vertebrae, each topped by a spinous process that anchors muscles and ligaments responsible for supporting the back. These processes are tallest in the withers region, where they angle backward before gradually becoming shorter and more upright farther down the back. The condition clusters between roughly the thirteenth and eighteenth thoracic vertebrae, with the fifteenth being the most frequently affected. That stretch of spine bears the combined forces of a rider’s weight, the saddle, and the powerful muscles driving the hindquarters forward, so the mechanical load on those interspinous spaces is substantial.

Radiographic and postmortem studies have found anatomical evidence of narrowed or touching spinous processes in up to 92 percent of horses examined, yet many of those horses performed without apparent discomfort.¹ That disconnect between what the X-ray shows and what the horse feels is probably the most important thing to understand about kissing spine. A dramatic radiograph does not automatically equal a painful horse, and a relatively mild-looking image does not rule out significant discomfort.

What Causes It

There is no single, neat cause. Researchers describe kissing spine as multifactorial, meaning several things contribute, and they often stack on top of one another in the same horse.

Conformation plays a large role. Horses with shorter backs, naturally close-set spinous processes, or excessive lordosis (a dipped or swayed topline) start with less room between the bony projections, so even mild changes can close the gap. Poor posture from weak core and epaxial muscles can have a similar effect, because the back sags and the tops of the spinous processes drift toward each other.

Ridden work and equipment add mechanical forces that push the spine into extension. A study examining the effects of a saddle and rider weight on back movement found that carrying a saddle with weight caused overall extension of the back at the walk and trot, meaning the vertebrae shifted in a way that narrows the interspinous spaces. The researchers noted this may contribute to soft tissue injuries and kissing spine syndrome.² Ill-fitting saddles that concentrate pressure on a small area or restrict the horse’s ability to round its back likely amplify the problem.

Breed susceptibility is real but not straightforward. A genomic study found that Thoroughbreds, Warmbloods, and stock-type breeds including Quarter Horses and Paint Horses were at increased risk for developing clinical signs. Within that study population, however, breed group, age, height, and sex did not show a statistically significant independent effect, suggesting the picture is more about individual anatomy and genetics than about any one breed label.³ The same research identified specific genetic markers on chromosome 25 that correlated with increasing severity, pointing toward a heritable component that breeders may eventually be able to screen for.³

Young horses that begin intense training before their musculoskeletal system is fully developed may also be at higher risk. The muscles that stabilize the spine need time and progressive conditioning to handle the demands of collection, jumping, and carrying a rider. Without that conditioning, the skeleton absorbs forces the muscles should be dampening.

How Horses Show Pain

One reason kissing spine goes unrecognized is that the signs are often vague and overlap with dozens of other conditions. A horse might resist being saddled, flinch or dip when you press along the back, or show stiffness when asked to bend laterally. Under saddle the behavior can look like attitude problems: bucking, refusing fences, reluctance to go forward, swishing the tail, or hollowing the back instead of rounding through the topline. Some horses simply plateau in performance or gradually become less willing to work without any dramatic incident.

Lameness elsewhere can complicate the picture. Horses with sore hocks, stifles, or feet alter the way they move, and compensatory changes in gait mechanics can strain the back. Research on horses with facet joint arthritis in the thoracolumbar spine found that clinical features were significantly different depending on whether kissing spine was also present, and that the severity of signs related more to the presence of additional bony abnormalities than to any single lesion.⁴ In other words, kissing spine rarely exists in isolation, and failing to address concurrent problems is one reason some horses do not improve after treatment.

Getting the Diagnosis Right

Radiographs are the starting point. A lateral view of the thoracolumbar spine can show whether the spinous processes are close, touching, or overlapping, and several grading systems exist to rank severity. But the correlation between radiographic grade and clinical pain is unreliable. Many horses with dramatic narrowing perform without issue, while some with only mild changes are clearly uncomfortable.⁵ No definitive cut-off grade separates a sound horse from one with clinically meaningful disease, and researchers acknowledge such a line may never exist because pain tolerance varies between individuals.

Ultrasound adds useful information about the soft tissues surrounding those bony landmarks, including the interspinous and supraspinous ligaments, the epaxial muscles, and the thoracolumbar fascia. Nuclear scintigraphy (bone scan) can highlight areas of active inflammation and bone remodeling, but abnormal uptake has also been documented in horses that show no clinical signs at all.⁵

The most accurate approach combines imaging modalities with clinical examination and, critically, diagnostic local anesthesia. A study comparing radiology and scintigraphy found that neither alone was a reliable predictor of pain. The combination of both gave a more accurate picture, but the researchers stressed that injecting local anesthetic into the interspinous spaces to see whether the horse’s behavior improved was crucial for confirming the diagnosis.⁶ Even that step has limits: the neural anatomy of the equine back is complex and not fully mapped, and the wide variation in techniques used for localizing back pain may explain why some horses improve to local anesthesia while others do not.⁷

Non-Surgical Treatment

Most veterinarians start conservatively. A survey of equine practitioners in the United States found that about 72 percent recommended surgery only after non-surgical approaches had failed, and roughly 15 percent never recommended surgery at all. Encouragingly, 82 percent of respondents reported at least some improvement in clinical signs of primary back pain with rehabilitation alone.⁸

Corticosteroid injections are a mainstay. The steroid is typically placed directly into the narrowed interspinous spaces under radiographic guidance, where it reduces local inflammation and pain. This serves a dual purpose: it can confirm that the affected spaces are the pain source and provide weeks to months of relief.⁹ The downside is that the relief is often temporary. In one controlled study, 89 percent of horses treated with methylprednisolone initially resolved their signs of back pain, but the pain eventually recurred in half of those cases.¹⁰

Mesotherapy, shockwave therapy, and acupuncture are also used with varying degrees of evidence behind them. These modalities may help manage discomfort and improve muscle function, but they are generally considered part of a broader management plan rather than standalone fixes. Anti-inflammatory medications such as phenylbutazone can take the edge off during flare-ups and allow the horse to move more comfortably during rehabilitation exercises.

Surgical Options

When conservative management fails to hold, surgery offers two main approaches, and both can be performed with the horse standing under sedation rather than under general anesthesia.

Interspinous ligament desmotomy (ISLD) involves cutting the ligament that runs between adjacent spinous processes. By releasing that taut band, the procedure allows the spinous processes to separate slightly, reducing the mechanical contact that generates pain. In the controlled study mentioned earlier, 95 percent of surgically treated horses resolved their signs of back pain, and none of them experienced recurrence during the follow-up period, compared with recurring pain in half the medically treated group. Horses that underwent ISLD were 24 times more likely to experience long-term resolution of back pain than horses that received corticosteroid injections alone.¹⁰ Follow-up radiographs in that study confirmed that the interspinous spaces actually widened after surgery.

A longer-term follow-up of 71 horses treated with ISLD painted a more nuanced picture. About 91 percent returned to some level of performance, but only about 53 percent reached an equivalent or higher level than before the problem developed.¹¹ That gap between “returned to work” and “returned to the same level” is worth understanding. Getting rid of the pain source does not automatically rebuild the muscle, fitness, and confidence a horse lost during the months or years it was uncomfortable.

Subtotal ostectomy takes a more aggressive approach by removing part of the offending spinous processes. A cranial wedge technique applied to 25 horses resulted in roughly 79 percent returning to full work and another 18 percent showing improvement. Surgery time was short, often around 30 minutes, with no reported complications. The cosmetic outcome was described as excellent in over 80 percent of cases.¹² An earlier report confirmed that the affected portions of dorsal spinous processes could be removed safely in standing horses with minimal hemorrhage.¹³

Why Rehabilitation Makes or Breaks the Outcome

Surgery or injections address the immediate pain, but without a structured rehabilitation program the underlying weakness and movement patterns that contributed to the problem tend to persist. The evidence on this is fairly clear. A study surveying outcomes in Australian horses found that completing a rehabilitation program gave horses 10 times the odds of returning to their previous performance level, and having the saddle professionally fitted as part of post-treatment management also improved outcomes.¹⁴ A separate Australian survey identified five factors associated with higher odds of returning to ridden work: training at an intermediate to high level before the diagnosis, exercising more than 90 minutes per week, surgical management, completing a rehabilitation program, and having the saddle fitted afterward. All horses in the surgical group also completed rehabilitation, and 84 percent returned to ridden work within a median of 14 weeks.¹⁵

Rehabilitation typically progresses through phases. Early work focuses on encouraging the horse to stretch and lift through its back, using ground-based exercises such as carrot stretches (baited stretches), walking over raised poles, and lunging in a frame that promotes topline engagement. Turnout is considered essential during this phase because free movement on pasture encourages natural posture and prevents stiffness. As strength improves, ridden work is reintroduced at a walk, then trot, with an emphasis on encouraging the horse to work “long and low” before gradually adding collection. Hill work, transitions, and lateral exercises are layered in to build the multifidus and longissimus muscles that support the spine.

Saddle fit deserves attention at every stage. A saddle that bridges (concentrating pressure at the front and back while lifting off the middle), pinches at the withers, or sits too far back can recreate the very forces that caused the problem. Having the saddle checked by a qualified fitter after rehabilitation, once the horse’s topline has changed shape, can prevent a relapse.

The Genetics Question

Whether kissing spine is heritable is a question breeders are starting to take seriously. The genomic study that identified markers on chromosome 25 found that each copy of a specific risk allele correlated with an increase of about one grade on the kissing spine severity scale.³ That is a meaningful effect for a single genetic locus in a condition influenced by many factors. A broader study of musculoskeletal conformation in Thoroughbred yearlings found that all conformational traits examined showed considerable genetic influence, with heritability estimates ranging from 0.16 to 1.00.¹⁶ While that study was not specific to kissing spine, it underscores that the structural features predisposing a horse to the condition, such as back length, vertebral spacing, and overall conformation, are at least partly inherited.

Genetic testing for kissing spine susceptibility is not yet standard practice in any breed registry, but the identification of candidate genomic regions suggests it could become feasible. For now, breeders can pay attention to whether a stallion or mare’s offspring repeatedly present with the problem and factor that information into breeding decisions alongside other health and performance data.

Baastrup’s Disease in Humans

Horses are not the only species that gets kissing spines. The human equivalent is Baastrup’s disease, named after the Danish radiologist who described it in 1933. In people, the condition involves close approximation of adjacent lumbar spinous processes, typically driven by degenerative changes and excessive lordosis. The mechanism is similar in principle: the spinous processes crowd together, the interspinous ligament degenerates, an adventitious bursa forms in the space, and repetitive shearing movements cause inflammation, sclerosis, and sometimes cyst formation.¹⁷

In humans, Baastrup’s disease tends to show up in older adults and is associated with degenerative disc disease, spondylosis, and changes in spinal curvature. It is often discovered incidentally on imaging and, much like in horses, the presence of close spinous processes on an MRI or X-ray does not always mean the patient is in pain. Treatment in people follows a similar ladder: anti-inflammatory medications, steroid injections into the interspinous space, and surgery (removal of the offending bone) reserved for refractory cases. The parallel between the two species is a useful reminder that the fundamental biomechanics of a vertebral column under load produce similar failure modes regardless of whether the patient walks on two legs or four.

Concurrent Back Pathology

One of the pitfalls in managing kissing spine is treating it as the sole problem when it is really one piece of a larger puzzle. Facet joint arthritis in the thoracolumbar spine commonly coexists with overriding spinous processes, and research has shown that the severity of clinical signs in horses with facet joint disease depended more on the presence of concurrent conditions like kissing spine than on the number of arthritic joints or the grade of those lesions.⁴ Sacroiliac dysfunction, hind-limb lameness, and poorly balanced feet can all feed into the same cycle of abnormal back movement and pain. Veterinarians who treat the back without evaluating the whole horse frequently see disappointing results because the root cause continues to load the spine in harmful ways.

This is one area where the diagnostic limitations of ambulatory practice become relevant. The deeper structures of the thoracolumbar spine, including the facet joints and vertebral bodies, require high-output radiographic equipment typically found only in hospital settings. A vet working out of a truck with portable X-ray can see the spinous processes clearly but may miss pathology underneath them, leading to a partial diagnosis.⁵ If a horse does not respond as expected to kissing spine treatment, a hospital-based workup that includes advanced imaging of the full spine is a reasonable next step.