ABSTRACT: In a previous article (VNJ 27 288-292, August 2012) I described how to investigate potential back pain in horses and discussed the difficulties of diagnosis and managing client expectations. In this complementary article, I explain my practical approach to the most commonly treatable conditions, with special reference to imaging, anaesthesia and nursing considerations.

Accurate (and successful) treatment requires an accurate diagnosis. The most commonly treatable conditions we diagnose are ‘kissing spines’, caudal cervical osteoarthritis, thoracolumbar intervertebral joint osteoarthritis, supraspinous ligament desmopathy, sacroiliac joint disease, lumbosacral joint disease and intertransverse joint/lumbosacral nerve pathology.

Impinging dorsal spinous processes – "kissing spines’

Kissing spines is an easy diagnosis to make. It is less easy to make a correct diagnosis of kissing spines. Impingement is a common radiographic finding but this does not mean that it is clinically relevant in all horses showing these radiographic abnormalities.

This is a crucial point – it is the author’s opinion that medical or surgical treatment of impingement can be very successful and that many failures are because there are other issues that have not been taken into account.

Peri-lesional corticosteroids

The simplest method for treating impingement is peri-lesional injection of corticosteroids. A number of methods exist for this, but I prefer to use triamcinolone acetonide (Adcortyl, up to a maximum of 30mg) or methylprednisolone acetate (Depo- Medrone, up to a maximum of 160mg) diluted in sterile saline (see below).

Methylprednisolone is probably preferable because of its longer duration of action when injected in this site, but this, owing to its correspondingly greater withdrawal period, sometimes creates problems for competition horses.

Typically, if three sites are to be treated (i.e. three affected interspinous regions), I will prepare, aseptically, three syringes, each with lOmg triamcinolone or 40mg methylprednisolone diluted in 10ml saline.

The sites for treatment will have been identified radiographically – radiopaque markers placed on the skin are standard practice for my cases, and their positions are then recorded by small clip marks in the hair after the procedure. Almost always, small areas of hair are clipped away at these sites, followed by a brief scrub and injection of a small bleb of local anaesthetic – this makes the procedure a lot less uncomfortable for the horse.

It is impossible to inject between the dorsal spinal processes (DSPs) – the aim is to deposit the medication either side of the lesion(s). I use a 1.5-inch 20 G needle, penetrating the skin in the midline. This is only advanced a few millimetres before it is lifted slightly, tenting the skin, and pushed to the left of midline (but keeping the needle vertical) before being advanced down the side of the interspinous space.

I then inject 5ml of the previously prepared 10ml solution and the needle is withdrawn – but not completely – before advancing it down the right side of the space and injecting the remaining 5ml of anaesthetic. The same technique is used for local anaesthesia during investigations to determine the significance of lesions.

Surgical resection

Surgical resection of affected DSPs is a highly effective method of treatment. It can be applied to cases that respond well – but only transiently – to medication, or those that have blocked out following diagnostic anaesthesia. In some cases, we will operate based only on imaging findings (radiographic and scintigraphic abnormalities) but rarely on the basis of radiographs alone.

The procedure can be carried out under general anaesthesia or, increasingly commonly, under standing sedation and local anaesthesia. The surgeons remove the top parts of the offending DSPs to create clear space between them (Figure 1).

Figure 1: Radiograph of equine spine showing impinging dorsal spinous processes ('kissing spines')

Horses are usually rested for eight weeks before beginning physiotherapy and resume ridden exercise around 12 weeks after the operation.

Caudal cervical osteoarthritis

Injection of the cervical joints is typically indicated for the treatment of osteoarthritis, most commonly affecting the caudal articulations (C5/6 and C6/7). The horse should be sedated reasonably heavily and made to stand square, preferably in stocks. It can be very useful to use a head rest so that the head and neck remain in the same position throughout the procedure.

Typically, I would use ultrasound to assess the joints before clipping the hair and sterilising the site using a liberal application of surgical spirit. The C5/6, C6/7 and C7/T1 joints appear similar and are most easily assessed with a micro-convex transducer, although it is possible to proceed with a high-frequency linear probe.

In most horses the musculature of the scapula/shoulder prevents imaging further back than C6/7, but in some patients the C7/T1 articulation can be imaged. This should be borne in mind – until recently I had assumed that the most caudal joint imaged was always C6/7, but this may not be the case.

If in doubt, a marker placed on the skin during radiographic assessment can be useful. Palpate the transverse processes of the caudal cervical vertebrae and place the transducer above these, just in front of the shoulder musculature, oriented vertically to produce a transverse image of the joints.

Once the joints have been identified, the hair can be clipped so that the site of injection and contact area for the transducer can be prepared. Once clipped and after a short scrub – I then place a small bleb of local anaesthetic at the site of needle placement.

The site for insertion of the needle is above (dorsal to) the contact point for the transducer and is estimated by imagining the course of the needle into the joint. However, the main rule is to start high as the angle of the needle direction should be steep to facilitate entry into the joint space, which is angled sharply from laterodorsal to medioventral (Figure 2).

Figure 2: The joint space is angled sharply from laterodorsal to medioventral

After placing the bleb, the site is prepared thoroughly. Usually, as it is most common to inject C5/6 and C6/7 on both sides of the neck, I prepare both joints on the left side first so that they can be scrubbed together and then injected in one sterile procedure.

As a rule, I inject 5mg triamcinolone acetonide into each joint, drawn up into 1.5ml sterile saline. A separate 3.5-inch 18G spinal needle is used for e
ach joint. The transducer is covered with a size 8 sterile glove – the glove is held open by the operator and a colleague fills a finger or two with sterile ultrasound couplant gel before dropping the transducer into the glove. The micro-convex probe fits easily down one of the fingers of the glove. Sterile couplant gel is applied to the gloved end of the probe and the joint to be injected is visualised.

As I am right-handed, I hold the probe in my left hand and place the needle with my right; for the left side I usually face towards the rear of the horse (and towards the front of the horse for right-sided joints). The transducer is positioned so that the joint space is seen at the side of the image, allowing more space to visualise the needle approaching the joint.

The needle is pushed through the skin bleb and advanced towards the joint (Figure 3). As mentioned previously, the angle of approach should be steep, mimicking the angle of the joint, to maximise the chance of successful entry.

Figure 3: Ultrasound-guided injection into cervical joint space

The most important thing to remember during any ultrasound-guided technique is to be guided by the ultrasound! Take time to ensure that the needle follows the path of the ultrasound beam – failures are often attributable to the path of the needle diverging from the ultrasound indication.

In some cases, the needle advances easily through the joint capsule and a distinct change in resistance indicates successful entry into the joint space. In other cases, the needle encounters bone. If the ultrasound image indicates that the tip of the needle is close to the joint, it is usually possible to ‘walk’ the needle into the joint. Once into the joint space, avoid advancing the needle too far as it is, in theory, possible to enter the vertebral canal.

Removal of the stilette occasionally prompts spontaneous flow of synovial fluid but, more commonly, aspiration with a syringe is necessary to confirm correct placement. Synovial fluid should be aspirated in all cases; although, as with other joints, there are occasions when the needle is correctly placed but fluid cannot be obtained.

I limit myself to two repeat placements of the needle (without coming back out through the skin – if this is necessary a new needle should be used) before, if I am confident about the position, I will inject without obtaining fluid. Of course, it gets easier with practice.

Generally, I do not hold the transducer while the injection is performed, preferring to put the probe to one side and hold the hub of the needle with one hand and the syringe with the other during the injection. The procedure is repeated for the next joint on the same side, before moving equipment to the other side of the horse and starting again.

The horse is treated with a single dose of intravenous non-steroidal anti¬inflammatory (usually phenylbutazone or firocoxib). I usually advise three weeks of restricted turn-out following treatment before any further exercise or physiotherapy.

In many cases, I suggest that the horse is fed from a height for this three-week period – attaching hay nets to fencing or similar. This is based on the observation that most affected horses display difficulty/discomfort when reaching to the floor and on several cases they displayed acute episodes of either severe pain or neurologic signs after grazing. It seems logical, therefore, to limit this (admittedly very normal activity!) for a short time after treatment.

Supraspinous ligament (SSpL) desmopathy

Once again, an easy diagnosis to make, but true SSpL injury as a cause of back pain is uncommon – at least in my case load. When genuine cases are encountered the most logical treatment is a period of rest, with ultrasonographic monitoring at three-month intervals to assess progress (although lesions can persist despite clinical resolution).

Other techniques, such as extracorporeal shock wave therapy (ESWT), seem logical for chronic, unresponsive cases; however, I would advise against its use in acute/subacute injuries. Similarly, intra- or peri-lesional injection of biologies such as platelet-rich plasma would also seem logical, although I have limited experience of their use in this site.

Intervertebral articular disease

Intervertebral articular disease typically affects the caudal thoracic and cranial lumbar joints and a diagnosis is usually made with a combination of scintigraphy, ultrasonography and radiography. Affected joints can be medicated, usually with corticosteroids, via ultrasound- guided injection (USGI).

Once the affected joints have been identified, a curvilinear transducer is placed at right angles to the midline and the joint localised. Two techniques are available but with similar results – the aim is to guide the 3.5-inch 18G spinal needle through the epaxial musculature and deposit the medication into, or close to, the joint.

Whether the probe is positioned close to the midline and the needle is directed from lateral or vice versa, the aim is to inject into the multifidus muscle, close to the joint. The site for injection is clipped and scrubbed before a small skin bleb of local anaesthetic is injected (again, this significantly reduces the discomfort to the patient). The site is then prepared aseptically before the injection is performed.

I would typically use 5mg of triamcinolone acetonide, diluted in 1.5ml of sterile saline, per joint (Figure 4). The needle should be visualised advancing towards the joint and then contacting bone before the steroids are injected.

Figure 4: Needle positioning for intervertebral articular corticosteroid injection

Lumbosacral and sacroiliac disease

Intervertebral disc disease is much less common in horses than in humans, owing, in part, to the differing morphology of the discs between the species (equine discs are comparatively much thinner and lack the liquid centre – the nucleus pulposus – of human discs). However, there is a large disc at the lumbosacral (LS) junction (usually L6/S1, but sometimes L5/L6 if there is sacralisation) that is prone to degeneration.

Assessment of this region is possible using ultrasonographic examination per rectum, at which time the
integrity of the LS disc can be evaluated. In addition, the sacroiliac joints (SIJs) can be viewed (although only the most medial aspects) and the inter-transverse joints.

The inter-transverse joints (ITJs) are interesting – it is likely that disease here may cause discomfort but it is their proximity to the L6 nerve roots that might be more significant. New bone formation around the ITJs will cause compression of the L6 nerve (a major contributor to the sciatic nerve). This region can be treated by ultrasound- guided injection via a cranial approach.

A curvilinear probe is positioned parallel to midline, over the cranial edge of the ilial wing, the aim being to guide the needle under the ilium, towards the lumbosacral region (Figure 5).

Figure 5: Ultrasound-guided approach to lumbosacral region

It is useful to have reference to an anatomical specimen when learning this approach. The close proximity of the ITJs, LSJ and SIJs can be appreciated, as well as the anatomy of the ilial wing – the change in bone thickness of the ilium is a useful guide during USGI (Figure 6).

Figure 6: Anatomical specimen showing the lumbosacral region

The 6-inch 18 G spinal needle effectively forms the hypotenuse of a right-angled triangle and it can be useful to use this principle when assessing the correct position for needle placement (Figure 7).

Figure 7: Annotated photograph showing correct needle placement for lumbosacral treatment

Following clipping, scrubbing and injection of a small volume of local anaesthetic into the skin, the needle is inserted, aiming caudoventrally, 3-6cm off midline and at a position cranial enough to the ilial wing to allow it to pass under the bone, parallel with its surface and toward the lumbosacral region. The position and course of the needle is carefully monitored ultrason- ographically to ensure accurate placement.

Again, typically, I would use lOmg triamcinolone or 40mg methylprednisolone, diluted in 10ml sterile saline, for this treatment. Left and right sides of the horse are always treated together. Contrast studies evaluated with computed tomography have demonstrated excellent dispersal of the injections throughout the targeted regions (Figure 8).

Figure 8: Computed tomography image showing dispersal of corticosteroid following lumbosacral injection


Marcus J Head BVetMed MRCVS

Marcus qualified from the Royal Veterinary College in 1994 and spent a period in mixed practice before an internship at the Animal Health Trust Equine Clinical Unit.

A second internship at The RVC was followed by a move to Rossdale & Partners in 1996, as an assistant dealing with horses in training. In 2004, he moved to the practice's diagnostic centre to work on referral cases full-time and he now sees orthopaedic cases from a wide variety of disciplines – occasionally escaping to attend Thoroughbred sales!

To cite this article use either

DOI: 10.1111/j.2045.0648.2012.00212.x or Veterinary Nursing Journal Vol 27 pp 383-38&


• VOL 27 • October 2012 • Veterinary Nursing Journal