Ex Vivo Assessment of an Ultrasound‐Guided Injection Technique of the Navicular Bursa in the Horse

Synovitis of the navicular bursa is common in performance horses. The objective of this study was to describe an ultrasound‐guided technique to inject a distended navicular bursa and to evaluate its feasibility for use by a clinician not trained in the technique. Twenty distal limbs of horses of various breeds and sizes were used. To produce synovial distension, the navicular bursa of each limb was injected with contrast medium using a lateral approach and radiography was performed to confirm that the contrast medium was distending the bursa. The digit was positioned with the distal interphalangeal joint in hyperextension. A microconvex ultrasound probe was placed in the hollow of the pastern, palmar to the middle phalanx and the region was assessed in a transverse plane slightly oblique to the horizontal plane. The ultrasound probe was rotated to visualize both the lateral and medial recesses and to select which side was more distended to inject. A 21G 0.8 × 50 mm needle was inserted abaxially to the probe in the plane of the ultrasound beam into the proximal recess of this navicular bursa and a methylene blue solution was injected. Following injection, dissection was performed to assess whether the navicular bursa had been successfully injected. This ultrasound‐guided technique was reliably performed with a success rate of 68%. The success of injection is influenced by hyperextension of the foot, quality of ultrasound images and degree of distension of the bursa.     

Ultrasound-guided coxofemoral arthrocentesis in horses

REASON FOR PERFORMING STUDY: Coxofemoral joint pain is probably underestimated due to difficulties in identifying hip pain. The deep location of the joint and proximity of the sciatic nerve make arthrocentesis based on external landmarks a difficult and potentially risky procedure in mature horses. OBJECTIVES: To describe an ultrasound-guided injection technique of the coxofemoral joint in standing horses and to evaluate its accuracy and potential difficulties/complications. METHODS: Nine mature horses had both pelvic areas prepared for sterile ultrasound examination (3.5 MHz curvilinear probe). Coxofemoral joints were located and penetrated at their craniodorsolateral aspect under ultrasonographic guidance and injected with sterile contrast medium. A standing ventrodorsal radiographic view of each hemipelvis centred on the hip was obtained for each horse to assess the injection site. Horses were evaluated for 10 days following injection for possible complications. RESULTS: Intra-articular injection was successful in all 18 joints. The procedure was well tolerated by horses under minimal restraint. Mean +/- s.d. needle repositionings required before accurate placement was 1.5 +/- 1.3 per joint. Once the needle was in the joint, synovial fluid was obtained in 7/18 joints. Minimal periarticular contrast medium was detected in 2/18 joints. Mean +/- s.d. ultrasonographic examination time required for coxofemoral localisation, accurate needle positioning and injection was 4.3 +/- 2.1 min. No complications were observed in the 10 days following injection. CONCLUSION: The ultrasound-guided coxofemoral arthrocentesis is an accurate, reliable and safe technique that offers a real time evaluation of needle introduction into the deep and narrow coxofemoral joint space. POTENTIAL RELEVANCE: Although this technique remains to be tested on clinical cases, it is a promising tool to facilitate diagnosis of coxofemoral pain, septic arthritis or administration of intra-articular medication.     

Ultrasound‐guided retrobulbar nerve block in horses: a cadaveric study

Objective To develop an ultrasound‐guided technique for retrobulbar nerve block in horses, and to compare the distribution of three different volumes of injected contrast medium (CM) (4, 8 and 12 mL), with the hypothesis that successful placement of the needle within the retractor bulbi muscle cone would lead to the most effective dispersal of CM towards the nerves leaving the orbital fissure. Study design Prospective experimental cadaver study. Animals Twenty equine cadavers. Methods Ultrasound‐guided retrobulbar injections were performed in 40 cadaver orbits. Ultrasound visualization of needle placement within the retractor bulbi muscle cone and spread of injected CM towards the orbital fissure were scored. Needle position and destination of CM were then assessed using computerized tomography (CT), and comparisons performed between ultrasonographic visualization of orbital structures and success rate of injections (intraconal needle placement, CM reaching the orbital fissure). Results Higher scores for ultrasound visualization resulted in a higher success rate for intraconal CM injection, as documented on the CT images. Successful intraconal placement of the needle (22/34 orbits) resulted in CM always reaching the orbital fissure. CM also reached the orbital fissure in six orbits where needle placement was extraconal. With 4, 8 and 12 mL CM, the orbital fissure was reached in 16/34, 23/34 and 28/34 injections, respectively. Conclusion and clinical relevance The present study demonstrates the use of ultrasound for visualization of anatomical structures and needle placement during retrobulbar injections in equine orbits. However, this approach needs to be repeated in controlled clinical trials to assess practicability and effectiveness in clinical practice.     

Femoral Head Osteotomy in Horses and Cattle

Femoral head ostectomy was performed in six horses, three ponies, and four cattle for treatment of fractures of the femoral capital physis, coxofemoral luxation, fractured acetabulum, or severe degenerative joint disease. The procedures were performed via a cranial approach that did not involve osteotomy of the greater trochanter. A dorsal approach for femoral head ostectomy via osteotomy of the greater trochanter was evaluated in three healthy adult ponies. Three animals (2 ponies, 1 calf) were euthanatized within a month and one horse was euthanatized at year 2 due to postoperative complications. Nine animals were discharged to owners and six of them fulfilled their intended functions of breeding, milking, and being kept as companions. One horse was lost to follow-up and two horses died of causes unrelated to the surgery. All surviving animals had a residual lameness that was described by owners as mild to moderate. None of the horses were used as riding animals. The mean age and weight of 10 animals that regained weight-bearing locomotion was 3.1 months and 84 kg; for three unsuccessful cases it was 34 months and 174 kg. We concluded that femoral head ostectomy was a viable salvage procedure for large animals with capital femoral physeal fracture, chronic coxofemoral luxation, or acetabular fracture. Surgical prognosis appeared to be favorable in young cattle and fair in young horses or ponies weighing less than 100 kg. Osteotomy of the greater trochanter resulted in superior exposure of the intact coxofemoral joint and allowed easier, less traumatic surgical luxation of the joint to facilitate femoral head ostectomy.     

COMPARISON OF ULTRASOUND‐GUIDED VS. STANDARD LANDMARK TECHNIQUES FOR TRAINING NOVICE OPERATORS IN PLACING NEEDLES INTO THE LUMBAR SUBARACHNOID SPACE OF CANINE CADAVERS

The standard technique for placing a needle into the canine lumbar subarachnoid space is primarily based on palpation of anatomic landmarks and use of probing movements of the needle, however, this technique can be challenging for novice operators. The aim of the current observational, prospective, ex vivo, feasibility study was to compare ultrasound‐guided vs. standard anatomic landmark approaches for novices performing needle placement into the lumbar subarachnoid space using dog cadavers. Eight experienced operators validated the canine cadaver model as usable for training landmark and ultrasound‐guided needle placement into the lumbar subarachnoid space based on realistic anatomy and tissue consistency. With informed consent, 67 final year veterinary students were prospectively enrolled in the study. Students had no prior experience in needle placement into the lumbar subarachnoid space or use of ultrasound. Each student received a short theoretical training about each technique before the trial and then attempted blind landmark‐guided and ultrasound‐guided techniques on randomized canine cadavers. After having performed both procedures, the operators completed a self‐evaluation questionnaire about their performance and self‐confidence. Total success rates for students were 48% and 77% for the landmark‐ and ultrasound‐guided techniques, respectively. Ultrasound guidance significantly increased total success rate when compared to the landmark‐guided technique and significantly reduced the number of attempts. With ultrasound guidance self‐confidence was improved, without bringing any significant change in duration of the needle placement procedure. Findings indicated that use of ultrasound guidance and cadavers are feasible methods for training novice operators in needle placement into the canine lumbar subarachnoid space.     

Validation of a cranial injection technique of the medial femorotibial joint in the horse

A cranial technique for entry of the medial femorotibial (MFT) joint in the horse was validated. It was hypothesised that the frequency of correct placement into the MFT joint using the described cranial technique would be similar to using a standard medial approach. Twenty‐four limbs from 15 horses were injected with a latex mixture. Limbs with stifle joint effusion and/or known stifle pathology were grouped separately to those without. When effusion/pathology was not present the medial technique (93%) was almost twice as successful as the cranial technique (53%), including when extremes of size were removed from the population. With the presence of femoropatellar or medial femorotibial effusion, the cranial technique (20%) was 5 times less successful than using the medial technique (100%). When extremes of size were removed from the population of horses with effusion, the success of the cranial technique was marginally improved (33%). When an ‘average size’ of horse is used with or without joint effusion, the cranial technique is successful in entering the MFT joint in approximately 50% of cases, and may hence be of use in a fractious animal without sedation, a stallion, or in the presence of a wound/skin infection at the site of the medial needle entrance.     

Arthroscopy of the Coxofemoral Joint of Foals

An arthroscopic procedure for examination of the coxofemoral joint was developed in nine foals (four cadavers, five anesthetized) to determine if access was sufficient for evaluation and surgical treatment of intra-articular lesions. The joint was distended and the arthroscope inserted through the notch (incisura trochanterica) between the cranial and caudal parts of the greater trochanter. This portal allowed examination of the cranial, lateral, and caudal aspects of the joint. Mechanical distraction of the joint through an instrument portal located 2 to 4 cm cranial and 1 to 2 cm ventral to the arthroscope portal allowed examination of the ligament of the head of the femur, the femoral head, and articular and nonarticular surfaces of the acetabulum. Adduction and rotation of the limb improved visualization of the craniomedial and caudomedial portions of the femoral head. Traction applied to the distal limb allowed visualization of the same structures that were observed when mechanical distraction was used. Traction also created space for placement of surgical instruments into the joint through the instrument portal. Access to most regions of the joint was adequate, but access to the caudal and medial aspects of the joint was limited. Three foals were killed while they were anesthetized, and their coxofemoral joints were dissected. Two foals were allowed to recover from anesthesia and were observed for 30 days after surgery. One foal was mildly lame for 2 days after surgery. The other foal was not lame after surgery. The incisions healed, and the coxofemoral joints were radiographically normal by postoperative day 30.     

Ultrasound‐guided arthrocentesis of the temporomandibular joint in healthy adult horses is equivalent to blind arthrocentesis

Equine temporomandibular joint (TMJ) diseases are increasingly recognized as a problem for the well‐being and performance of horses. Diagnosis is confounded by overlap of clinical signs associated with pathology of the oral cavity, poll, and cervical vertebrae. Arthrocentesis for intra‐articular analgesia, sampling of synovial fluid, and medication is needed for diagnostic and therapeutic purposes. Ultrasound features of the normal TMJ and a blind arthrocentesis technique have been described, but a systematic approach to ultrasound‐guided (USG) arthrocentesis has not been reported. Ultrasound guidance allows visualization of the TMJ that may prove beneficial in cases when pathology, abnormal anatomy, or clinician inexperience make blind arthrocentesis difficult. We hypothesized that USG arthrocentesis would result in fewer needle repositions than blind arthrocentesis. We also aimed to assess synovial fluid parameters for normal equine TMJs. A prospective randomized method comparison with crossover experimental design compared the number of needle positionings required for accurate injection of the TMJ using each technique. Arthrocentesis technique and operator experience were tested using cadavers and two operators. Injection success was confirmed using CT. The radiologist then applied both techniques in normal live horses. No statistically significant difference was noted between arthrocentesis techniques or operators (P > .05). No complications were observed in live horses following either technique. Synovial fluid parameters were largely within the normal range expected for other synovial joints. Either blind or USG arthrocentesis of the equine TMJ can be performed with minimal prior operator experience. Ultrasound‐guided arthrocentesis is an alternative method and can be considered in cases with altered anatomy.     

Evaluation of a tenoscopic approach for desmotomy of the accessory ligament of the deep digital flexor tendon in horses

Objective: To develop a tenoscopic approach for desmotomy of the accessory ligament of the deep digital flexor tendon (AL‐DDFT) in horses. Study Design: Experimental. Animals: Cadaveric forelimbs (n=10) and 4 forelimbs from 2 horses anesthetized for terminal teaching procedures, and 12 forelimbs of 6 experimental horses. Methods: Saline distention of the carpal flexor sheath facilitated insertion of an arthroscope into the distal medial aspect of the sheath between the AL‐DDFT and deep digital flexor tendon (DDFT). Location of an instrument portal on the lateral aspect of the metacarpus was identified with a needle. The lateral aspect of the AL‐DDFT was transected and the arthroscope and instrument were switched to transect the remaining fibers on the medial aspect. Cadaveric specimens were dissected for evaluation. Experimental horses were monitored for 30 days postoperatively. Results: Minor complications including incomplete division of the AL‐DDFT and shallow incision into the suspensory ligament were observed in some cadaver specimens. The AL‐DDFT was completely transected in all experimental horses with no suspensory ligament damage. Mean±SD surgical time (incision to skin closure) was 28.3±11.8 minutes. On ultrasonographic examination, transection of the AL‐DDFT was complete in all experimental horses. Minor DDFT fiber disruption was noted in 1 limb during ultrasound examination at day 30. Conclusions: A tenoscopic approach through the carpal flexor sheath provided adequate access for desmotomy of the AL‐DDFT.     

Use of Intraoperative Ultrasonography in Six Horses

Intraoperative ultrasonography was used in six horses to aid localization and removal of bone fragments (3 horses) and foreign bodies (3 horses). The ultrasound transducer was enclosed in a sterile sleeve containing sterile aqueous gel and the examination was performed after aseptic preparation of the surgical site. Using ultrasound guidance a needle was placed in contact with the bone fragment or foreign body and an incision was made along the path of the needle to expose and remove the object. This technique resulted in decreased operative time and minimal tissue dissection.     

ULTRASOUND‐GUIDED CERVICAL CENTESIS TO OBTAIN CEREBROSPINAL FLUID IN THE STANDING HORSE

Horses with intracranial lesions and severe ataxia are not good anesthesia candidates; however, only one method to obtain cerebrospinal fluid (CSF) from the cervical region in a standing horse has been reported. This method is not performed routinely due to the difficulty for sample acquisition. Our hypothesis is that standing cervical centesis can be performed in horses without complication. Ultrasound‐guided centesis of the CSF between C1 and C2 in 11 clinically normal horses and two horses with neurologic signs were performed. Horses were sedated and ultrasound was used to identify the subarachnoid space and spinal cord between C1 and C2. With ultrasound guidance, a needle was introduced into the dorsal aspect of the subarachnoid space using a lateral approach. Ten milliliters of CSF was obtained and analyzed. Two normal horses in this study had moderate red blood cell contamination in the CSF (940 and 612 RBC/μl). One horse had 11 RBC/μl and the remaining horses had <4 RBC/μl. The total procedure time was approximately 2 min. No reaction was observed and no complications were detected up to 48 h after the procedure. Ultrasound‐guided centesis between C1 and C2 is a rapid procedure that causes minimal to no reaction in standing, sedated horses used in this study. The use of ultrasound to guide a standing C1–2 centesis of the subarachnoid space provides an additional route to obtain CSF for analysis in the equine patient.     

Osteochondral dysplasia of the coxofemoral joints in a Friesian foal: Clinical findings and methods of diagnosis

Diseases of the coxofemoral (hip) joint are infrequently diagnosed in horses. Most cases are presented as an unilateral condition and usually are of traumatic origin. This case report describes a Friesian foal with a clinically obvious thoracolumbar kyphosis, combined with a weight‐shifting stance and a shortened stride of both hindlimbs. General clinical and lameness examinations, computed tomographic examination of the pelvis, and macroscopic and histopathological examinations of the coxofemoral joints were performed. This revealed a final, phenotypical diagnosis of a primary osteochondral dysplasia of both coxofemoral joints with secondary osteoarthritis. Similar to the occurrence of this condition in other species and considering the small genetic basis of the Friesian horse breed, a genetic predisposing factor is suspected to play a key role in the developing mechanism of dysplastic coxofemoral joint disease in horses as illustrated with this case. Computed tomography scanning appears to be a useful imaging technique in the detection of coxofemoral joint disease in small horses and foals.     

Ultrasound-guided periarticular injections of the sacroiliac region in horses: a cadaveric study

REASON FOR PERFORMING STUDY: The traditional techniques for injection of the sacroiliac (SI) region are based on external landmarks. Because of the depth of the SI joint and pathological modifications, SI injections are sometimes challenging in horses. HYPOTHESIS: An ultrasound-guided techniques would allow placement of the needle without depending on external landmarks. METHODS: Fourteen pelvic specimens were isolated from mature horses. A 20 cm bent spinal needle was positioned with ultrasonographic guidance under both iliac wings aiming for SI joints using 5 approaches: cranial, craniomedial, medial and 2 caudal approaches. The length of needle inserted was recorded and 2 ml of latex injected. The distance from latex to the closest sacral articular margin, the contact between latex and the SI interosseous ligament or the contact with the neurovascular structures emerging from the greater sciatic foramen were recorded at the time of dissection. RESULTS: Latex was identified under the iliac wing in all injections but one. The distance from the latex to the closest sacral articular margin was significantly shorter (P = 0.02) for the 2 caudal approaches compared to the cranial, craniomedial and medial approaches. Contact between latex and the SI interosseous ligament was significantly more frequent (P = 0.01) with the cranial, craniomedial and medial approaches (38/73) compared to the caudal approaches (1/24). Contact between latex and the neurovascular structures was significantly less frequent (P = 0.005) for the cranial and craniomedial approaches (0/47) compared to the medial and caudal approaches (8/60). Four erratic injections were encountered. CONCLUSIONS: Ultrasonographic guidance allowed the needle to engage under the iliac wing without being dependent on external landmarks. The caudal approaches allowed deposition of liquid extremely close to the SI joint although retroperitoneal injections occurred. CLINICAL RELEVANCE: Each approach has advantages/drawbacks that could be favoured for selected purposes, but additional work is required to evaluate them on clinical cases.     

ULTRASOUND CRITERIA AND GUIDED FINE‐NEEDLE ASPIRATION DIAGNOSTIC YIELDS IN SMALL ANIMAL PERITONEAL,MESENTERIC AND OMENTAL DISEASE

Peritoneal, mesenteric, and omental diseases are important causes of morbidity and mortality in humans and animals, although information in the veterinary literature is limited. The purposes of this retrospective study were to determine whether objectively applied ultrasound interpretive criteria are statistically useful in differentiating among cytologically defined normal, inflammatory, and neoplastic peritoneal conditions in dogs and cats. A second goal was to determine the cytologically interpretable yield on ultrasound‐guided, fine‐needle sampling of peritoneal, mesenteric, or omental structures. Sonographic criteria agreed upon by the authors were retrospectively and independently applied by two radiologists to the available ultrasound images without knowledge of the cytologic diagnosis and statistically compared to the ultrasound‐guided, fine‐needle aspiration cytologic interpretations. A total of 72 dogs and 49 cats with abdominal peritoneal, mesenteric, or omental (peritoneal) surface or effusive disease and 17 dogs and 3 cats with no cytologic evidence of inflammation or neoplasia were included. The optimized, ultrasound criteria‐based statistical model created independently for each radiologist yielded an equation‐based diagnostic category placement accuracy of 63.2–69.9% across the two involved radiologists. Regional organ‐associated masses or nodules as well as aggregated bowel and peritoneal thickening were more associated with peritoneal neoplasia whereas localized, severely complex fluid collections were more associated with inflammatory peritoneal disease. The cytologically interpretable yield for ultrasound‐guided fine‐needle sampling was 72.3% with no difference between species, making this a worthwhile clinical procedure.     

Treatment of a coxofemoral luxation in a pony using a prosthetic capsule technique

Objective: To report open reduction and joint stabilization using a prosthetic capsule technique for treatment of a craniodorsal coxofemoral luxation in a lightweight pony. Study Design: Clinical report. Animals: Pony stallion (8 years old, 113 kg). Methods: A craniodorsal approach to the coxofemoral joint was performed under general anesthesia. After open reduction, a prosthetic capsule technique was used. Four 3.5 mm self‐tapping cortical anchor screws were implanted: 2 screws were inserted above the cranial aspect of the dorsal acetabular rim, and 2 screws were inserted into the femoral neck. Subsequently, a #5 USP ultra‐high molecular weight polyethylene suture was placed through these anchor screws in figure‐8 fashion to serve as a prosthetic capsule. Routine closure and a medial patellar desmotomy completed the surgical procedure. Postoperatively, the pony was cross tied for 4 weeks to prevent it from lying down. Results: Neither short‐term nor long‐term complications occurred and the pony was sound at the follow‐up examination performed 15 months postoperatively. Conclusions: A prosthetic capsule technique can be used successfully to stabilize the coxofemoral joint after open reduction of a craniodorsal luxation in a lightweight pony.     

Radiographic diagnosis of craniodorsal coxofemoral luxation in standing equids

Radiographic evaluation of the pelvis in standing horses has been used to diagnose fractures of the pelvis, head and greater trochanter of the femur, and luxations of the coxofemoral joint. Coxofemoral luxation injuries are more common in smaller horse breeds and donkeys, but, due to their size, the standing ventrodorsal projection is not possible, as there is insufficient space to place the radiography equipment under the animal's abdomen. The objective of the study was to report the advantages and limitations of the use of an oblique radiographic projection to diagnose unilateral craniodorsal coxofemoral luxation in 3 ponies and a donkey performed with the animals standing under light sedation. All cases had severe unilateral hindlimb lameness and asymmetry of the gluteal region; 2 also had concurrent intermittent upward fixation of the patella. A standing dorsolateral 20–30° ventral oblique radiograph of the affected coxofemoral joint was performed in all cases to obtain a definitive diagnosis. Radiography of the coxofemoral joint in standing ponies and donkeys can be carried out to identify craniodorsal coxofemoral luxation avoiding the need for general anaesthesia.     

Comparison of medial and lateral ultrasound-guided approaches for periarticular injection of the thoracolumbar intervertebral facet joints in horses

Objective: To compare ultrasound‐guided lateral and medial approaches for periarticular injections of the thoracolumbar intervertebral facet joints (IFJ). Study Design: Experimental cadaveric study. Sample Population: Adult equine cadavers (n=4). Methods: IFJ (T12–T13 to L5–L6) were identified by ultrasound (transducer perpendicular to the spine axis) and insertion of a 13 cm, 18 g spinal needle monitored until bone contact using medial (right side) and lateral (left side) approaches. Number of needle insertions at each site, needle repositioning, and insertion depth were recorded. On bone contact 2 mL latex was injected. Intraarticular deposition, distance of latex from the closest articular margin, and presence of latex in the multifidus muscle were established by dissection. Results: Of 96 attempts, only 1 site require reinsertion of the needle; however, 46% of the injections required needle repositioning. Mean ± SD insertion depth was 8.5 ± 1.1 cm. Most injections (86%) were intraarticular and 96% were at or within 0.5 cm of the closest articular margin. Needle insertion relative to the transducer (lateral, medial) had no effect on the distance from the latex to the closest articular margin and all injections were performed into the multifidus muscle. Conclusions: Ultrasound‐guidance facilitated accurate periarticular injection of thoracolumbar IFJ irrespective of using a medial or lateral approach.     

FEASIBILITY FOR ULTRASOUND‐GUIDED INJECTION OF THE COLLATERAL LIGAMENTS OF THE DISTAL INTERPHALANGEAL JOINT IN HORSES

Desmitis of the collateral ligament of the distal interphalangeal joint is a cause of lameness in performance horses. The objective of this prospective, experimental, ex vivo feasibility study was to evaluate the success of ultrasound‐guided injection of the collateral ligaments of the distal interphalangeal joint in the equine forelimb. Seventy‐six ultrasound‐guided dye injections of the collateral ligament of the distal interphalangeal joint were performed on horses’ cadaver limbs. The hooves were sectioned transversely to verify the location of the dye relative to the collateral ligaments and surrounding structures. Evaluations of transverse sections were performed independently by two experienced observers. A scoring system was used to assess injection of the collateral ligament of the distal interphalangeal joint at the proximal, middle, and distal aspect over the length of the ligament. The collateral ligament was injected at any point in 97.4% of cases. The ligament was injected over the entire scored length in 43.2% of cases (32/74), over two scored length areas in 45.9% of cases (34/74), and in one area in 10.8% of cases (8/74). The distal interphalangeal joint and the common digital extensor tendon were also injected in 81.6% (62/76) and 43.4% (33/76) of the cases, respectively. Use of the ultrasound had a positive and negative predictive value of 98% and 9%, respectively. In this study, ultrasound guidance was useful for confirming injection of the collateral ligament of the distal interphalangeal joint but did not prevent injecting the distal interphalangeal joint and the common digital extensor tendon.     

Fracture of the Supraglenoid Tubercle Treatment and Results in Five Horses

Five horses with fractures of the supraglenoid tubercle (tuber scapulae) were presented from 6 weeks to 1 year after injury. Clinical signs included lameness characterized by a shortened cranial phase of the stride and shoulder muscle atrophy. Radiographically, the fractures differed in the degree of cranioventral displacement of the fragment as well as the extent of joint involvement. Three horses were treated by resection of the fractured tubercle, including one using a new grid approach to the area of the fracture. Two of these horses have returned to athletic activity and one was euthanized due to postoperative infection. One horse with minimal displacement and joint involvement was treated with rest and raced successfully. One horse seen with a fracture of 1 year duration also was treated with rest and was a pasture sound broodmare.