Raphael Labens, MagMedVet, DEC-EQ, CertES(Orth), MVM, MRCVS
College of Veterinary Medicine, North Carolina State University, 4700 Hillsborough Street, Raleigh NC 27606, USA

Lance C. VOÛTE BVSc, CertES(Orth), MRCVS
Weipers Centre for Equine Welfare, Division of Companion Animal Sciences, Institute of Comparative Medicine, University of Glasgow Veterinary School, Bearsden Rd, Glasgow, G61 1QH, UK

Introduction:

Navicular disease (ND) is a chronic forelimb disease associated with pain arising from the navicular bone (NB) and closely related structures, including the collateral ligament (CL) of the navicular bone, the distal sesamoidean impar ligament (DSIL), the navicular bursa (NBU), and the deep digital flexor tendon (DDFT). The navicular apparatus (NA) is the functional unit composed of the NB, CL, DISL, NBU and DDFT (Wissdorf et al., 1998). The clinical presentation of ND can be variable, ranging from a severe and sudden onset of unilateral lameness to a slow and progressive development of bilateral forelimb lameness (Dyson, 2003). An accelerated aging process caused by overloading of the navicular apparatus is thought to be responsible for the development of the disease (Wright and Douglas, 1993; Wright et al., 1998). A number of different imaging modalities have been reported to be useful for examination of the NA, including radiography (Butler et al., 2000a), ultrasonography (Busoni and Denoix, 2001; Kristoffersen and Thoefner, 2003), nuclear scintigraphy (Trout et al., 1991; Dyson, 2002), computed tomography (CT) (Widmer et al., 2000; Garcia-Lopez, 2003), magnetic resonance imaging (MRI) (Dyson et al., 2003b) and bursoscopy (Wright et al., 1999).

This report describes a case of ND causing severe bilateral fore limb lameness, its clinical presentation including scintigraphic, radiographic and ultrasonographic findings, as well as treatment, progress and post mortem examination of the affected animal.

Case details

History: A six year old grey Thoroughbred cross gelding, used for showjumping, was presented to the Weipers Centre for Equine Welfare at the University of Glasgow for investigation of marked right fore (rf) limb lameness of 10 days’ duration. The onset was reported to be sudden and had apparently occurred while the gelding had been turned out in the field.

Clinical findings: No abnormalities were detected on general physical examination. Orthopaedic examination revealed mediolateral foot imbalance and increased digital pulses in the rf. Hoof testers applied to the rf heels elicited a mild pain response. At trot in a straight line on a hard surface the horse was 3/5 (AAEP, 1999) rf limb lame. This lameness was completely abolished by a palmar digital nerve block (Bassage and Ross, 2003), using mepivacaine (Intra-epicaine1), and the horse rendered 1/5 lf limb lame. On examination the next day the horse had become 4/5 rf limb lame and very reluctant to walk, and it was therefore considered inadvisable to perform further diagnostic local anaesthesia. As an alternative metabolic and anatomical diagnostic imaging, including scintigraphy, radiography and ultrasound was performed.

Scintigraphic examination: For this purpose a gamma camera with a 53 x 39 cm field of view, fifty-five photomultiplier tubes and low energy general purpose collimator was used 2. Foot images were acquired for 100 000 counts using a matrix size of 256 x 256 x 16 and were processed with Micas X plus software3.
An indwelling jugular vein catheter (Intraflon4, 14 G, L. 80 mm) was placed and the horse was injected with 5 GBq of technetium 99m-methylene diphosphonate (99mTc-MDP)5. Pool phase and bone phase images, within 15 minutes and from 2 ½ hours after injection respectively, were obtained under sedation with detomidine (Domosedan6 0.001 mg/kg bwt i.v.) and butorphanol (Torbugesic7 0.025mg/kg bwt i.v.). Pool phase images consisted of lateral views and bone phase images of lateral and solar views of both front feet (Figure 1, 2).

Pool phase imaging detected two areas of intense focal increased radiopharmaceutical uptake (IRU) distal to the coronary band and palmar to P2/P3 in the rf foot. Only a mild IRU was visible in the same region in the lf foot. Using a quantitative method of scintigraphic assessment (Dyson, 2002) bone phase imaging illustrated a greater than 50 % focal IRU in the central area of both fore NBs relative to P3 (Figure 3 a, b).

Figure 1: Lateral pool phase images (RF Pool 1 at 5 and Pool 2 at 15 minutes, LF pool 1 at 10 and Pool 2 at 12 minutes post injection with 99mTc-MDP)

Figure 1: Lateral pool phase images (RF Pool 1 at 5 and Pool 2 at 15 minutes, LF pool 1 at 10 and Pool 2 at 12 minutes post injection with 99mTc-MDP)
Note the IRU in the normally photopenic palmar foot area (arrow).

Figure 2: Lateral and solar bone phase images
Note the focal intense IRU over the NBs

Figure 3 a: Solar bone phase images with regions of interest (ROI’s) drawn in the region of the (1) NB, (2) the region of insertion of the DDFT, (4) the toe region of the distal phalanx, (3) the medial and (5) lateral aspects of the distal phalanx and (6) the central area of the NB.

 

Figure 3 b: Difference (%) in counts between ROI 1 and 6 with ROI 3, 4 and 5 mean counts.

Radiographic examination:
In both fore limbs a lateromedial (LM) and dorsopalmar (DP) view of the foot (LM: 58 kV/5.25 mAs; DP: 60 kV/ 5.25 mAs)* dorsoproximal-palmarodistal oblique view of the distal phalanx (upright pedal bone: 48 kV/5.25 mAs), dorsoproximal-palmarodistal oblique view of the NB (upright NB: 76 kV/5.25 mAs/ N70 parallel grid9) and palmaroproximal-palmarodistal oblique view of the distal phalanx at differing (40-50˚) angulations (skyline NB: 58 kV/50 mAs) were obtained using a ceiling mounted10 (max. 100 KV/ 400 mAs) and portable x-ray machine11 (max. 100 KV/30 mAs).
On the proximal, proximolateral and proximomedial border of both fore NBs entheseous bone was present (Figure 4 a, b). The NBs were classified as shape two and grade three (Dik and Vandenbroek, 1995; Dik et al., 2001b). At the distal border, six (rf NB)/ seven (lf NB) lucencies were visualised, of which three in the lf NB had a “lollipop” appearance but were not associated with sclerosis. On the lf skyline NB view, a loss in corticomedullary definition and sclerosis of trabecular bone was evident. The palmar cortex of the flexor surface of the lf NB, lateral to the sagittal ridge, appeared irregular and flattened with cortical radiolucent areas suggestive of bone lysis (Figure 5). On the rf skyline NB view, a similar but indistinct radiolucency lateral to the sagittal ridge was suspected (Figure 6).

Figure 4: Right (a) and left (b) upright navicular bone view
Note the entheseous new bone at the insertion site of the CLs and the proximal border of the NB flexor cortex (broad white arrows). The proximal articular border of the NB has an undulating appearance (narrow white arrows). Distal to the left NB a mineral radiodensity (stone) can be seen giving the false impression of an osteochondral fragment at the distal NB border (black arrow).

Figure 5: LF skyline NB view
Loss in corticomedullary definition and trabecular sclerosis can be appreciated. Note the flattening and cortical radiolucency lateral to the sagittal ridge (arrow).

Figure 6: RF skyline NB view
A small cortical radiolucency lateral to the sagittal ridge is suspected (arrow).

 

Ultrasonographic examination:

Transcuneal ultrasonography of the rf and lf navicular apparatus was performed (Siemens Elegra 25612) using a 7.5 MHz linear transducer. Patient preparation for this approach followed published guidelines (Busoni and Denoix, 2001). Images of the DDFT, flexor surface of the NB, overlying fibrocartilage, NBU and DSIL were obtained in both front feet and compared to each other. Entheseous new bone at the insertion site of the rf DSIL (Figure 7) and a decrease in distance between the palmar aspect of the lf NB and dorsal DDFT border were appreciated (Figure 8). The latter was speculated to either represent a loss of fibrocartilage covering the flexor surface of the lf NB or to reflect an effusion of the rf NBU.

Figure 7: Sagittal ultrasonographic view of the rf DISL insertion site
Note the roughened bone surface at the insertion site of the DISL (arrow).

Figure 8: Sagittal ultrasonographic view of the rf and lf NB (1), DDFT (2) and NBU and fibrocartilage (arrow)
Note the difference in distance between the NB flexor surface and dorsal DDFT border (arrow).

 

Differential diagnosis: Nerve block results located the pain process in the distal limb and radiographic and scintigraphic examinations identified abnormalities of the NA which therefore excluded differential diagnoses for foot pain other than ND. The exact nature and extent of the lesions involved remained uncertain and included the possibility of a DDFT, DISL or CL lesion additionally to the identified NB pathology.

Diagnosis: A diagnosis of navicular disease in both front feet was made. Soft tissue damage such as DISL entheseopathy/desmitis and tendonitis of the DDFT was a suspected cause of the exacerbation of the rf lameness.

Prognosis: Given the patient’s young age, intended use, severe lameness and the presence of NB changes the horse was given a guarded prognosis for full recovery (Hickman, 1989).

Treatment: Corrective trimming was performed and orthopaedic shoeing consisted of graduated egg bar shoes with a medial wedge and a squared toe in both fore feet. A course of phenylbutazone13 (2.2 mg/kg bwt bid PO for 5 days then 2.2 mg/kg bwt sid PO for 10 days) and isoxuprine hydrochloride14 (1 mg/kg bwt bid PO for 4 weeks) was prescribed. The horse was to be box rested first and after completion of the course of non steroidal anti- inflammatory drugs (NSAIDs), exercised at walk for an additional 2 weeks. Re-assessment was scheduled for 4-6 weeks following release from the hospital.

Follow up: On follow up 6 weeks later, the owner reported that the horse had been frequently non-weight bearing on the rf when stabled and no improvement had been appreciated. Re-assessment showed that the rf limb lameness had deteriorated to grade 4/5. The horse was very reluctant to walk but when persuaded to to do so placed the rf foot toe first to the ground.
In light of the negative response to treatment and given that the prognosis for the horse returning to high level competition was poor no further treatment was attempted and euthanasia performed.

Post-mortem (pm) examination: Both fore NBs showed marked to severe loss of fibrocartilage at and abaxial to the sagittal ridge of the flexor surface (Figure 9). Both DDFTs adjacent to the flexor surface of the NBs showed yellow discolouration, superficial tendon fibre disruption and some localized intra-tendinous haemorrhage (Figure 10). Complete fibrocartilage loss and some adhesions of disrupted DDFT fibres were apparent in the area lateral to the sagittal ridge of the left fore NB (Figure 11). Radiographic imaging of the isolated NBs clearly demonstrated small areas of cortical bone lysis lateral to the sagittal ridge in the rf NB (Figure 12, 13).

Figure 9: Both fore NBs isolated at pm examination demonstrating the fibrocartilage erosion on their flexor surface (arrow).

Figure 10: Picture demonstrating the flexor surface of the rf NB and dorsal aspect of the DDFT after disarticulation at the level of the DIP joint. The DISL has been severed and the NB flipped proximally to expose its flexor surface. Intra-tendinous haemorrhage and superficial tendon fibre disruption can be appreciated on the dorsal surface of the DDFT.

Figure 11: Close up of the flexor surface of the lf NB
Note the full thickness erosion of the fibrocartilage and the loss of cortical bone exposing trabecular bone (blue arrow). Adhesions of the DDFT have been disrupted in the area enclosed by the box.

Figure 12: Skyline view of the lf NB after isolation at pm examination
Note the area of cortical bone lysis lateral to the sagittal ridge (arrow) and the area of trabecular bone sclerosis (box).

Figure 13: Skyline view of the rf NB after isolation at pm examination
Note the area of cortical bone lysis lateral to the sagittal ridge (arrow).

 

Histopathology:

Samples of the fore and hind DDFTs at the level of the NBs were taken and sections in transverse and longitudinal planes were obtained. In contrast to sections of the normal hind DDFTs the samples of the rf DDFT were very friable and easily disintegrated (Figure 14). On microscopic examination of the sections of the rf DDFT bundles of tendon fibres were interspersed with connective tissue showing a high vascularity (Figure 15). These bands of connective tissue gave the specimen an heterogeneous macroscopic appearance. They were interpreted as fibrous tissue filling areas of ruptured tendon fibres.

Figure 14: Tendon specimens and sections for comparison Gross inspection of the original samples revealed differences in tissue homogeneity. Sections of the rf DDFT frequently disintegrated due to altered composition.

Figure 15: Transverse section of the rf DDFT
Note the increased amount of connective tissue between tendon fibres and its high vascularity (black arrows) and cellularity.
(1=tendon fibres, 2=connective tissue)

 

Discussion:

This case report illustrates some of the common clinical features of horses affected with ND - bilateral but asymmetric forelimb lameness aggravated on a circle on a hard surface and abolished by palmar digital nerve block (Wright, 1993a). The young age and severity of the rf limb lameness though represent rather uncommon clinical features. In horses affected with the disease a mean age of 9,2 years, with a range of 3-18 years has been reported (Wright, 1993a). This horse’s age of six years lies within the range reported but is still below the mean age of horses affected with ND. Due to the progression of lameness on initial presentation no further more specific diagnostic analgesia such as a DIP joint or NBU block were performed which would have localised the pain process more precisely (Schramme, 2002). Using different diagnostic imaging techniques such as scintigraphy, radiography and ultrasound, abnormalities of the NA were detected. This together with nerve block results led to the diagnosis of ND.
ND is hypothesised to be a degenerative disease influenced by certain predisposing factors such as foot conformation and hoof balance (Wright et al., 1998). In the case reported here a marked mediolateral foot imbalance only was appreciated in both fore limbs, whereas collapsed heels, flat feet and a broken back foot/pastern axis have been identified as predisposing factors (Wright et al., 1998). Similarly to other degenerative orthopaedic diseases the clinical presentation does not necessarily correlate with diagnostic imaging results. The relatively mild radiographic and ultrasonographic findings compared to the actual pathological changes seen at pm examination demonstrate this discrepancy.

Ultrasonographic examination identified a smaller distance between the lf palmar subchondral NB surface and corresponding dorsal surface of the DDFT (Figure 8), which may have represented fibrocartilage loss of the lf NB or distension of the rf NBU. Pm examination however identified a full thickness defect of fibrocartilage and also loss of underlying cortical bone (Figure 11). The DDFT thickness was measured in a sagittal ultrasonographic view and appeared similar in both fore limbs. Additionally, no intra-tendinous lesions were appreciated. Post mortem microscopic examination of transverse sections of the fore DDFTs at the level of the NB displayed clear evidence of chronic tendonitis, which was more pronounced in the DDFT sections of the lamer rf limb (Figure 15).

Identifying and assessing lesions with transcuneal ultrasonography requires them to be located in the sagittal NB region. Abnormalities ab- or axial to this position are difficult to detect ultrasonographically (Busoni and Denoix, 2001; Kristoffersen and Thoefner, 2003). The selection of a linear transducer for this examination reduced the size of the acoustic window compared to a sector transducer, which additionally could also explain the author’s failure to detect soft tissue pathology.

Radiographic imaging of the lf NB suggested the presence of a localised palmar cortical bone defect (Figure 5) but its full extent was only revealed on pm examination (Figure 12). When the rf navicular bone was isolated a small area of cortical bone lysis was easily appreciated radiographically (Figure 13) in contrast to the images taken ante mortem (Figure 6). In a study investigating the significance of radiographic findings flexor cortex defects and entheseophytes at the proximal border were over represented in animals exhibiting higher grades of lameness (Wright, 1993a, b). In this context it is interesting that in the case reported, the more severely lame limb showed less pronounced pathology. Obtaining a skyline NB image after injection of contrast agent into the NBU might have improved assessment of the NB flexor surface and contributed to the detection of further abnormalities (Turner, 1998).

Using a radiographic grading system (Dik and Vandenbroek, 1995; Dik et al., 2001b) both fore NBs were classified as shape 2 and grade 3 which represent a high predisposition for the development of ND in Dutch Warmblood horses. Assuming that the shape of the proximal articular border of the NB determines the distribution of biomechanical forces that predispose to the development of ND (Dik et al., 2001a) the grading system could also apply to Thoroughbreds (as in this case) and other breeds. The findings in this case supported the evidence that scintigraphy is a more sensitive diagnostic imaging technique than radiography in the investigation of ND (Trout et al., 1991). In the report by Trout et al (1991) results of pool or soft tissue phase imaging were indicative for ND when the distinctive radiopharmaceutical uptake pattern of the equine foot was obscured by uptake in the normally photopenic area between the coronary band and lamellar region. Pool phase images in this case did show this characteristic especially the images of the rf foot (Figure 1). The position of the marked rf IRU was suggestive of DDFT pathology (Trout et al., 1991; Dyson, 2002). Results of bone phase imaging demonstrated an over 50 % increase of radiopharamaceutical uptake at the central NB region in both fore limbs indicating significant bony involvement in the disease complex (Figure 3).

Further investigation of the navicular apparatus could have been achieved with MRI, which would have represented a useful tool in assessing the nature and exact extent of the lesions. In a recent study by Dyson (Dyson et al., 2003a) the clinical findings and diagnostic imaging results with MRI, scintigraphy, ultrasonography and radiography of 46 horses suffering from DDFT tendonitis in the digit were discussed. An over-representation of the latter lesion in jumping horses was appreciated. Scintigraphy was interpreted as having a low sensitivity but high specificity in the detection of DDFT lesions with pool or bone phase imaging. Ultrasonographic investigation, including the transcuneal and heel bulb approaches, did not contribute to the identification of significant abnormalities. Most of the DDFT lesions detected with MRI were seen abaxially within one of its lobes, which in the author’s opinion explained the inability of transcuneal ultrasonography to detect DDFT lesions in this case. In this case ultrasonography of the digit via the heel bulb, which allows access to the more proximal portion of the DDFT, was not performed. Positioning of the transducer head and the ability to get a near perpendicular beam to the DDFT determines the quality and successful interpretation of the images obtained (Reef, 1998) and in the author’s experience this can be difficult. In contrast to the transcuneal approach where thickness and moisture of the frog influences success with imaging (Busoni and Denoix, 2001) however, the heel bulb approach allows easy access in all horses without special preparation of the site. CT and bursoscopy are two further diagnostic imaging modalities that may be of use in the investigation of ND. In a study assessing radiography, MRI and CT for evaluation of navicular syndrome in horses MRI and CT were superior detecting bony and soft tissue lesions (Widmer et al., 2000). The advantage of bursoscopy is that it allows direct inspection of the NBU, palmar cortex of the NB, dorsal surface of the DDFT and the debridement of any lesions. Lesions out with the NBU and its margins however, would not be detected by bursoscopy.

Gross, histological and histomorphometric parameters of diseased, aged matched healthy and immature navicular bones have been documented (Wright et al., 1998). Observations from this survey identified a number of characteristic pathological findings in horses with ND: marked palmar surface fibrocartilage lesions; palmar cortex erosions; DDFT fibrillation and adhesion formation; distal border fragmentation; areas of medullary lysis; and entheseous bone formation at the proximal border. All of the above except distal border fragmentation and medullary lysis were found on pm examination in this case. Conservative treatment options for ND include rest, corrective trimming and remedial farriery, systemic administration of NSAIDs, modulators of NB blood flow and metabolism, intra-articular medication of the DIP joint and NBU with corticosteroids and hyaluronic acid and systemic administration of polysulphated glucosaminoglycans. Surgical options such as navicular suspensory desmotomy and palmar digital neurectomy have been discussed for the treatment of ND (Turner and Tucker, 1989b; Wright and Douglas, 1993; Dabarainer and Carter, 2003; Denoix et al., 2003).
In an initial approach treatments for this case consisted of rest, improvement of foot balance, application of a modified egg bar shoe according to published guidelines (Willemen et al., 1999; Wilson et al., 2001), to facilitate break over and reduce the biomechanical forces exerted on the NA, and the administration of phenylbutazone and isoxsuprine. Tiludronate, a bisphosphonate, would have been included in the initial treatment regimen had it been readily available to the author.

Following re-assessment, further treatment was not pursued given the indications that the horse’s prognosis was poor. Navicular suspensory desmotomy would have represented a valid treatment option. Palmar digital neurectomy however, was considered inappropriate due to concerns about progression of the DDFT lesion. Intra articular or intra bursal medication of the DIP joint or the NBU with corticosteroids were felt to be unlikely to result in more than a temporary, and perhaps incomplete, remission of clinical signs and were therefore not used.
The prognosis of horses affected with ND is very difficult to judge as there are no reports on the correlation of lesions with outcome of treatments. After the development of classical lesions the disease can be considered incurable (Hickman, 1989). Acceptable success rates with different treatment options have been reported but the results can be variable (Ostblom et al., 1984; Turner and Tucker, 1989a; Wright, 1993c; Denoix et al., 2003).
This case illustrated that a severe and sudden onset of lameness, similar to the clinical presentation of a fracture, may occur in a horse with ND. Interpretation of diagnostic findings has to be done cautiously as they may underestimate the pathology present. Using additional imaging modalities in this case, such as MRI, CT imaging or bursoscopy, may have enabled a more complete assessment to be made ante mortem. Having identified the gross pathological changes at pm examination, the concerns for the full recovery of the patient which were based on the clinical severity of the disease and the failure to respond to treatment, were justified.

*Screen/Film combination for all x-rays: U-V Rapid/Curix Ultra UV-G8

 

Veterinary products and manufacturers:

1. Arnolds Veterinary Products Ltd, Cartmel Drive, Harlescott, Shrewsbury, UK, SY1 3TB
2. Nucline Mediso, Budapest, Hungary
3. Bartec Technologies Ltd, Farnborough, UK
4. Vycon, 95440 Ecouen, France
5. Technetium methylene diphosphonate, Western Infirmary, Glasgow, UK
6. Pfizer Animal Health Ltd, Walton Oaks, Tadworth, Surrey, UK, KT20 7NS
7. Fort Dodge Animal Health, Flanders Road, Hedge End, Southampton, Hants, UK, SO30 4QH
8. Agfa-Gevaert N.V., 2640 Mortsel-Belgium
9. X-ograph Imaging Systems Ltd, Hampton Street, Tetbury, Glos GL8 8LD, UK
10. APR-VET/FL Veterinary Systems - HF Generators, Sedecal, Madrid, Spain
11. Atomscope HF 300, Veterinary X-Rays, Beaconsfield, Buckinghamshire, HP9 2QZ, UK
12. Siemens Aktiengesellschaft, Bereich Medizinische Technik, D-91050, Erlangen, Germany
13. Equipalazone Powder, Arnolds Veterinary Products Ltd, Cartmel Drive, Harlescott, Shrewsbury, Shropshire, SY1 3TB, UK
14. Navilox, Vetoquinol UK Ltd, Wedgwood Road, Bicester, Oxfordshire, OX26 4UL

• Case study