SEARCH

SEARCH BY CITATION

Keywords:

  • distal border fragments;
  • MRI;
  • navicular

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. ACKNOWLEDGMENTS
  8. REFERENCES

Distal border fragments of the navicular bone are increasingly being detected due to the improved capabilities of magnetic resonance imaging (MRI), but their clinical significance remains unclear. The purpose of this retrospective study was to describe the location, size, and frequency of fragments in a cohort of horses presented for MRI of the foot and to compare MRI findings with severity of lameness. Archived MRI studies and medical records were searched from March 2006 to June 2008. Horses were included if a distal border fragment of the navicular bone was visible in MRI scans. Confidence interval comparisons and linear regression analyses were used to test hypotheses that fragments were associated with lameness and lameness severity was positively correlated with fragment volume and biaxial location. A total of 453 horses (874 limbs) were included. Fragments were identified in 60 horses (13.25%) and 90 limbs (10.3%). Fifty percent of the horses had unilateral fragments and 50% had bilateral fragments. Fragments were located at the lateral (62.2%), medial (8.89%), or medial and lateral (28.9%) angles of the distal border of the navicular bone. There was no increased probability of being categorized as lame if a fragment was present. There was no significant difference in fragment volume across lameness severity categorizations. Confidence intervals indicated a slightly increased probability of being classified as lame if both medial and lateral fragments were present. Findings indicated that distal border fragments of the navicular bone in equine MRI studies are unlikely to be related to existing lameness.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. ACKNOWLEDGMENTS
  8. REFERENCES

Osseous fragments located at the distal border of the distal sesamoid (navicular) bone of the forelimb were first described in horses in 1973.[1] In recent years, the detection of distal border fragments has increased due to the improved imaging capabilities of magnetic resonance imaging (MRI) of the distal limb, but their relationship with clinical lameness remains unclear. Distal border fragments may occur at the medial and lateral angles of the distal border of the navicular bone in the forelimb and are embedded within the origin of the distal sesamoidean impar ligament.[2] Proposed pathogenesis theories include avulsion fracture of the navicular bone, fracture of an enthesiophyte at the origin of the impar ligament, separate center of ossification, and dystrophic mineralization within the impar ligament or adjacent soft tissue structures.[3-6] Fragments are typically located adjacent to a smooth, concave defect in the distal border of the navicular bone,[2, 7, 8] supporting the theory of an avulsion fracture fragment associated with the impar ligament.

Conventional film screen and computed radiography have low sensitivity for identification of fragments (39% and 37.8%, respectively), although specificity is high (99.7% and 100%).[9] Because fragments are often difficult to identify on radiographic images, this may have led to a previous underestimation in the prevalence of these osseous fragments. A good correlation has been found between the presence of distal border fragments identified on MRI and postmortem examination, with the sensitivity and specificity of MRI being 92% and 93%, respectively.[2]

The presence of these osseous fragments has been described in both lame and nonlame horses.[2, 4, 5, 10, 11] The reported radiographic prevalence of navicular bone distal border fragments is 8.8% in young stallions[7] and 7% in clinically healthy horses.[10] In contrast, 40% of middle-aged horses with clinical navicular syndrome had an osseous fragment at the distal border of the navicular bone.[6] Similarly, in a study of horses undergoing radiographs of the front feet, fragments were identified in 3.6% of sound horses, 8.7% of lame horses, and 24.1% of horses with navicular bone pathology,[8] suggesting that distal border fragments are a significant finding in horses with navicular syndrome.

A recent report of MRI in horses with foot pain identified distal border fragments in 13.6% of horses and 9.8% of feet examined.[12] Another study compared MRI findings of horses with recently diagnosed and chronic navicular syndrome. The prevalence of distal border fragments was similar between the two groups: 37.5% and 36%, respectively.[13] Distal border fragments can be identified on MRI as isolated abnormalities, but significant associations between the presence of a fragment and other navicular bone abnormalities[12] suggest that they are a part of the complex of pathological findings that indicate navicular syndrome.

The purpose of this study was to describe the location, size, and frequency of distal border fragments of the navicular bone and investigate the relationship of fragments to concurrent lameness in horses that underwent MRI examination of the front feet at a single MRI referral center for horses between March 2006 and June 2008. The following hypotheses were tested: (1) fragments are associated with lameness, (2) fragment volume is positively correlated with lameness severity, and (3) biaxial fragments are positively correlated with lameness severity.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. ACKNOWLEDGMENTS
  8. REFERENCES

Magnetic resonance imaging reports were reviewed for all horses that underwent MRI of one or both front feet at the Alamo Pintado Equine Medical Center (Los Olivos, CA) between March 2006 and June 2008. Horses were selected for MRI based upon localization of lameness to the foot by diagnostic analgesia and a lack of radiographic, ultrasonographic, or scintigraphic findings that conclusively isolated the source of pain. Horses were included in the current study if a navicular bone distal border fragment of one or both forelimbs was recorded on the MRI report. For all included horses, magnetic resonance (MR) images were acquired using a 1.5-T MRI scanner with a high-resolution eight-channel knee radio frequency coil (Magnetom Espree MRI System, Siemens Medical, Malvern, PA). The following imaging sequences were required for both limbs (Table 1): dual echo – proton density (PD) + T2 weighted—axial and sagittal sequences; T1-weighted volume interpolated enhancement (VIBE) fat suppressed (FS)—axial, sagittal, and dorsal sequences; T2-weighted short tau inversion recovery (STIR)—axial, sagittal, and dorsal sequences; PD FS—dorsal oblique sequence (impar axial). Axial images were obtained using a field of view of 150 mm and slice thickness 3 mm for dual echo and T2 STIR and 2 mm for T1 VIBE. Sagittal and dorsal images were obtained using a field of view 150–175 mm and slice thickness 3 mm for dual echo and T2 STIR and 2 mm for T1 VIBE. PD FS impar axial images were obtained using a field of view 150 mm and slice thickness 2 mm. Some scans included an isotropic T1 VIBE FS Sag with a field of view 150 mm, a slice thickness of 0.7 mm, and a voxel of 0.7 × 0.7 × 0.7 mm. Each MRI evaluation was performed under general anesthesia with the horse placed in right lateral recumbency.

Table 1. Technical parameters used for MRI examinations of equine front feet in the current study
    Gap  Percent    
   Slice(% distance  FOVEchoImaging  
 TRTEthicknessof slice  phasetrainfrequenceyFlipInversion
Sequence(ms)(ms)(mm)thickness)AveragesFOVencodinglength(bandwith)angletime (ms)
  1. TR, repetition time; TE, echo time; ms, millisecond; mm, millimeter; FOV, field of view.

Proton density transverse3000304101320100563.638068150 
T2 transverse30001194101320100563.638068150 
Proton density sagittal3000304101320100563.638068150 
T2 sagittal30001194101320100563.638068150 
T1 VIBE FS transverse9.313.572102320100163.63804912 
T1 VIBE FS sagittal9.313.422102320100163.63805712 
T1 VIBE FS frontal9.313.232102320100163.63802412 
T2 STIR transverse4380744102320100963.638068150150
T2 STIR sagittal4140743102320100963.638068150150
T2 STIR frontal3030743102320100963.638068150150
Proton density FS impar axial2000322102384100563.638018170 

All MR studies meeting inclusion criteria were retrieved and interpreted by a board certified veterinary radiologist (TS) and a board certified equine surgeon (CEJ). The images were evaluated independently by each interpreter, and then discussed to reach a consensus. Each sequence was evaluated individually for the presence or absence of distal navicular border lesions identified on other sequences. If a distal border fragment was present, the location (left or right forelimb and lateral and/or medial position) and size of the fragment were recorded. Fragment size was measured using the diagnostic workstation software's linear measurement tool (eFilm, Merge Healthcare, Chicago, IL) and fragment volume was recorded in cubic millimeters. Volume was calculated based on medial-lateral, proximal-distal, and dorsal-palmar measurements recorded from dorsal and sagittal planes on T1 VIBE sequences. If more than one fragment was present on a foot, individual fragment volumes were recorded and the total fragment volume was determined based on the sum of individual fragment volumes.

Clinical findings were recorded from medical records and consisted of signalment, history, and results of lameness evaluation. Lameness severity was graded according to the American Association of Equine Practitioners lameness grading scale.[14] Lameness examination was performed either by a clinic staff veterinarian or the referring veterinarian. Lameness evaluation was performed on firm ground, at the walk and trot in a straight line and circling in both directions. Diagnostic analgesia was performed first in the lamest limb. Lameness was localized to the foot by palmar nerve block performed just proximal to the collateral cartilages of the foot or at the base of the proximal sesamoid bones. Once diagnostic analgesia was completed in the lamest limb, the horse was reevaluated. If the contralateral limb was then observed to be lame, diagnostic analgesia was then performed on that limb to determine the lameness grade for both limbs. Individual limbs were excluded from the lameness analyses if the lameness grade was not recorded for that limb.

To describe the size and location of the fragments, the mean and standard deviation (SD) of fracture volume for medial and laterally located fragments were calculated. Statistical tests were selected and performed by a trained statistician (CPM) to evaluate associations between fragment presence or volume and lameness diagnosis. A two by two contingency table and a Z-test[15] were used to test the hypothesis that limbs with a fragment detected were more likely to be classified as lame. This test compared the proportion of limbs with a fragment classified as lame (number of limbs with a fragment classified as lame/the total number of limbs) to the proportion of limbs with a fragment classified as not lame (number of limbs with a fragment but not diagnosed as lame/the total number of limbs) against the expected proportions, if there is no effect. Average volume and associated variance per limb were calculated for each lameness categorization (i.e., 0, 1, 2, 3, 4; there were no 5s in the dataset). Using the SD and the estimated mean, the 95% confidence interval (CI) was calculated for each lameness category (CI = mean ± 1.96 × SD) to compare the mean volumes per categorization.[15] If the estimated mean volume for one lameness categorization fell within the 95% CI, it was concluded that no statistical difference was present between the means.[15] Further, to determine if fragment volume was positively associated with increasing lameness score when an animal was known to be lame and known to have a fragment, the mean fragment volume for each level of lameness was again calculated and means were compared using 95% CIs. A least squares regression analysis[16] was used to determine whether lameness scores (the dependent variable) increased with increasing volume of fragments (the independent variable). An F-test was used to determine whether the slope of the estimated regression line was positive and significantly greater than 0. If the slope was positive and greater than 0, it would mean that lameness scores tended to increase as fragment volume increased. Furthermore, the R2 calculated in the regression analysis was a measure of how much variation in the data could be explained by the proposed linear regression model. An R2 value of 0 meant that no variation in the data would be explained by the model and a value of 1 would mean all the variation in the data would be explained by the model. An R2 value >0.25 was considered to be a strong correlation.[16]

Lastly, the hypothesis that fragment location may affect the probability of being lame was tested. The proportion of limbs with medial fragments only, lateral fragments only, or biaxial fragments that were also categorized as lame was calculated. The 95% CIs were estimated for each proportion to compare the calculated proportions inline image; where p is the proportion of limbs in category i, j (e.g., i, j = limbs with only lateral fragments and are lame), and n is the total number of limbs in category i (e.g., all limbs with only lateral fragments).[15] If the proportion (presented as a percentage) fell outside the CI of another proportion, it was considered significantly different from the other proportion. The 95% CIs for mean lameness score were compared for limbs with medial, lateral, or biaxial fragments to see if limbs with medial, lateral, or biaxial had higher lameness scores. Similar to above, if the mean lameness score of one type of fragment fell within the 95% CI of one or both of the other categories it was not significantly different.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. ACKNOWLEDGMENTS
  8. REFERENCES

A total of 453 horses underwent MRI of one or both front feet during the time period reviewed, with 874 front feet available for evaluation. Distal border fragments were identified in 60 horses (13.25% prevalence) and 90 feet (10.3% prevalence). Of these 60 horses, breeds represented included Warmbloods (36.7%), Quarter horses (26.7%), Arabians (10%), Thoroughbreds (8.3%), Appaloosa (1.7%), Irish Sport horse (1.7%), Paint (1.7%), Percheron (1.7%), and mixed breed or unknown (11.7%). These breed percentages were comparable to the overall hospital caseload. The age range was 3–18 years and the mean age was 9.5 years. Geldings accounted for 63.3%, mares for 26.7%, and stallions for 10%.

Distal border fragments appeared as round or oval areas of low signal intensity on dorsal and sagittal MR images (Fig. 1). The fragments generally had smooth margins, even if the contour varied from linear to irregular. On sagittal plane images, some fragments had round to triangular shapes. Fragments were best visualized on T1-weighted VIBE sagittal and dorsal planes images and some were seen on PD impar images. Subjectively, only the largest fragments were visible on the TSE PD, T2W, or STIR sequences. The threshold size for visualization on these sequences was not measured. Comparison figures illustrating TSE PD, T2W, and STIR images of the same horse shown in Fig. 1 are provided (Figs. 2-4).

image

Figure 1. A and B. Sagittal and dorsal plane T1-weighted VIBE fat-suppressed MR images of the front foot for one of the horses, illustrating a distal border fragment of the navicular bone, appearing as a round or oval area of low signal intensity.

Download figure to PowerPoint

image

Figure 2. A sagittal plane proton density MR image from the same horse as Fig. 1.

Download figure to PowerPoint

image

Figure 3. A sagittal plane T2-weighted MR image from the same horse as Fig. 1.

Download figure to PowerPoint

image

Figure 4. (A) A dorsal plane T2 STIR MR image from the same horse as Fig. 1. (B) A sagittal plane T2 STIR MR image from the same horse as Fig. 1.

Download figure to PowerPoint

Thirty horses (50%) had bilateral fragments and 30 (50%) had unilateral fragments. Of the 30 horses with unilateral fragments, 13 (43.3%) occurred on the right foot and 17 (56.7%) occurred on the left foot. Fragments were located at the lateral angle (56/90; 62.2%), medial angle (8/90; 8.89%), or medial and lateral (26/90; 28.9%) angles of the distal border of the navicular bone. No more than one fragment was identified on either angle. Fragment volume ranged from 1 mm3 to 260 mm3. Lateral fragment volumes ranged from 3 mm3 to 260 mm3 and medial fragment volumes ranged from 1 mm3 to 108 mm3. The average volume for lateral fragments was 66.3 mm3 and for medial fragments was 41.3 mm3. No significant difference in lateral vs. medial fragment volume was found.

There were 60 horses with distal border fragments identified on one or both forefeet, and therefore there were 120 forelimbs belonging to these horses. Of the 120 forelimbs, record of lameness evaluation was available for 94 limbs. Sixty-two limbs were classified as lame and 32 limbs were classified as not lame. The mean lameness grade for all forelimbs was 1.63/5. The mean lameness grade for forelimbs without a distal border fragment was 1.80/5. The mean lameness grade for forelimbs with at least one fragment was 1.57/5.

Based on the Z-test, the proportion of limbs that had a fragment present and were diagnosed as lame was 0.62 and the proportion of limbs that had no fragment and were not diagnosed as lame was 0.76. Results indicated that there was no significant difference in these proportions (Z = −1.322, p = 0.09), that is, there was no increased probability of being categorized as lame if a fragment was present. The negative Z value indicated that there was a slightly increased probability, though not statistically significant, of being categorized as lame when no fragment was present.

Mean fragment volume for limbs with lameness grade 0 was 54.6 mm3 (SD = 72.5); mean volume for lameness grade 1 was 89.8 mm3 (SD = 56.4); mean volume for lameness grade 2 was 40.3 mm3 (SD = 69.6); mean volume for lameness grade three was 71.7 mm3 (SD = 71.1); mean volume for lameness grade 4 was 40.0 mm3 (SD = 49.9). There was broad overlap in the 95% CIs for mean fragment volume separated by lameness grade (Fig. 5). All estimated mean fragment volumes fell within the CIs of all other categorizations (Fig. 5) indicating that there was no significant difference in the size of fragments across grades of lameness. The CIs for each lameness grade also encompassed zero, indicating that there was no significant difference in lameness grade for horses with fragments vs. without fragments. These results did not support the hypothesis that navicular bone fragments are associated with lameness diagnoses or severity. When only considering limbs with detected fragments and presence of clinical lameness (i.e., excluding limbs with a lameness grade of zero or limbs with no detected fragments), the pattern did not change. There was still broad overlap of CIs and no significant differences among the mean volumes for each lameness grade (Fig. 6). The linear regression model showed no significant association between fragment volume and lameness grade (b1 = 0.0006 (slope), R2 = 0.003; Fig. 7).

image

Figure 5. Comparison of mean fragment volume (± 95% CI) for limbs classified in each lameness grade (0–4).

Download figure to PowerPoint

image

Figure 6. Mean volume of distal border fragments by lameness grade (± 95% CI) for horses in which the limb was lame and a fragment was present.

Download figure to PowerPoint

image

Figure 7. Linear regression of the association between fragment volume (independent variable) and lameness grade (dependent variable) for limbs with a distal border fragment.

Download figure to PowerPoint

The percentage of limbs with lateral fragments only that were lame was 57.14% (SE = 14.97%). The percentage of limbs with medial fragments only that were lame was 33.3% (SE = 37.72%). The percentage of limbs with biaxial fragments that were lame was 80.95% (SE = 16.79%). The proportion of limbs with biaxial fragments was greater than the upper bound of CIs for limbs with only medial (CI upper bound = 71.05%) and only lateral fragments (CI upper bound = 72.11%), though the CIs did overlap (Fig. 8). Conversely, the proportion of limbs with only medial or only lateral fragments fell below the lower bound of the CI for the proportion of limbs that had biaxial fragments (CI lower bound = 64.15%). This indicated that there is a slightly increased probability of being classified as lame, if fragments were present in both locations as compared to one. There was no measurable difference in the proportion of limbs classified as lame with either medial only or lateral only fragments.

image

Figure 8. Comparison of percentage of limbs classified as lame (± 95% CI) for limbs with medial only, lateral only, or both medial and lateral fragments.

Download figure to PowerPoint

There was broad overlap in the 95% CIs comparing mean lameness scores for limbs with medial (mean lameness = 1.17, SD = 1.83), lateral (mean lameness = 1.47, SD = 1.36), or biaxial (mean lameness = 1.86, SD = 1.12) fragments. A linear regression model confirmed that, although there was a slight increase in lameness from medial to lateral to biaxial fragments, the increase was not significant (b1 = 0.359, R2 = 0.025; Fig. 9).

image

Figure 9. Comparison of mean limb lameness grade for limbs with medial fragments only, lateral fragments only, or medial and lateral fragments including a linear regression showing a nonsignificant increasing trend.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. ACKNOWLEDGMENTS
  8. REFERENCES

Contrary to the hypothesis, there was no significant association between the presence of an osseous fragment at the distal border of the navicular bone in MRI studies and the likelihood of the limb being classified as lame. There was no significant difference in the size of fragments across grades of lameness, either when including or excluding limbs with a lameness grade of zero (sound limbs). There was a small significant increase in the likelihood that limbs with fragments located at both the medial and lateral angles of the distal border would be classified as lame compared to limbs with a fragment located at the medial angle or the lateral angle alone. However, there was no significant association between the location of the fragment(s) and the severity of lameness.

During this study period, prevalence of distal border fragments was 13.25% among scanned horses and 10.3% among scanned front feet. This prevalence was similar to a recent study of horses that underwent high-field MRI in which fragments were observed in 13.6% of horses and 9.8% of feet.[12] The prevalence was lower than several studies that reported distal border fragments in populations of horses diagnosed with clinical navicular syndrome.[6, 8, 13] Of the horses that had osseous fragments at the distal border, 50% had unilateral fragments and 50% had bilateral fragments. This finding was similar to a previous study in which 56.9% of horses had unilateral fragments and 43.1% had bilateral lesions.[12] In the present study, fragments were more commonly located at the lateral angle of the distal border of the navicular bone (62.2%) compared to the medial (8.89%) or medial and lateral (28.9%) angles. A comparable distribution was reported in a study that identified 72.2% of fragments on the lateral angle, 17.5% on the medial angle, and 10.5% on both the medial and lateral angles.[7] An elongation of the proximolateral extent of the navicular bone has been noted as a frequent finding on radiographs of the navicular bone in both sound horses[10] and those with clinical navicular syndrome.[5] The same forces that contribute to this remodeling process may play a part in the etiology of lateral distal border fragments.[12] No apparent association between the presence of a fragment and breed, age, or sex was noted in the present study, as the distribution of fragments among these categories approximated the clinic caseload.

In a study looking at horses with chronic navicular syndrome with fragments in one limb only, fragments were associated with the lamest limb in 64% of the time.[13] Similarly, fragments observed on radiological evaluation were more frequent in lame horses (8.7%) compared to sound horses (3.6%).[8] In the present study, no significant statistical association was found between the presence of a fragment and the likelihood of being classified as lame. In addition, the mean lameness grades of limbs with and without fragments were similar. In fact, the mean lameness grade was slightly lower for limbs with fragments. This finding was not indicative of a true protective effect as there was no statistical significance. Therefore, the simple presence of a fragment was not associated with the presence or increased severity of current clinical lameness.

Previous studies have reported that distal border fragments appear to be more prevalent in horses with navicular syndrome[6, 8, 12] and are often associated with other pathological findings. One study found that 88.3% of navicular bones with distal border fragments had other detectable abnormalities, and there was a significant association between the presence of a fragment and overall navicular bone grade, osseous cyst-like lesions, increased number and size of the synovial invaginations, and proximal extension of abnormal signal intensity in the adjacent navicular bone.[12] Another study found that limbs with distal border fragments often had additional abnormalities, such as impar desmopathy and/or edema-like signal in the distal third of the navicular bone identified on MRI. [13] This would suggest that traumatic or degenerative etiologies may result in distal border fragmentation. While the presence of a distal border fragment appears to be associated with navicular syndrome, and it may be anticipated that limbs with evidence of significant navicular bone pathology would be more likely to be lame, our study findings indicated that identification of a fragment does not cause the limb to have a higher grade of lameness simply because of its presence.

The purpose of this study was neither to assess the primary cause of lameness nor to designate the amount of lameness in each horse that was directly attributable to the fragment. The objective was to describe associations among presence, location, and volume of distal border fragments vs. concurrent diagnosed lameness. There were multiple abnormalities identified in the horses included in the study and the association of fragments with other pathological findings was not evaluated. Causes of lameness are abundant, and future research might examine the association of distal border fragments with lameness when combined with other pathology.

In the current study, distal border fragments were measured on the T1 VIBE gradient echo sequence. The thin-slice thickness and excellent bony resolution of this sequence provided a superior method of viewing these small fragments. The T1 VIBE was performed in three planes, allowing approximation of fragment volume. Fragment volume ranged widely and no significant difference was found between the size of fragments located at the lateral and medial angles of the distal border. There was also no significant difference in the size of fragments across grades of lameness. This suggested that limbs with larger total fragment volume were not more likely to be classified as lame or have a greater degree of lameness than limbs with smaller total fragment volume.

There was a small significant increase in the probability of a limb with fragments located on both the medial and lateral angles being classified as lame, compared to a limb with a fragment on only one side. Navicular bones with more than one fragment may represent a more advanced form of navicular bone degeneration or impar desmopathy, and hence be associated with a greater chance of being clinically lame. While there was also a slight increase in the severity of lameness on limbs with both medial and lateral fragments, the results were not statistically significant.

Several limitations existed for this study. A longer study time period and greater number of horses may have resulted in greater statistical significance in some of the trends that were observed. In addition, because of the retrospective nature of the study, records of lameness evaluation were not available for both front limbs of all horses, resulting in a reduced number of limbs that were able to be included in the lameness calculations. Lameness evaluations were performed by both the staff veterinarians and by multiple referring veterinarians, possibly resulting in some variability of lameness interpretation. Seventy percent of the primary lameness exams were performed by staff veterinarians and the remaining 30% were performed by referring veterinarians. When the primary lameness exam was performed by a referring veterinarian, the findings were confirmed by a staff veterinarian prior to MRI in the majority of cases. The use of multiple veterinarians to assess the lameness grade assumed that each practitioner was trained according to and followed closely the American Association of Equine Practitioners (AAEP) guidelines. However, it is possible that some lack of standardization in lameness assessments across practitioners reduced the significance of some of the patterns and trends observed in this study (e.g., the positive but statistically insignificant correlation between fragment location and lameness score).

In an ideal world, this study would have been completed using a control group of sound horses with distal border fragments. However, MRI is a technique usually reserved for evaluation of lameness, and not generally used as a screening tool for sound horses. It was not feasible to perform MRI on a group of sound horses numerous enough to provide a control group. Hence, almost all of the horses included in the study were lame in one or more feet. Because of this, the population evaluated was skewed toward lame horses and may not be truly representative of the frequency and significance of distal border fragments in the general population of horses. The limbs that were included in the sound group were determined to be free of lameness after diagnostic analgesia of the contralateral limb and served as internal controls.

The work performed here supports the need for further research. In this study, the association of distal border fragments to current lameness was evaluated, but the contribution to future lameness should be further clarified. The presence of a fragment in a nonlame limb may be an indicator of the presence of subclinical degeneration of the navicular bone. For example, it would be useful to determine whether a distal border fragment, found incidentally on a prepurchase examination, would likely be an important factor in the prognosis for future soundness.

In conclusion, findings from the current study indicated there was no association between the presence of an osseous fragment at the distal border of the navicular bone in MRI studies and the likelihood of the limb being classified as lame. There was no significant difference in the size of fragments across grades of lameness. There was a small significant increase in the likelihood that a limb with fragments located at both the medial and lateral angles of the distal border would be classified as lame, however, there was no significant association between the location of the fragment(s) and the severity of lameness. Therefore, while distal border fragments may occur at a higher frequency in horses with navicular syndrome, the simple presence of a fragment is unlikely to be related to existing lameness.

ACKNOWLEDGMENTS

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. ACKNOWLEDGMENTS
  8. REFERENCES

The authors thank Ms. Elizabeth Files for technical assistance.

Any use of trade, product, or firm names is for descriptive purposes only and does not imply endorsement by the US Government.

REFERENCES

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. ACKNOWLEDGMENTS
  8. REFERENCES
  • 1
    Numans SR, Van de Watering CC. Navicular disease: podotrochlitis chronica aseptica podotrochlosis. Equine Vet J 1973;5:17.
  • 2
    Schramme M, Murray R, Blunden A, Dyson S. A comparison between magnetic resonance imaging, pathology, and radiology in 34 limbs with navicular syndrome and 25 control limbs. Proc Am Assoc Equine Pract 2005;51:348358.
  • 3
    Blunden A, Dyson S, Murray R, Schramme M. Histological findings in horses with chronic palmar foot pain and age-matched control horses. Part 1: navicular bone and related structures. Equine Vet J 2006;38:1522.
  • 4
    Poulos PW, Brown A, Brown E, Gamboa L. On navicular disease in the horse. A roentgenological and patho-anatomic study. Part II. Osseous bodies associated with the impar ligament. Vet Radiol 1989;30:5053.
  • 5
    Wright I. A study of 118 cases of navicular disease: radiological features. Equine Vet J 1993;25:493500.
  • 6
    Wright IM, Kidd L, Thorp BH. Gross, histologic and histomorphometric features of the navicular bone and related structures in the horse. Equine Vet J 1998;30:220234.
  • 7
    Claerhoudt S, Pille F, Vanderperren K, et al. Association between navicular bone fragmentation and shape in Belgian Warmblood horses. Vet Comp Orthop Traumatol 2011;2:132136.
  • 8
    Biggi M, Dyson S. Distal border fragments and shape of the navicular bone: radiological evaluation in lame horses and horses free from lameness. Equine Vet J 2012;44:325331.
  • 9
    Biggi M, Dyson S. Comparison between radiological and magnetic resonance imaging lesions in the distal border of the navicular bone with particular reference to distal border fragments and osseous cyst-like lesions. Equine Vet J 2010;42:70712.
  • 10
    Kaser-Hotz B, Ueltschi G. Radiographic appearance of the navicular bone in sound horses. Vet Radiol Ultrasound 1992;33:917.
  • 11
    Dyson S. Radiological interpretation of the navicular bone. Equine Vet Educ 2008;20:268280.
  • 12
    Biggi M, Dyson S. High-field magnetic resonance imaging investigation of distal border fragments of the navicular bone in horses with foot pain. Equine Vet J 2011;43:309316.
  • 13
    Sampson SN, Schneider RK, Gavin PR. Magnetic resonance imaging findings in horses with recent and chronic bilateral forelimb lameness diagnosed as navicular syndrome. Proc Am Assoc Equine Pract 2008;54:419434.
  • 14
    Lameness exams: Evaluating the lame horse. Am Assoc Equine Pract 2005. Available at: http://www.aaep.org/health_articles_view.php?id=280 (accessed May 1, 2013).
  • 15
    Sokal RR, Rohlf FJ. Biometry, 3rd ed. New York: W.H. Freeman and Company, 1995;139–143, 593608.
  • 16
    Neter J, Kunter MH, Nachtsheim CJ, Wasserman W. Applied linear statistical models. Boston: McGraw Hill, 1996;395.