Caudal foot placement superior to toe elevation for navicular palmaroproximal‐palmarodistal‐oblique image quality

Summary Background Palmaroproximal‐palmarodistal oblique (PaPr‐PaDiO) radiographs are regularly obtained for a full evaluation of the navicular bone (NB). Despite their routine use, different acquisition techniques are described. Objectives To determine optimal foot placement and beam angle for obtaining PaPr‐PaDiO views. Study design In vitro experiment. Methods A convenience sample of 26 disarticulated forelimbs were placed in six different positions using a leg press to mimic the weight‐bearing position. In each position, navicular PaPr‐PaDiO images were obtained with eight different beam angles. The resulting 1248 radiographs were graded for their diagnostic quality and the compacta spongiosa demarcation of the NB. Results Diagnostic quality and compacta‐spongiosa demarcation was graded higher for feet positioned caudally and angle between 40° and 45°. Elevation of the toe significantly decreased the NB palmar border angle (elevated mean: 40.66, SD: 4.46, non‐elevated mean: 42.06, SD: 4.70) (P < .01), but seemed to have no obvious positive influence on radiographs. Main limitations Using disarticulated legs could only mimic positions but, using a press, weight‐bearing positions were replicated as closely as possible. The use of a convenience sample makes the results of the study exploratory only. Conclusions Caudal foot placement seems to improve the image quality of the navicular PaPr‐PaDiO view. The widely used standard beam angle of 45° appears to be the favourable angle for acquisition with a varied range of −5°. Elevation of the toe, standard in most commercially available navicular skyline cassette holders, does not influence the obtained image quality.


| INTRODUC TI ON
Routine protocols for radiographic evaluation of the navicular bone (NB) usually consist of lateromedial (LM) views, dorsoproximalpalmarodistal oblique (DPr-PaDiO) views and palmaroproximalpalmarodistal oblique (PaPr-PaDiO) views. 1 The latter has been described with different techniques, with different positions of the limb and different angles of the primary x-ray beam. In the first description of the PaPr-PaDiO view, placing the foot as far caudally underneath the horse as possible was recommended. 2 Additionally, the primary beam angle should be parallel to the palmar aspect of the digit resulting in the primary beam being in the same plane as the palmar compact bone (PB). 2,3 In a later study, a beam angle of 45° was specified but the position of the limb was not mentioned, 4 whilst others recommended a caudal position of the foot for this beam angle. 5 Another study used a beam angle depending on the foot conformation for the PaPr-PaDiO view, but neither commented on foot position or beam angle nor how foot conformation influenced acquisition parameters in detail. 6 In a recent study, a standard beam angle between 55-65° and 35-45° for additional views was suggested. 7 Different angles depending on the position of the foot are described in the literature, angulation should be 45° when the foot is placed flat on the floor and an angle of 30° is recommended when a wedged block for toe elevation is used. 8 In contrast, some authors suggest the heels should be elevated rather than the toe. 9,10 One study investigating different limb positions and primary beam angles concluded an angle of 47° was favourable, independent of the limb positions. 9 However, in that study, a wedge block to elevate the toe was not used.
In summary, there is no universal agreement on either the beam angle or foot placement for the acquisition of the PaPr-PaDiO view.
Therefore, the aim of the current study was to investigate the optimal position of the leg and the best angle of the x-ray beam for obtaining the PaPr-PaDiO view. We hypothesised that a far caudal position with an elevated toe and a primary beam angle of 45° is superior to other positions and primary beam angles.

| MATERIAL S AND ME THODS
A convenience sample of 26 forelimbs was obtained from an abattoir and breed, age and reason of death were unknown. The legs were disarticulated at the middle carpal joint and a hole was drilled into the third carpal bone and the proximal end of the third metacarpal bone, to apply a custom-made hydraulic leg press to simulate weight-bearing positions as closely as possible. The pressure was applied to achieve full weight-bearing of the sole ( Figure 1) and to the position of the metacarpal bone perpendicular to the floor (positions 1 and 4) and with the foot 3 (positions 2 and 5) and 6 cm caudally (positions 3 and 6). To avoid artefacts, shoes were removed, feet were thoroughly cleaned and frogs were packed with commercial putty (Play-Doh ® , Hasbro UK Ltd).

| Image acquisition
Palmaroproximal-palmarodistal oblique views of all limbs were ac-

| Image rating
All acquired PaPrPaDiO images (n = 1248) were anonymised and each image was randomly assigned a number between 1 and 1248.  Project) and averages were used for further statistical analysis.

| Data analysis
Statistical analyses were performed using IBM SPSS Statistics ver- . Image C was graded highest for diagnostic quality and compacta spongiosa demarcation compared to all other images. Images (A, B and D) were rated with a grade 3 for diagnostic quality, however, compacta spongiosa demarcation was lower in (A and B) compared to (D). Images (E and F) were graded lowest for diagnostic quality and compacta spongiosa demarcation were not normally distributed. The significance of all statistical tests was set at P < .05.

| Angle measurements in different positions
Angle measurements are presented in  Overall, images were graded lower by the diplomate than by the veterinary surgeon.

| Grading of images
Both observers graded the DQ of images in position 6 significantly higher than images obtained in the other positions ( Figure 5). Except for images acquired with an x-ray beam of 40°, the DQ was graded significantly higher by both observers for images obtained with an x-ray beam angle of 45° compared to the other angles.
Compacta-spongiosa-demarcation was graded significantly This beam angle is in accordance with most previous studies, 4,5,8 but not used as standard in a recent study. 7 In the latter, this was instead used as an additional angle, and the routine beam angle used was between 55° and 65°. However, angles above 55° were not found to provide high rated radiographs. Interestingly, the additional angle used in the recent study helped identify erosions of the PB. 7 In combination with the results of the current study, using a beam angle between 40° and 45° is recommended for evaluation of the NB on PaPrPaDiO radiographs. Furthermore, these beam angles are similar to the angle of the PB measured on the lateromedial radiographs and are similar to previously published data. 9 Therefore, these results suggest the best beam angles equal the angle of the PB, which is not surprising as for optimal radiographs the beam should be parallel to the PB. 14 For positions with a raised toe, a flatter angle of 30° has been recommended. 8  Caudal placement of the foot was assumed to cause hyperextension of the DIPJ and, therefore, avoid superimpositions and increase the diagnostic quality. 15 This was also found in the current study, where caudal placement increased the angle of the DIPJ, but not SAP3. Whilst toe elevation changes SAP3, it appears to have less influence on the angle of the DIPJ. The DB and PB appeared to be lower when the toe was raised, but the caudal placement of the foot did not alter these significantly.
Images obtained in positions with the metacarpal bone perpendicular to the floor were graded significantly lower than the most caudal positions, either with or without toe elevation. Therefore, acquiring radiographs in these positions might lead to insufficient DQ and an abnormal decrease in CSD. In some horses, positioning the leg caudally and on the plate might be challenging and lifting the contralateral limb will aid positioning in these cases, but will result in a position with the metacarpal bone perpendicular to the floor. 8,14 However, based on the results of the current study, the negative influence on the resulting radiograph should be considered. It appears to be more appropriate to contemplate alternative restraining methods rather than lifting the contralateral limb for sufficient image quality.
Agreement between both observers was substantial for DQ as well as the SUM and moderate for grading CSD. The latter is in accordance with previously published results, where CSD showed good intra-and interobserver agreement. 6 The better agreement for DQ highlights that judging radiographs for their DQ appears to be a relatively easy task when proper guidelines are available.
However, the more experienced observer graded the images lower overall; this could be due to their more critical judgement of radiographs.
The main limitation of the current study is the use of disarticulated limbs rather than obtaining the images in live horses.
However, obtaining the number of views done per limb in the current study would have led to major radiation safety concerns. The number of legs included was chosen at convenience, and whether a higher study population would have led to more significant differences is unknown. Finally, as no information about the orthopaedic health status of the limbs was available, it is possible that some NB were diseased and had an abnormal CSD. However, as all limbs were radiographed with the same set of image acquisition parameters and a comparison of these were made; it is presumed that this might have limited the influence of disease on the results of the current study.
The current results suggest that a beam angle between 40° and 45° should be used for the acquisition of a PaPrPaDiO radiograph of the NB to obtain an image of high DQ and optimal CSD. Caudal foot placement seems to be more beneficial for image quality than toe elevation.

ACK N OWLED G EM ENT
We thank Dr Yu-Mei Chang for her statistical advice.

CO N FLI C T O F I NTE R E S T S
No competing interests have been declared.

Ethics approval has been given by the Clinical Research and Ethical
Review Board at the Royal Veterinary College (URN: M2018 0148).

I N FO R M ED CO N S ENT
Not applicable: equine samples were sourced via an abattoir.

PEER R E V I E W
The peer review history for this article is available at https://publo ns.com/publo n/10.1111/evj.13563.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.