Objective: Canine hip dysplasia is a common orthopaedic disease in dogs. The Norberg angle value is a measure of hip joint laxity. The aim of this study was to quantify the degree of rotation necessary to consider the radiograph as inadequately positioned and to determine the influence of rotation on the Norberg angle.
Study design: Three sets of radiographs with different positioning and projections were acquired using 10 canine cadavers. Rotation of the pelvis was simulated by rotating the X-ray tube without changing the position of the patient.
Results: In dorsoventral projections, the Norberg angle value is increased by 3.2 to 5.8%. Due to rotation along the long axis, the Norberg angle increases on one side and decreases on the contralateral side by approximately the same value. Rotation of greater than 2 degrees in a lateral direction is visually perceived by the observer as tilted. Rotation of the projection in the caudoventral to craniodorsal direction causes mild increase of the Norberg angle (≤1%), while rotation of the projection in a cranioventral to caudodorsal direction causes moderate (≤2%) decrease of the Norberg angle. Rotation of less than -10 degrees (caudoventral to craniodorsal) or greater than 10 degrees (cranioventral to caudodorsal) is visually perceived as tilted.
Conclusion: Tilted images that are subjectively perceived evaluable have only little effect on the Norberg angle and are therefore probably acceptable. Dorsoventral projections can be recognized based on the proximal position of the patella and should be excluded from evaluation, as the Norberg angle value can be falsely increased.