Evaluation of the effect of pelvic tilt in the coronal plane on the Norberg angle measured in ventrodorsal radiographic views of a canine hip joint bone model

Abstract

To determine the effect of pelvic tilt on the Norberg angle (NA) measured in ventrodorsal radiographic views of a hip joint in a canine femoral-pelvic bone model. Radiographic views of a femoral-pelvic bone model with 1 articulated hip joint (without evident laxity). The bone model was placed on a radiography table. With the joint fully extended, the pelvic coronal plane was tilted from 0° in 5° increments caudally and cranially to a maximum of 30° in each direction (femoral frontal plane was maintained parallel with the radiograph cassette); a digital ventrodorsal radiographic view was obtained at each tilt position. In each image, the NA was measured 3 times by 2 observers (operators). The effect of pelvic tilt on the NA was evaluated by use of a repeated-measures analysis; within-and between-operator variances were also assessed. Within-operator repeatability was excellent for assessments of hip joint NAs in radiographic views obtained when cranial or caudal tilt was applied to the femoral-pelvic model; replicate measurement variation was < 2.5% (coefficient of variation < 5%). Mean NA values determined by the 2 operators in cranial or caudal tilt views did not differ. Mean between-operator difference in caudal tilt measurements was -0.55° (95% confidence interval, -1.69° to 0.59°) and in cranial tilt measurements was -0.13° (95% confidence interval, -1.33° to 1.06°). Cranial or caudal pelvic tilt did not affect measurement of the NA. Model-derived results indicated that pelvic tilt during acquisition of ventrodorsal radiographic views of the hip joints of a dog should not affect measurement of the NA.