A new “angle” on aortic neck angulation measurement

AbstractObjectiveInfrarenal aortic neck angulation is one of the most powerful predictors of endovascular aneurysm repair failure. Whereas the “gold standard” to measure this angle is three-dimensional (3D) reconstruction and centerline measurement, many surgeons rely on estimations of angulation based on two-dimensional (2D) views of computed tomography imaging. Unfortunately, these views do not accurately represent the true angle, particularly if aortic angulation is oblique to the standard views. In response to this issue, our group has developed a novel trigonometric formula that uses coronal and sagittal measured angles to calculate the true angle. The purpose of this study was to compare the paired angle formula with 3D centerline measurements for estimating true aortic neck angulation. MethodsFifty randomly selected patients treated by endovascular aneurysm repair at The Ottawa Hospital between 2010 and 2015 were studied. The 3D centerline aortic neck angle measurements were made by a radiology staff physician. The paired angle formula was applied by a vascular surgeon, resident, and student using 2D coronal and sagittal angles from computed tomography imaging to estimate the true angle. ResultsThe average age was 78 years; 74% of patients were male, and average preoperative aneurysm diameter was 5.7 cm. The mean neck length was 1.9 cm (1.1-3.2 cm), and mean neck angulation calculated by the gold standard measurements was 39 degrees (2-84 degrees). Linear regression demonstrated strong association between 3D measurements and the paired angle formula, with correlations comparable to the intraobserver variability (intraclass correlation coefficient values range, 0.74-0.87). The average user estimates deviated minimally from the gold standard (absolute difference, 6 degrees; 95% confidence interval, 4-8 degrees) without systemic bias. The paired angle formula accurately ruled out severe angulation >60 degrees with an overall negative predictive value of >99%. Compared with isolated 2D measurements, application of the paired angle formula significantly decreased the false-negative rate of unappreciated severe angulation >60 degrees from 4.8% to 0.7% ( P = .032). ConclusionsThe paired angle formula detects significantly more severe angles than isolated 2D measurements and can accurately rule out severe angulation >60 degrees compared with the 3D measurements. The implementation of this angle estimation method is a useful adjunct in the measurement of aor