Interobserver variability in the measurement of internal carotid stenosis

Objective: To assess interobserver variability in the measurement of caroti d stenoses from digital subtraction angiograms displayed in different ways (nonmagnified or magnified, white or black arteries); and to compare human readers with computer-generated densitometric measurements of vessel stenos is. Methods: Digital subtraction angiograms of 20 proximal internal carotid artery stenoses were laser printed in the following ways: (1) Nonmagnified white artery on a black background; (2) Magnified white artery on a black background; (3) Nonmagnified black artery on a white background; (4) Magnif ied black artery on a white background. This resulted in 80 images of inter nal carotid artery stenoses. These stenoses were independently measured by 4 radiologists using the North American Symptomatic Carotid Endarterectomy Trial method. A computer-generated densitometric measurement of the black n onmagnified images was also obtained. Results: The most reliable stenosis m easurements were obtained from the nonmagnified black and white artery imag es. The interobserver variability in the measurement of internal carotid st enoses using these images was quite small. Variability increased with the u se of magnification. The computer generated stenosis measurements were cons istently much higher than those of the radiologists. Conclusion: There was significant variability in measurements made from magnified images and betw een human readers and computer-generated measurements. This has great clini cal significance. Readers of digital angiographic images must determine the most reliable, reproducible images generated by their equipment, as these measurements significantly affect treatment of patients with symptomatic in ternal carotid artery stenosis.