Diagnostic Accuracy of Pulmonary CT Angiography at Low Tube Voltage: Intraindividual Comparison of a Normal-Dose Protocol at 120 kVp and a Low-Dose Protocol at 80 kVp Using Reduced Amount of Contrast Medium in a Simulation Study

Abstract

The purpose of this study was to simulate pulmonary emboli (PE) and image quality at low tube energy and reduced contrast material volume in normal-dose pulmonary CT angiography (CTA) images and to analyze the diagnostic accuracy with normal- and low-dose pulmonary CTA. Normal-dose pulmonary CTA examinations using 120 kVp and 100 mL of contrast material in 10 patients with no PE were retrospectively selected. The image characteristics of an 80-kVp low-dose pulmonary CTA protocol (patient exposure reduction, 57%) with 75 mL of contrast material were simulated. Four different sets of filling defects were computer simulated in identical locations in each normal-dose and corresponding low-dose examination, equaling 783 PE in 40 normal-dose and 40 low-dose datasets. Ten normal-dose and 10 low-dose examinations contained no emboli and were used as controls. The 100 pulmonary CTA studies were randomly assessed by three readers blinded to PE location and image quality. The results were assessed by nonparametric tests and Student t tests. No difference was found between the CT protocols in terms of sensitivity, specificity, and positive and negative diagnostic likelihood ratios at all ramification levels of the pulmonary arteries (p = 0.343-1). The overall sensitivity and specificity with the normal and simulated low-dose protocols were 79.9% versus 81.3% and 98.0% versus 98.2% (p = 0.444 and 0.702), respectively. The diagnostic confidence (2.81 ± 0.39 vs 2.77 ± 0.47; p = 0.297) and overall image quality (3.92 ± 0.52 vs 3.83 ± 0.54; p = 0.216) were similar at 120 kV and 80 kV. The intraindividual comparison of diagnostic accuracy with normal-dose and simulated low-dose pulmonary CTA protocols revealed no difference under experimental conditions.