Optimizing CT pulmonary angiography: comparison of radiation dose, contrast volume, and image quality between standard and low-dose protocols
摘要
Pulmonary embolism (PE) is a potentially life-threatening condition that requires prompt and accurate diagnosis. Computed tomography pulmonary angiography (CTPA) is the reference standard for PE detection; however, it carries risks of radiation exposure and contrast-induced nephropathy, particularly in younger patients and those requiring repeated scans. This study aimed to evaluate whether a combined low-dose, low-contrast CTPA protocol could achieve diagnostic image quality comparable to the standard protocol.
MethodsIn this prospective comparative study, 44 patients with clinically suspected PE were enrolled. The control group (n = 22) underwent standard CTPA (120 kVp, 90 mL contrast), while the experimental group (n = 22) received an optimized protocol (100 kVp, 70 mL contrast, automatic tube current modulation, and 50% ASiR-V iterative reconstruction). Radiation dose parameters and contrast usage were recorded. Subjective image quality was graded by two blinded radiologists using a 5-point Likert scale, and quantitative metrics Hounsfield units (HU), signal-to-noise ratio (SNR), and contrast-to-noise ratio (CNR) were compared between groups.
ResultsThe optimized protocol achieved a 39.6% reduction in effective dose (2.65 ± 0.54 mSv vs. 4.39 ± 0.62 mSv; p ≤ 0.001) and a 22.2% reduction in contrast volume (70 mL vs. 90 mL). Subjective image quality was comparable between groups (p = 0.606). Quantitative analysis revealed no significant differences in SNR, CNR, or vascular attenuation (p > 0.05). Interobserver agreement was excellent (ICC = 0.91).
ConclusionsA low-dose, low-contrast CTPA protocol can substantially reduce patient radiation and contrast exposure while maintaining diagnostic image quality. This approach supports the ALARA principle and may serve as a practical, cost-effective standard for routine PE imaging, especially in resource-limited or high-risk patient settings.