Low Dose Contrast Enhanced CT Thorax Protocol: Comparison of Low Kilovoltage, Low Contrast Volume Using Iterative Reconstruction Technique with Standard Protocol.

Abstract

Background: Computed enhanced computed tomography (CECT) of the thorax is an effective imaging technique for diagnosing lung diseases. However, the increased use of CECT thorax scans has raised concerns regarding cancer risk and contrast-induced nephropathy (CIN). The iterative reconstruction (IR) method, specifically iDose4, enhances image quality (IQ) and reduces artifacts at low doses (LD). This study aimed to evaluate the image quality (IQ) and radiation dose (RD) of low-dose, low-volume (LD-LV) CECT thorax with iDose4, compared to standard dose (SD) CECT thorax (iDose4).

Methods: Group A consisted of 40 patients who underwent SD CECT thorax (120 kVp, 60 ml), while Group B included 40 patients who underwent LD-LV CECT thorax (80 kVp, 40 ml). All CECT thorax scans were performed using a 128-slice Incisive CT (Philips Healthcare Systems). A qualitative analysis of thoracic structures in both the lung and mediastinal windows was performed. Quantitative parameters, including Hounsfield units (HU) for the pulmonary artery (CTPA) and infraspinatus muscle (CTISM), noise (SD), and contrast-to-noise ratio (CNR), were also assessed. The Mann-Whitney U test and independent t-test were used to compare IQ and radiation dose between the two groups.

Results: Qualitative analysis of thoracic structures in the lung and mediastinal windows revealed no significant difference (p > 0.001) between the two groups. Quantitative parameters, such as CTPA, CTISM, and CNR, showed statistically significant differences (p < 0.001), with higher values observed in the LD-LV group compared to the SD group. The effective dose (ED) was reduced by 65.2% in the LD-LV group.

Conclusion: Our LD-LV CECT thorax protocol using iDose4 demonstrated a significant reduction in effective dose and iodine contrast volume, while maintaining image quality and enhancing diagnostic confidence.