Abstract
OBJECTIVE: To investigate the visualization of mediastinal lymph nodes during thoracic CT employing a multiphasic contrast media (CM) protocol. METHODS: Institutional review board approved retrospective study consisting of 300 patients with known chest malignancy. Patients were allocated to one of two CM protocols: Protocol A, consisted of dual bolus (Phase 1:100 ml CM followed by 100 ml saline chaser) i.v. injected at 2.5 ml s(-1); Protocol B employed 100 ml of CM using a multiphasic injection protocol (Phase 1 and 2:60 ml contrast and saline, followed by Phase 3 and 4:40 ml contrast and saline injected at 2.5 ml s(-1)) with a fixed scan delay of 70 s for each acquisition. Attenuation profiles of the thoracic arteries and veins were calculated as well as the arterio-venous contrast ratios (AVCR). Receiver operating characteristic (ROC), visual grading characteristic (VGC), and Cohen's kappa analysis were assessed. RESULTS: Arterial opacification was up to 24% (p < 0.032) higher in protocol B than A, whereas, in the veins it was significantly lower in protocol B than A, with a maximum reduction of up to 84% (p < 0.0001). There was no statistical significance between the central and peripheral pulmonary arteries [>263 Hounsfield units (HU)] in each protocol. Protocol B, demonstrated significant improvement in AVCR at various anatomical sites (p < 0.002). Radiation dose was significantly reduced in protocol B compared to A (p < 0.004). Both ROC and VGC demonstrated significantly higher Az score for protocol B compared to A (p < 0.0001) with an increased inter reader agreement from poor to excellent. CONCLUSION: Employing a multiphasic CM protocol significantly improves opacification of the thoracic vasculature and visualization of mediastinal lymph nodes during thoracic CT. ADVANCES IN KNOWLEDGE: Uniform opacification between thoracic arteries and veins increases the delineation between vasculature and lymph nodes, reduces radiation dose when employing a multiphase contrast media injection protocol.