Added value of photon-counting CT for triple rule-out imaging: A propensity-matched comparison with energy-integrating CT

Purpose To compare the technical performance of photon-counting detector (PCD)-CT and energy-integrating detector (EID)-CT in triple rule-out (TRO) imaging. Methods In this IRB-approved single-center study, consecutive TRO-CT patients underwent dual-source EID-CT or PCD-CT with 75 or 98.5 ​mL iopromide. Propensity score matching generated four groups: EID-75, EID-98.5, PCD-75 and PCD-98.5. Axial polychromatic images were reconstructed at 1.0 ​mm for whole-thorax and 0.6 ​mm for coronary assessment using a vascular kernel (Bv36). Volume CT dose index (CTDI vol ), contrast-to-noise ratio (CNR), and diagnostic confidence (Likert 4, excellent ) were compared across vascular territories. Diagnostic performance for coronary stenosis was assessed in patients undergoing invasive coronary angiography. Results A total of 96 subjects (mean age 62.0 ​± ​13.1 years; 62.5% female, n ​= ​24 per group) were included. TRO-CT ruled out aortic, coronary and pulmonary pathology in 69/96 (71.9%) patients. PCD-CT showed 32.3% lower CTDIvol than EID-CT (29.3 [23.5–41.4] vs 43.3 [36.6–51.0] mGy; p ​< ​0.001). PCD-98.5 yielded the highest CNR (aorta, 32.2 [25.5–44.1]; coronaries, 24.5 [19.5–31.3]; and pulmonary arteries, 36.8 [28.0–45.3]; all p ​< ​0.001). Pulmonary CNR with PCD-75 was comparable to EID-98.5 despite lower contrast volume (22.4 [19.1–28.8] vs 24.2 [20.3–29.6]; p ​= ​0.437). Segmental pulmonary diagnostic confidence increased from fair (2 [1–3]) with EID-98.5 to excellent (4 [3–4]) with PCD-98.5 (p ​< ​0.001). TRO-CT showed coronary segment-based sensitivity of 94.4% (17/18) and specificity of 93.2% (218/234), with no AUC difference between EID-CT and PCD-CT (0.961 vs 0.936; p ​= ​0.361). Conclusion PCD-CT enabled robust TRO imaging with lower radiation dose, higher multi-territory CNR, and greater segmental pulmonary diagnostic confidence, addressing key limitations of conventional TRO protocols.