Photon-counting CT versus energy-integrating CT of the chest in acute respiratory distress syndrome: A retrospective observational study of image quality and radiation exposure
摘要
Chest CT is central to the diagnosis and monitoring of acute respiratory distress syndrome (ARDS). Photon-counting detector CT (PCD-CT) may improve image quality and enable radiation dose optimization compared with Energy-integrating detector CT (EID-CT). However, evidence in critically ill ARDS patients undergoing contrast-enhanced chest imaging is limited. The purpose of this study was to intraindividually compare image quality and radiation exposure between PCD-CT and EID-CT in ARDS patients.
MethodsThis retrospective study included patients with ARDS who underwent both contrast-enhanced chest PCD-CT and EID-CT between April 2023 and May 2025. CT dose index volume (CTDIvol), dose–length product (DLP), effective and organ doses were extracted from dose monitoring software. Subjective image quality and diagnostic acceptability were rated independently by two radiologists using a Likert scale; inter-reader agreement was assessed with Cohen’s κ. Objective image quality was assessed using signal-to-noise ratios in representative anatomical regions. Intraindividual comparisons were performed using the Wilcoxon signed-rank test.
ResultsAmong 531 ARDS patients, 80 (15.1%) underwent chest PCD-CT, including 59 (73.8%) contrast-enhanced examinations (median age: 51.7 years (IQR: 41.6–59.5); 62.7% male). Intraindividual imaging was performed in 32 patients (54.2%). PCD-CT demonstrated significantly higher subjective and objective image quality than EID-CT (p < 0.005), with moderate inter-reader agreement. Median CTDIvol and DLP were not significantly different between PCD-CT (8.2 mGy; 435 mGycm) and EID-CT (7.7 mGy; 297 mGycm).
ConclusionIn contrast-enhanced chest CT of ARDS patients, PCD-CT provides superior image quality without a significant difference in radiation dose, potentially facilitating more precise diagnostic assessment in this critically ill population.
Trial registrationDRKS-ID: DRKS00037916, 15 September 2025, retrospectively registered.
Graphical abstract