Background: Computed tomography (CT) is widely used to diagnose peritoneal metastases (PM), with debated accuracy. Dual-energy CT (DECT) may improve accuracy, yet its diagnostic performance is still unknown. We explored the potential of DECT for PM detection and quantification. Materials and methods: We retrospectively included patients undergoing staging DECT for cancers with a high risk of peritoneal involvement, followed by staging laparoscopy/laparotomy, which served as the reference standard. Nine readers with varying experience levels (three expert, three intermediate, and three inexpert) reviewed two sets of images, separated by ≥ 60 days, considering the presence/absence of PM, abdominal region(s) involved, and calculated the radiological peritoneal cancer index (PCI). The first set included contrast-enhanced delayed-DECT scans reconstructed as virtual 120-kVp images; the second set also included virtual monoenergetic, 40-keV images and iodine maps. Performance metrics, receiver operating characteristic (ROC) analysis, McNemar, DeLong, and Wilcoxon tests were applied. Results: Twenty patients (mean age 64.2 years; 12 females) were included, 10 with PM. At per-patient analysis, the addition of monoenergetic 40-keV images and iodine maps slightly increased the performance and improved inter-reader agreement, with significant benefit for inexperienced readers only (p = 0.010). Per-region analysis demonstrated a significant advantage with an area under the ROC curve ranging from 0.709 to 0.766 (p < 0.001), confirmed for each reader group; in addition, the inter-reader agreement significantly improved. Quantitative analysis showed a reduction in the differences between CT results and surgical PCI by DECT (4 ± 12 versus 2 ± 9, p < 0.001). Conclusion: DECT-derived reconstructions in the delayed-phase enhanced PM detection and quantification. Relevance statement: Delayed-phase DECT reconstruction showed superior accuracy over conventional CT in detecting and quantifying peritoneal metastases. These findings could help establish a new standard CT protocol for malignancies with peritoneal tropism. Key Points: CT is the most widely used technique for assessing peritoneal metastases. The accuracy of CT for peritoneal metastases is debated; dual-energy CT shows promise. In our study, delayed-phase dual-energy CT provided significant advantages for all readers. © The Author(s) 2025.

Bagnacci, G., Perrella, A., Di Meglio, N., Di Martino, V., Sansotta, L., Gentili, F., et al. (2025). Improved detection and quantification of peritoneal metastases using delayed contrast-enhanced dual-energy CT scans. EUROPEAN RADIOLOGY EXPERIMENTAL, 9(1) [10.1186/s41747-025-00627-5].

Improved detection and quantification of peritoneal metastases using delayed contrast-enhanced dual-energy CT scans

Bagnacci, Giulio
;
Perrella, Armando
;
Di Meglio, Nunzia
;
Di Martino, Vito;Sansotta, Letizia;Gentili, Francesco;Guerrini, Susanna;Ruggeri, Silvia;Intrieri, Cristina;Piccioni, Stefania;Marrelli, Daniele;Mazzei, Maria Antonietta
2025-01-01

Abstract

Background: Computed tomography (CT) is widely used to diagnose peritoneal metastases (PM), with debated accuracy. Dual-energy CT (DECT) may improve accuracy, yet its diagnostic performance is still unknown. We explored the potential of DECT for PM detection and quantification. Materials and methods: We retrospectively included patients undergoing staging DECT for cancers with a high risk of peritoneal involvement, followed by staging laparoscopy/laparotomy, which served as the reference standard. Nine readers with varying experience levels (three expert, three intermediate, and three inexpert) reviewed two sets of images, separated by ≥ 60 days, considering the presence/absence of PM, abdominal region(s) involved, and calculated the radiological peritoneal cancer index (PCI). The first set included contrast-enhanced delayed-DECT scans reconstructed as virtual 120-kVp images; the second set also included virtual monoenergetic, 40-keV images and iodine maps. Performance metrics, receiver operating characteristic (ROC) analysis, McNemar, DeLong, and Wilcoxon tests were applied. Results: Twenty patients (mean age 64.2 years; 12 females) were included, 10 with PM. At per-patient analysis, the addition of monoenergetic 40-keV images and iodine maps slightly increased the performance and improved inter-reader agreement, with significant benefit for inexperienced readers only (p = 0.010). Per-region analysis demonstrated a significant advantage with an area under the ROC curve ranging from 0.709 to 0.766 (p < 0.001), confirmed for each reader group; in addition, the inter-reader agreement significantly improved. Quantitative analysis showed a reduction in the differences between CT results and surgical PCI by DECT (4 ± 12 versus 2 ± 9, p < 0.001). Conclusion: DECT-derived reconstructions in the delayed-phase enhanced PM detection and quantification. Relevance statement: Delayed-phase DECT reconstruction showed superior accuracy over conventional CT in detecting and quantifying peritoneal metastases. These findings could help establish a new standard CT protocol for malignancies with peritoneal tropism. Key Points: CT is the most widely used technique for assessing peritoneal metastases. The accuracy of CT for peritoneal metastases is debated; dual-energy CT shows promise. In our study, delayed-phase dual-energy CT provided significant advantages for all readers. © The Author(s) 2025.
2025
Bagnacci, G., Perrella, A., Di Meglio, N., Di Martino, V., Sansotta, L., Gentili, F., et al. (2025). Improved detection and quantification of peritoneal metastases using delayed contrast-enhanced dual-energy CT scans. EUROPEAN RADIOLOGY EXPERIMENTAL, 9(1) [10.1186/s41747-025-00627-5].
File in questo prodotto:
File Dimensione Formato  
s41747-025-00627-5 (1).pdf

accesso aperto

Descrizione: Articolo
Tipologia: PDF editoriale
Licenza: Creative commons
Dimensione 3.38 MB
Formato Adobe PDF
3.38 MB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11365/1302394