EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Multi-Organ Fat Fraction Quantification with Photon-Counting CT: Determinants, Regional Heterogeneity, and Agreement Between Spectral Fat Maps and Attenuation-Derived Fat Fraction in 270 Patients

Flavio Spoto, Martina Roman, Giulia Vinco, Leonardo Monterubbiano, Emanuele Demozzi, Giuseppe Sala, Carmelo Cicciò, Chiara Longo, Luca Ordofendi, Andrea Santovito, Martina Castellan, Sofia Maria Avesani und 4 weitere

Journal of Clinical Medicine · 14.11.2026 · 15 (22) · 8924

DOI: 10.3390/jcm15228924 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Clinical Medicine, online erschienen am 14.11.2026. Verfasst von Flavio Spoto und 15 weiteren Autor:innen. Die Arbeit nennt 19 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background/Objectives: Photon-counting CT (PCCT) spectral fat maps allow fat quantification on routine scans. We quantified the fat fraction (FF) in the liver, spleen, pancreas, and psoas, assessed determinants and regional heterogeneity, and compared spectral FF with an attenuation-derived PDFF estimate (HU-PDFF).

Methods

In total, 270 consecutive patients underwent abdominal PCCT. On the unenhanced acquisition, 14 ROIs per patient were placed, recording spectral FF and 70 keV attenuation; no external reference standard was available. Agreement was assessed via ICC and Bland–Altman analysis.

Results

Mean FF was 4.2% (liver), 12.3% (spleen), 23.1% (pancreas), and 15.2% (psoas); 34.8% had hepatic steatosis (FF ≥ 5%). In exploratory analyses, BMI was the main independent correlate of hepatic and pancreatic FF, and psoas FF was associated with BMI, age, and male sex; splenic FF was unrelated to BMI and age. Hepatic FF showed no systematic supra-/infra-portal difference (ICC 0.97); the pancreatic tail was leaner than the body and head. Spectral FF and attenuation correlated strongly in the liver, pancreas, and psoas (r = −0.82 to −0.92), but spectral FF was 4.6 points lower than HU-PDFF in the liver and 2.7–3.4 points higher in the pancreas and psoas; liver attenuation ≤ 58.7 HU corresponded to FF ≥ 5% (AUC 0.93; cross-validated sensitivity and specificity 86%).

Conclusions

PCCT fat maps are feasible for multi-organ fat measurement. Spectral FF and HU-PDFF are highly correlated but not interchangeable because of organ-specific offsets, whose source and absolute accuracy must be established against MRI-PDFF.

Zitieren

APA

Spoto, F., Roman, M., Vinco, G., Monterubbiano, L., Demozzi, E., Sala, G., Cicciò, C., Longo, C., Ordofendi, L., Santovito, A., Castellan, M., Avesani, S. M., Cardobi, N., Robertis, R. D., D’Onofrio, M., & Foti, G. (2026). Multi-Organ Fat Fraction Quantification with Photon-Counting CT: Determinants, Regional Heterogeneity, and Agreement Between Spectral Fat Maps and Attenuation-Derived Fat Fraction in 270 Patients. Journal of Clinical Medicine, 15(22), 8924. https://doi.org/10.3390/jcm15228924

Vancouver

Spoto F, Roman M, Vinco G, Monterubbiano L, Demozzi E, Sala G, et al. Multi-Organ Fat Fraction Quantification with Photon-Counting CT: Determinants, Regional Heterogeneity, and Agreement Between Spectral Fat Maps and Attenuation-Derived Fat Fraction in 270 Patients. Journal of Clinical Medicine. 2026;15(22):8924. doi:10.3390/jcm15228924

BibTeX
@article{Spoto2026,
  author = {Spoto, Flavio and Roman, Martina and Vinco, Giulia and Monterubbiano, Leonardo and Demozzi, Emanuele and Sala, Giuseppe and Cicciò, Carmelo and Longo, Chiara and Ordofendi, Luca and Santovito, Andrea and Castellan, Martina and Avesani, Sofia Maria and Cardobi, Nicolò and Robertis, Riccardo De and D’Onofrio, Mirko and Foti, Giovanni},
  title = {Multi-Organ Fat Fraction Quantification with Photon-Counting CT: Determinants, Regional Heterogeneity, and Agreement Between Spectral Fat Maps and Attenuation-Derived Fat Fraction in 270 Patients},
  journal = {Journal of Clinical Medicine},
  year = {2026},
  volume = {15},
  number = {22},
  pages = {8924},
  doi = {10.3390/jcm15228924}
}
Kostenlos starten →