Spectral CT in patients with colorectal cancer: quantitative contrast lesional assessment of mucinous and non-mucinous peritoneal metastases using virtual monoenergetic imaging.

PURPOSE: To objectively evaluate quantitative image parameters-signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR)-of peritoneal metastases (PMs) from colorectal cancer (CRC) on virtual monoenergetic images (VMIs) reconstructed from dual-layer spectral CT (DLCT) technology.

MATERIALS AND

METHODS

This prospective single-center study included consecutive patients scheduled for cytoreductive surgery (CRS) with hyperthermic intraperitoneal chemotherapy (HIPEC) between September 1, 2023, and May 1, 2025. All patients underwent contrast-enhanced DLCT < 7 days before surgery. SNR, CNR between PMs and mesenteric fat (CNR MF), CNR between PMs and adjacent organ (CNR AO) were calculated on VMIs ranging from 40 to 200 keV. PMs were classified as mucinous (mPMs) or non-mucinous (nmPMs) based on histopathology. PM conspicuity and sharpness were also evaluated between 40 and 120 keV.

RESULTS

A total of 92 PMs (24 mPMs, 68 nmPMs) in 55 patients were analyzed. For both mPMs and nmPMs, attenuation, SNR, CNR MF, CNR AO, were highest at 40 keV (42.47 ± 23.81, 3.00 ± 2.01, 12.36 ± 3.78, 7.30 ± 3.71, and 131.16 ± 50.17, 9.88 ± 5.38, 21.89 ± 10.57, 5.62 ± 3.07, respectively) and decreased with increasing energy levels. In nmPMs, CNRAO and SNR declined significantly from at 50 keV onward (p < 0.05). Image noise remained stable across the 40-200 keV range. PM conspicuity and sharpness were significantly higher at 40 keV (p < 0.05).

CONCLUSION

Low-energy VMIs obtained from spectral CT significantly improve SNR and CNR of PMs, particularly for nmPMs, supporting their integration into preoperative reading protocols.