Related Experiment Video
Updated: Sep 14, 2025
Robot-Assisted Transcanal Endoscopic Ear Surgery for Congenital Cholesteatoma
Published on: December 15, 2023
Diagnostic performance of dual-layer spectral detector CT in differentiating middle ear cholesteatoma from chronic
Shuang Yuan Zhou1,2, Heng Liu3,4,5, Lu Jiang3,4,5
1Department of Radiology, Xiangya Hospital, Central South University, Changsha, China. zsy_amory1003@163.com.
Objective:
To investigate the diagnostic performance of dual-layer spectral detector CT (DLCT) in differentiating middle ear cholesteatoma and chronic suppurative otitis media (CSOM).
Materials And Methods:
This prospective, institutional review board-approved study consecutively enrolled patients with suspected middle ear cholesteatoma or CSOM from January to July 2024. All participants underwent DLCT scanning before surgery. Clinical average hearing threshold (AHT) was collected. Four blinded readers, with 2, 3, 11, and 13 years of experience, independently provided diagnoses, based on high-resolution CT (HRCT) or HRCT combined spectral images (integrated mode). Virtual mono-energetic images (VMIs) at 40 keV, spectral attenuation slope (λHU), and effective atomic number (Zeff) were measured. Intraoperative or pathological findings served as reference standards. Receiver operating characteristic (ROC) curve analysis was used.
Results:
Sixty-six participants (median age, 47.0 [33.0-55.0] years; 23 men, 33 with cholesteatoma) were included. The sensitivity of the integrated mode was 87.9%, significantly higher than HRCT alone (68.2%; p < 0.001), while specificities were 73.5% and 79.5% (p = 0.169), respectively. Diagnostic confidence scores of the integrated mode were significantly higher (p < 0.001). The Zeff, λHU, and VMI 40 keV of cholesteatoma were significantly lower (p < 0.001) than those of CSOM. AUCs of combined spectral model, integrated model and integrated model + AHT were 0.820 (95% CI: 0.744-0.882, p = 0.01), 0.817 (95% CI: 0.744-0.891, p = 0.017),0.853 (95% CI: 0.789-0.917, p = 0.001), respectively, all significantly higher than that of conventional HRCT (AUC = 0.741; 95% CI: 0.657-0.813).
Conclusion:
HRCT combined with spectral images enhances the sensitivity of cholesteatoma diagnosis, and DLCT multi-parameters can quantitatively distinguish cholesteatoma from CSOM.
Key Points:
Question HRCT is limited in differentiating middle ear cholesteatoma and CSOM. Findings Integrating HRCT and spectral images in DLCT increases the diagnostic sensitivity for cholesteatoma. Clinical relevance DLCT can provide valuable multi-parametric images. Its combination with HRCT significantly improves the diagnostic performance of middle ear cholesteatoma and CSOM, which may help clinicians make reasonable treatment plans.
