Pituitary Photon-Counting Detector CT for Cushing Disease: Preoperative Lesion Localization, Intraoperative Findings,...

MRI fails to localize a tumor in up to 40% of cases of adrenocorticotropic hormone-dependent Cushing disease (CD), impacting patient outcomes and creating uncertainty in the diagnosis. Photon-counting detector CT (PCD-CT) is a new CT technology with better imaging performance than conventional CT. PCD-CT is a novel imaging technique for CD, especially for MR-negative cases, however, its application to CD remains unknown.

We present a cohort of consecutive patients with a preoperative PCD-CT undergoing transsphenoidal surgery for CD at a single institution from September 2023 to January 2025. All patients underwent gadolinium-enhanced pituitary MRI for baseline work-up and PCD-CT as part of their preoperative planning imaging. Both imaging modalities were used to localize the lesion and guide surgical planning. The accuracies of MRI and PCD-CT were based on confirmation of the lesion location by a combination of intraoperative visualization, pathologic confirmation, and postoperative biochemical remission, and were compared with each other via Fisher exact test.

Twenty-five patients with CD were included in this analysis. The cohort's median age was 50 years old, and 88% were women. At diagnosis, 88% had central obesity, 76% had hypertension, 48% had type 2 diabetes mellitus, and 20% had osteoporosis. Contrast-enhanced MRI accurately identified 56% of lesions, while it was negative in 24% and misleading in 20%. Dynamic contrast-enhanced PCD-CT was superior in localizing CD (P = .01), accurately identifying 92% of lesions, while it was negative in 4% and misleading in 4% of cases. Early postoperative biochemical remission was achieved in 88% of cases, while 84% remained in remission at follow-up.

PCD-CT had a high rate of adenoma localization confirmed intraoperatively, even among MR-negative cases, potentially leading to high remission and gland preservation rates. Its role in the clinical workflow of preoperative evaluation for CD requires further investigation, but current evidence supports its value in lesion localization, especially in MR-negative disease.