Computed tomography perfusion parameters may aid preoperative differentiation of canine pheochromocytomas from...

To characterize CT perfusion (CTP)-derived time to highest observed adrenal tissue enhancement (T-peak) and arterial blood flow (AF), evaluate associations with triple-phase CT findings, and assess the diagnostic performance of T-peak, AF, and the tumor-to-contralateral adrenal AF ratio for differentiating pheochromocytomas from adrenocortical tumors.

Historical CTP data from 5 healthy Beagles served as controls. Client-owned dogs evaluated from November 21, 2024, through May 31, 2026, underwent triple-phase CT, CTP, adrenalectomy, and histopathology. Time to highest observed adrenal tissue enhancement was the interval from contrast administration to the greatest sampled adrenal attenuation; AF was calculated with a single-input maximum-slope model.

20 dogs with 22 tumors (8 adrenocortical adenomas, 7 adrenocortical carcinomas, and 7 pheochromocytomas [PHEOs]) were included. Time to highest observed adrenal tissue enhancement was longer in both adrenocortical tumor groups than in PHEOs; a post hoc cutoff < 36.0 seconds yielded apparent 100% sensitivity and specificity. Arterial blood flow was lower in both adrenocortical tumor groups than in controls and higher in PHEOs than in adrenocortical carcinomas. Arterial blood flow had an area under the curve of 0.923; a cutoff > 100.60 mL/min/100 mL yielded apparent 85.7% sensitivity and 100% specificity. Among 17 dogs with unilateral tumors, the tumor-to-contralateral AF ratio completely separated PHEOs from adrenocortical tumors. Arterial blood flow correlated positively with CT wash-in and washout indices, whereas T-peak correlated negatively with arterial- and venous-phase percentage enhancement washout.

Time to highest observed adrenal tissue enhancement and AF differed among adrenal tumor types and were associated with CT enhancement characteristics.

Computed tomography perfusion may provide adjunctive information for preoperative adrenal tumor characterization, but the proposed thresholds require validation in larger independent cohorts.