Comparison of computed tomographic angiography and ultrasonography for the detection and characterization of portosystemic shunts in dogs.

The purpose of this retrospective study was to compare the accuracy of computed tomographic angiography (CTA) and abdominal ultrasonography in detecting and characterizing portosystemic shunts (PSS) in dogs. Medical records of 76 dogs that underwent CTA and/or abdominal ultrasonography suspected to have PSS were reviewed. Presence or absence, and characterization of PSS (when present) on CTA were reviewed by a board-certified veterinary radiologist that was blinded to the clinical findings. The abdominal ultrasonography findings were reviewed from the medical records. Visualization and description of the origin and insertion of PSS on CTA and abdominal ultrasonography were related with laboratory, surgical, or mesenteric portographic confirmation of the presence or absence of PSS. The sensitivity for detection of PSS with CTA (96%) was significantly higher than abdominal ultrasonography (68%; P < 0.001). The specificities for CTA and abdominal ultrasonography were 89% and 84%, respectively (P = 0.727). Computed tomographic angiography detected the correct origin in 15 of 16 dogs and correct insertion in 15 of 16 dogs with congenital PSS. Abdominal ultrasonography detected the correct origin in 24 of 30 dogs and correct insertion in 20 of 33 dogs with congenital PSS. Multiple acquired PSS were seen in four of five dogs and in one of six dogs on CTA and abdominal ultrasonography, respectively. Computed tomographic angiography was 5.5 times more likely to correctly ascertain the presence or absence of PSS when compared to abdominal ultrasonography (P = 0.02). Findings indicated that CTA is a noninvasive diagnostic modality that is superior to abdominal ultrasonography for the detection and characterization of PSS in dogs.