Introduction

Computed tomographic MIP images have been shown to be superior to standard axial images for the detection of pulmonary nodules and assessment of vascular structures in people and also reduce the effects of reviewer experience.

Objectives

To compare standard axial CTEU images to MIP images for the diagnosis of canine ectopic ureters, assess MIP slab thickness, and evaluate inter-observer variability and confidence.

Methods

Two blinded, board-certified veterinary radiologists and a small animal internist independently reviewed randomized and anonymized standard axial post-contrast CTEU images as well as two-, five-, and 10-slab thick MIP images of 24 dogs suspected of having ectopic ureters. Each reviewer classified each ureter as normal or ectopic and recorded their level of confidence in their diagnosis on a scale from 1–3, which were compared to the surgical or cystoscopic diagnosis.

Results

There was no significant difference in a diagnosis between viewing methods (p=0.1797–0.3173) with significant agreement between reviewer diagnoses and the surgical/cystoscopic findings (p=<0.0001). There was good inter-observer agreement for all viewing methods (p=0.0363 to <0.0001). Reviewer confidence in their diagnosis was significantly higher for the two- (p=0.008) and five- (p=0.0009) slab thick MIP images.

Conclusion

CTEU standard axial and MIP images appear to have similar accuracy in diagnosing ectopic ureters in dogs with good inter-observer agreement regardless of experience. Two- and five-slab thick MIP CTEU images improve reviewer confidence and thus should be considered when assessing canine patients for ectopic ureters.