Comparison of excretory urography and ultrasonography for detection of experimentally induced pyelonephritis in dogs.

Pyelonephritis was experimentally induced in 10 clinically normal dogs by nephropyelocentesis and introduction of Proteus mirabilis into the randomly chosen right or left renal pelvis. Dogs were examined by nephrosonography and excretory urography before and 2 weeks after infection. The major nephrosonographic findings of pyelonephritis were renal pelvic dilatation, usually with proximal ureteral dilatation, and a hyperechoic mucosal margin line within the renal pelvis, proximal portion of the ureter, or both. In addition, at least one or more of the following were observed: generalized hyperechoic renal cortex, focal hyperechoic areas within the medulla, and focal hyperechoic or hypoechoic cortical lesions. Interpretation of excretory urograms resulted in 3 false-negative and 1 false-positive conclusions, compared with the histologic findings. Interpretation of nephrosonograms resulted in 2 false-negative and no false-positive conclusions. Of the kidneys with histologic evidence of pyelonephritis, 73% were detected by excretory urography, whereas 82% were detected by nephrosonography. Nephrosonography appeared to be useful for detection of mild to moderate cases of acute pyelonephritis that may be be interpreted as such by excretory urography.