Functional assessment of canine kidneys after acute vascular occlusion on Gd-DTPA-enhanced dynamic echo-planar MR imaging.

Abstract

To assess the alteration in renal transit of gadolinium diethylenetriaminepentaacetic acid (Gd-DTPA) in dog kidneys after acute vascular occlusion on dynamic echo-planar imaging (EPI). Dynamic 240-ms EPI series (repetition time/echo time/inversion time [TR/TE/TI] = 3000/42.1/100 ms) of the midcoronal plane of both kidneys of dogs anesthetized by intravenous administration of phenobarbital sodium and ketamine hydrochloride were obtained before and after ligation of the left renal vein (n = 6) or artery (n = 6) for 40 minutes after a 2-second-rate bolus injection of a 0.05 mmol/kg dose of Gd-DTPA. Renal Gd-DTPA transit was analyzed on the time-DeltaR2* curves in each layer of the outer cortex (OC), juxtamedullary cortex and outer zone of the medulla (JMC-OM), and the inner zone of the medulla (IM). The results were compared with those in six normal animals and those of a fast gradient-echo T1-weighted dynamic study performed in other vein- (n = 6) or artery- (n = 6) occluded animals and six normal animals. The histopathological basis of the altered Gd-DTPA transit was also evaluated. The dynamic EPI showed rapid Gd-DTPA transit through each of the five concentric layers, with three separate peaks on the time-DeltaR2* curves. The vein-occluded kidneys showed immediate swelling, with a significant increase in the cross-sectional area proportion of the JMC-OM layer compared with normals (32% +/- 2% vs 24% +/- 2%, P < 0.0001) and intensely congested capillaries, tubular, obliterated material, and gradual and persistent enhancement of the OC and JMC-OM layers but poor Gd-DTPA migration to the IM layer. The artery-occluded kidneys showed a significant reduction in the entire cross-sectional area compared with normals (1352 +/- 69 vs 1432 +/- 47 mm(2), P < 0.05) and poor enhancement, with significant decreases in the area under the time-DeltaR2* curve of the OC and JMC-OM layers compared with the vein-occluded kidneys (79 +/- 50 vs 324 +/- 108 and 82 +/- 42 vs 326 +/- 113, respectively; both P < 0.0001), despite minimal histological damage. In both models, the nonaffected kidneys showed significant increases in the area under the time-DeltaR2* curves compared with baseline. The time course of vascular and tubular Gd-DTPA transit was more detailed by the EPI study than by the T1-weighted imaging study. Echo-planar imaging has an excellent ability to follow the rapid, renal Gd-DTPA transit through the regional anatomy of the canine kidney. After venous occlusion, the JMC-OM layer may be the most affected site, primarily causing renal swelling and interruption of tubular Gd-DTPA transit and concentration. In contrast, an initial block of vascular Gd-DTPA inflow is the primary effect of arterial occlusion. Nonaffected kidneys seem to compensate by increasing excretion of Gd-DTPA.