Renal vein renin measurements accurately identify renovascular hypertension caused by total occlusion of the renal artery

OBJECTIVE: To assess the usefulness of indexes derived from renal vein renin measurements. DESIGN: A 12-year prospective study. SETTING: A tertiary institutional referral centre. PATIENTS AND METHODS: Between 1988 and 2000, we studied 152 consecutive hypertensive patients with a high pre-test probability of renovascular hypertension (RVH). RVH was diagnosed retrospectively on the basis of reduction in blood pressure after correction of ischaemia at follow-up. Renal vein renin measurements were used to calculate the ratios: Visch/Vctl (renal vein renin ratio; RVRR); Vctl/Viivc; (Visch - Viivc)/Viivc; (Vctl - Viivc)/Viivc, where Visch and Vctl indicate plasma renin activity (PRA) in the ischaemic and contralateral renal veins, respectively, and Viivc denotes PRA in the infrarenal inferior vena cava. A receiver operator characteristics (ROC) curve analysis was used to determine the cut-off value of renal vein renin measurement indexes that provided the best discrimination between patients with and without RVH and to identify patients with RVH caused by total occlusion of the renal artery. RESULTS: Sixty-seven patients were diagnosed as having RVH: 51 had significant renal artery stenoses (RVH non-occluded) and 16 had total renal artery occlusion (RVH occluded). Of the remaining 85 patients in whom RVH was excluded (non-RVH group), 27 had reno-parenchymal hypertension and 58 had essential hypertension. Of the renal vein renin measurement indexes, only RVRR and (Visch - Viivc)/Viivc in RVH-occluded patients differed significantly (P < 0.005) from those in the non-RVH group and showed the best performance by ROC curve analysis. This analysis also showed that, at any cut-off value, RVRR was far more accurate for identification of RVH-occluded patients than for identification of RVH non-occluded patients, both in the subgroup with unilateral and, even more so, in those with bilateral renal artery lesions. The best trade-off between sensitivity and false-positive rate was provided by cut-off values of 1.55 and 1.70 of the RVRR for identification of non-occluded and occluded RVH, respectively. CONCLUSIONS: RVRR is more useful for establishing an indication for nephrectomy in patients with renal artery occlusion than for identifying those patients with renal artery stenosis who will benefit from revascularization. In patients with RVH with bilateral renal artery lesions, lateralization of renin secretion occurs only in the presence of total renal artery occlusion. Different cut-off values are necessary for identification of non-occluded and occluded RVH.