The diagnostic work-up of renovascular hypertension is still controversial. The efficacy of renal scintigraphy with technetium-99m diethylene triamine pentaacetate (DTPA) before and after captopril (scintigraphic captopril test) was evaluated in a multicentre study. All 380 hypertensive patients in the study underwent renal arteriography; 125 had renal arterial stenosis > or = 70%, and 54 had a technically successful intervention to correct the stenosis. The post-captopril study had a sensitivity of 93% and a specificity of 100% for predicting blood pressure response to intervention, if renal function was normal and a combination of quantitative parameters was applied (individual kidney uptake index < 40%, time to peak activity < 2 min or > 10 min). In the entire population renal artery stenosis > or = 70% was detected with a sensitivity of 83% and a specificity of 93% if renal function was normal. In patients with abnormal renal function the performance of the test was worse, owing to a lower specificity which could be increased by using only time parameters. The performance of the test was optimal when the post-captopril findings were examined; no improvement was achieved by evaluation of the changes induced by captopril from the baseline. The test can thus be simplified by performing only a post-captopril study for routine use: a negative test would exclude a curable form of renovascular hypertension in > 80% and a positive test would predict it in > 90% of the patients selected for suspicion of the disease.(ABSTRACT TRUNCATED AT 250 WORDS)