Objective: Renal perfusion is a key parameter of kidney function and the decrement of renal perfusion is a marker of target organ damage caused by hypertension. Detecting these changes in renal perfusion could help to manage antihypertensive therapy and evaluate patients[Combining Acute Accent] prognosis. Measurement of renal perfusion by MRI arterial spin labelling (ASL) is a non-invasive and non-time-consuming method without the need to inject any contrast agent. This study examined reproducibility of renal perfusion measured by 1.5 Tesla MRI. Design and method: Renal perfusion was measured by ASL technique using an 1.5 Tesla MRI scanner. Subjects were scanned 3 times at two different days in an interval of two weeks to assess the test-retest reproducibility. Renal perfusion was automatically calculated for the cortex and medulla of the kidney by dedicated software. Results: 14 patients were included with mean age 48.9 ± 12.7 and mean office blood pressure 132 ± 16/82 ± 10mmHg and estimated glomerular filtration rate> 60 ml/min/1.73m2. The change of the mean total, cortical and medullary renal perfusion from the first examination to the second examination was 0.37 ± 13/0.62 ± 18/0.00 ± 12 ml/min/100 g kidney weight (p = 0.915/p = 0.898/p = 0.998), respectively. There was also no significant difference between the three renal perfusion measurements at one time point. For clinical trials these data indicate that to detect a 5% (10%) difference of cortical renal perfusion due to an intervention (vs placebo) only 38 (14) patients are required in face of the observed standard deviation for the change in renal perfusion. Conclusions: The inter and intra-session reproducibility of cortical renal perfusion assessed by MRI ASL 1.5 Tesla is excellent and small study cohorts can be used for examination of renal perfusion.