ABSTRACT
OBJECTIVES: To investigate prospectively the repeatability of pancreatic perfusion measurements using arterial spin labelling (ASL) and to determine the increase in perfusion due to secretin stimulation. MATERIAL AND METHODS: An (FAIR)-TrueFISP ASL sequence was applied to determine the perfusion of the pancreatic head in a 3T MRI scanner. Ten healthy volunteers (four men, six women: mean age 28.5 ± 4.6 years; age range 25-40 years) were investigated twice within 1 week. The inter-individual variability was calculated using the standard deviation. Intra-individual agreement between the first and second scan was estimated using the Pearson correlation coefficient. A paired Wilcoxon rank-sum test was used to compare perfusion at baseline (BL) and during secretin stimulation. RESULTS: The mean BL perfusion of the pancreatic head was 285 ± 96 mL/100 g/min with an intra-individual correlation coefficient of 0.67 (strong) for repeated measurements. Secretin stimulation led to a significant increase (by 81%) in perfusion of the pancreatic head to 486 ±156 mL/100 g/min (p=0.002) with an intra-individual correlation of 0.29 (weak). A return to BL values was observed after 239 ± 92 s with a moderate intra-individual correlation coefficient of 0.42 for repeat measurements. CONCLUSION: Dynamic non-invasive ASL imaging of the pancreas permitted quantification of pancreatic perfusion in a clinically applicable setting. KEY POINTS: ⢠ASL imaging of the pancreas permitted quantification of pancreatic perfusion ⢠Secretin stimulation led to a significant increase in pancreatic perfusion ⢠The intra-individual correlation coefficient for baseline perfusion was strong for repeated measurements.
