Technical Note JOURNAL OF MAGNETIC RESONANCE IMAGING 39: (2014)

JOURNAL OF MAGNETIC RESONANCE IMAGING 39:448 454 (2014) Technical Note Assessment of Exocrine Pancreatic Function by Secretin-Stimulated Magnetic Resonance Cholangiopancreaticography and Diffusion-Weighted Imaging in Healthy Controls Gaute K. Wathle, MD, 1 Erling Tjora, MD, 2,3 Lars Ersland, MSc, PhD, 4,5 Georg Dimcevski, MD, PhD, 6,7 Øyvind O. Salvesen, MSc, PhD, 8 Anders Molven, MSc, PhD, 9,10 Pål R.Njïlstad, MD, PhD, 2,3 and Ingfrid S. Haldorsen, MD, PhD 1,5 * Purpose: To characterize and quantify exocrine pancreatic function by secretin-stimulated magnetic resonance cholangiopancreaticography (s-mrcp) and diffusionweighted imaging (DWI) in healthy subjects and compare these findings to morphological features, ie, pancreatic volume and secretin-stimulated peak bicarbonate concentration measured in pancreatic juice. Materials and Methods: Pancreatic magnetic resonance imaging (MRI) (1.5 T) was performed in 20 healthy volunteers among which 10 underwent gastroscopy with duodenal intubation. MRI included T2-weighted imaging and DWI acquired before and 1, 5, 9, and 13 minutes after 1 Department of Radiology, Haukeland University Hospital, Bergen, 2 Department of Pediatrics, Haukeland University Hospital, Bergen, 3 KG Jebsen Center for Diabetes Research, Department of Clinical Science, University of Bergen, 4 Department of Clinical Engineering, Haukeland University Hospital, Bergen, 5 Section for Radiology, Department of Clinical Medicine, University of Bergen, 6 Department of Medicine, Haukeland University Hospital, Bergen, 7 Institute of Medicine, University of Bergen, 8 Unit for Applied Clinical Research, Department of Cancer Research and Molecular Medicine, Norwegian University of Science and Technology, Trondheim, 9 Section for Pathology, Gade Institute, Department of Clinical Medicine, University of Bergen, 10 Department of Pathology, Haukeland University Hospital, Bergen, Contract grant sponsor: Western Norway Regional Health Authority; Contract grant sponsor: University of Bergen; Contract grant sponsor: Research Council of Norway; Contract grant sponsor: Innovest; Contract grant sponsor: KG Jebsen Foundation; Contract grant sponsor: European Research Council (to P.R.N.); Contract grant sponsor: Research Funds at the Department of Radiology at Haukeland University Hospital. *Address reprint requests to: I.S.H., Department of Radiology, Haukeland University Hospital, 5021 Bergen, E-mail: ingfrid.haldorsen@helse-bergen.no Received June 8, 2012; Accepted March 13, 2013 DOI 10.1002/jmri.24167 View this article online at wileyonlinelibrary.com. secretin administration. Secreted pancreatic juice volumes were calculated based on the sequential T2- weighted images and pancreatic volumes and apparent diffusion coefficient (ADC) values were estimated. Results: The mean pancreatic secretion rate declined from 9.5 ml=min at 1 5 minutes (postsecretin) to 2.9 ml=min at 9 13 minutes. Pancreatic head ADC values significantly increased from baseline (1.29 10 3 mm 2 =s) to 1 minute postsecretin (1.48 10 3 mm 2 =s) (P ¼ 0.003). Secreted pancreatic juice volume at 1 minute after secretin correlated positively with peak bicarbonate concentration (n ¼ 10, P ¼ 0.05). Conclusion: Secretin-stimulated MRCP and DWI can characterize and quantify exocrine pancreatic function in healthy subjects. These imaging methods may prove relevant for patients with exocrine pancreatic dysfunction. Key Words: pancreas; diffusion-weighted MRI; secretinstimulated MRCP; exocrine pancreatic function J. Magn. Reson. Imaging 2014;39:448 454. VC 2013 Wiley Periodicals, Inc. PANCREATIC EXOCRINE DYSFUNCTION is a hallmark of various pancreatic diseases, eg, chronic pancreatitis and cystic fibrosis, and is also commonly reported in diabetes (1 3). Traditionally, pancreatic exocrine dysfunction is assessed by indirect methods with biochemical tests of the feces, eg, fecal elastase-1 and chymotrypsin (4). These tests, however, are not sensitive enough to diagnose slight to moderate exocrine pancreatic insufficiency (4). At present, duodenal intubation with duodenal juice collection after stimulation with secretin is considered the most accurate functional test of the exocrine pancreas (5). Unfortunately, this test is hampered by lack of test standardization and is uncomfortable for the patient due to prolonged duodenal intubation (5,6). VC 2013 Wiley Periodicals, Inc. 448

Exocrine Pancreatic Function Assessed by MRI 449 Magnetic resonance imaging (MRI) is a noninvasive diagnostic imaging method without ionizing radiation and provides detailed information of pancreatic parenchymal and ductal morphology with potential relevance for pancreatic exocrine function. Recently, secretinstimulated magnetic resonance cholangiopancreaticography (s-mrcp) has gained increasing interest as a noninvasive imaging method enabling visualization and quantification of various aspects of pancreatic exocrine function (7 10). Furthermore, diffusion-weighted imaging (DWI), a novel functional MRI technique whereby the contrast is derived from random motion of water molecules in the tissue, has been applied in an attempt to visualize functional characteristics of the pancreas in chronic pancreatitis related to alterations in secretin response (11). These advanced pancreatic imaging methods have potential value in patients with exocrine pancreatic dysfunction and in patients with pancreatic disease in general (8 12). The aim of this study was to characterize and quantify exocrine pancreatic function with s-mrcp in relation to pancreatic morphology in healthy subjects. We further aimed to compare pancreatic imaging findings to biochemical test results based on an endoscopic aspiration-based bicarbonate test of the same subjects. MATERIALS AND METHODS Study Population This study was performed under Institutional Review Board-approved protocols with written informed consent from the subjects. Twenty healthy volunteers (10 females, 10 males) with a mean age of 36 years were included and underwent an abdominal MRI. None of the subjects reported a former history of chronic abdominal pain, gastrointestinal problems, or alcohol consumption that affected daily life. Three of the study subjects were children (aged 8 12 years), whereas 17 were adults (aged 18 65 years). Ten of the adults also underwent gastroscopy with duodenal juice sampling. Thirteen subjects had normal body mass index (BMI), whereas seven adults were overweight or obese with BMIs of 25 31.2. The subjects were recruited among relatives of the staff at the hospital. Endoscopic Exocrine Pancreatic Function Test Ten subjects had gastroscopy with duodenal intubation using a rapid endoscopic aspiration-based test (13). The participants fasted >8 hours and the procedure commenced 25 minutes after administration of intravenous (IV) injection of secretin (1 CU=kg, maximum dose 70 CU; Secrelux, Sanochemia, Neuss, Germany). The duodenoscope was placed into the stomach; after a luminal examination, all gastric fluid was aspirated and discarded. The tip of the duodenoscope was then placed close to the ampulla of Vater 30 minutes after secretin stimulation, and three samples of duodenal juice were collected, each sampling lasting for 5 minutes (Fig. 1). The highest concentration of bicarbonate among these three samples was considered the peak bicarbonate concentration. Median interval between endoscopic pancreatic function test and MRI was 224 (range, 82 308) days. MRI Protocol We performed MRI after >4 hours fasting on a 1.5-T Siemens Avanto running Syngo MR B17 (Erlangen, Germany) using a 24-channel spine matrix coil and a six-channel body coil. A water-containing bag (serving as reference phantom for water) was placed parallel to the individual s torso in order to ensure the presence of voxels containing pure water for calculation of fluid volumes in the bowel. Details regarding the imaging protocol are given Fig. 1. DWI with b-values of 50 and 800 sec=mm 2 was acquired and apparent diffusion coefficient (ADC) maps generated by calculating the slope of the logarithmic decay curve for signal intensity against b-value with use of the MRI software (Syngo, Siemens). Axial DWI and coronal T2-weighted imaging (for estimation of secreted pancreatic volumes) were acquired prior to administration of IV secretin and 1, 5, 9, and 13 minutes postsecretin (Fig. 1). Secretin was administered over a period of 3 minutes. Image Analysis and Quantification Standard anatomical images including MRCP sequences were reviewed by two experienced radiologists with >5 and >10 years experience with pancreatic imaging, respectively. The presence of pancreatic cysts, pancreatic duct anomalies, or other pathological findings was recorded. All measurements of pancreatic volume and ADC value and calculation of secreted pancreatic juice were performed by the same radiologist with >5 years experience in this field. The images were read and analyzed using Agfa Impax 6.4 (Agfa Healthcare, Mortsel, Belgium), NordicICE 2.3.12 (NordicNeuroLab, Bergen Norway), and Vitrea workstation 6.2 (Vital Images, Minnetonka, MN). We calculated secreted pancreatic juice in a similar way as done in previous studies on quantitative secretin-stimulated MRCP (7,9,10,14,15). On the coronal T2-weighted images (obtained 12 and 4 minutes prior to secretin and 1, 5, 9, and 13 minutes postsecretin), a region of interest (ROI) was drawn by free hand around the duodenum and the proximal jejunum on each slice. Using a software tool, the bowel content in the ROI was segmented with 250 as the lower signal intensity limit and maximum pixel intensity in the slice as the upper intensity limit. Lower signal intensity limit was decided based on a visual inspection of which lower signal intensity limit that correctly segmented fluid filled bowel content but excluded surrounding soft tissue (Fig. 2a,b). The segmented areas of each slice were added and multiplied by slice thickness (10 mm) to obtain the total volume of the duodenal and jejunal content. Assuming that the highest signal intensity in each volume (typically seen in the water bag or the gallbladder) represented pure water signal, the calculated volumes within the bowel were

450 Wathle et al. Figure 1. Inclusion of patients. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.] multiplied by mean signal intensities of the segmented volumes divided by maximum signal intensities of the imaging volumes (10). Based on this equation, we calculated the bowel fluid volumes at the different timepoints (12 and 4 minutes before secretin and 1, 5, 9, and 13 minutes after secretin). The secreted fluid volumes after secretin were calculated by subtracting the fluid volume at 4 minutes before secretin from fluid volumes after secretin. We estimated pancreatic volumes using fatsaturated axial T1-weighted images (volume interpolated breath-hold examination [VIBE]) on which the contour of the pancreas was encircled on every slice, considering each area to represent a volume of 2.5 mm thickness (Fig. 2c). The areas were added together to estimate the pancreatic volume. We measured pancreatic ADC values in three ROIs in both the pancreatic head and body (Fig. 2d) from the diffusion series acquired before secretin and at 1, 5, 9, and 13 minutes after secretin. When pancreatic cysts were present (observed in two subjects), these were not included in the ROIs. The median value from each region was used for statistical analyses. were normally distributed. Correlations between continuous variables were analyzed using the Pearson bivariate correlation test. We analyzed ADC values before and after secretin using the Student s paired t- test. The data were analyzed using SPSS 18.0 (Chicago, IL). All reported P-values were two-sided and considered to indicate statistical significance when less than 0.05. RESULTS Pancreatic Morphological Findings In two adults one small cystic lesion (diameter 5 mm) was identified in the pancreatic head and tail, respectively, and another adult was diagnosed with pancreas divisum. None of the patients had signs of pancreatic pseudocysts. The mean pancreatic volume for all 20 subjects was 80 ml (95% confidence interval [CI], 66 95; range, 36 152) with corresponding volumes of 86 ml (95% CI, 70 102; range, 39 152) for adults (n ¼ 17) and 48 ml (95% CI, 7 89; range, 36 67) for children (n ¼ 3). Statistical Analysis We used the nonparametric Kolmogorov Smirnov test for normality to test whether the continuous variables Pancreatic Secretion After Secretin Mean peak bicarbonate concentration for the adults who underwent endoscopic aspiration-based

Exocrine Pancreatic Function Assessed by MRI 451 Figure 2. Pancreatic imaging for quantification of pancreatic secretion and for calculation of pancreatic volume and ADC value. Coronal T2-weighted imaging (a,b) with ROIs that were manually drawn encircling the duodenum and proximal jejunum. Bowel content having signal intensities >250 (marked in red) was segmented. Segmented area on image obtained 12 minutes before secretin (a) is markedly increased on the corresponding image 13 minutes after secretin (b). Axial fat-saturated T1-weighted image (c) with encircled pancreatic boundaries drawn manually was used for calculation of pancreatic volume. ADC map (d) with drawn ROIs (corresponding to the pancreatic body) was used to measure pancreatic ADC value. bicarbonate test (n ¼ 10) was 117 meq=l (95% CI, 107 127; range 97 138). Based on MRI, pancreatic secretion rate after secretin was highest initially with a mean secretion rate of 9.5 ml=min at 1 5 minutes declining to 4.8 ml=min at 5 9 minutes and 2.8 ml=min at 9 13 minutes (Fig. 3a). Mean pancreatic secretion rates during the first 5, 9, and 13 minutes were 10.7 ml=min, 8.1 ml=min, and 6.4 ml=min, respectively and were similar for adults and children. Before secretin, the secretion (from 12 to 4 minutes before secretin) was negligible, with a mean secretion rate of 0.3 ml=min (95% CI 0.1 0.8). Pancreatic ADC Changes Postsecretin Sequential DWI before and after secretin showed a significant increase in mean ADC value of the pancreatic head from presecretin images (1.29 10 3 mm 2 =s, 95% CI 1.17 1.41) to 1 min postsecretin images (1.48 10 3 mm 2 =s, 95% CI 1.33 1.63) (P ¼ 0.003) with a gradual decrease in ADC values on consecutive images back to baseline values (Fig. 3b). We observed the same trend in the pancreatic body with an increase in mean ADC value from presecretin images (1.23 10 3 mm 2 =s) to 1 minute postsecretin images (1.36 10 3 mm 2 =s; P ¼ 0.07). Correlations Between Pancreatic Imaging Characteristics and Clinical Findings Secreted pancreatic volumes at 5, 9, and 13 minutes postsecretin were positively correlated with pancreatic head ADC values at all timepoints, and secreted volumes also correlated with pancreatic body ADC values at some timepoints (Table 1). We identified no significant correlation between changes in ADC values after secretin and secreted volumes. Peak bicarbonate concentration was positively correlated with secreted pancreatic volumes at 1 minute postsecretin (r ¼ 0.64, P ¼ 0.04: Fig. 4). The same tendency was observed between peak bicarbonate concentration and secreted pancreatic volumes at 5 minutes (r ¼ 0.38, P ¼ 0.29), 9 minutes (r ¼ 0.52, P ¼ 0.12), and 13 minutes (r ¼ 0.30, P ¼ 0.39) postsecretin, although not statistically significant. There was no significant correlation between peak bicarbonate concentration and pancreatic volume or ADC values. DISCUSSION This study presents morphological and functional MRI characteristics of the pancreas in healthy individuals. Pancreatic MRI with secretin-stimulated MRCP

452 Wathle et al. Figure 3. Estimated bowel fluid volumes (a) and pancreatic head ADC values (b) at different timepoints before and after administration of secretin. Mean fluid volumes (ml; 95% CI displayed as error bars) in the duodenum and proximal jejunum increased after administration of secretin (a). Mean pancreatic ADC values (10 3 mm 2 =s; 95% CI displayed as error bars) increased initially after administration of secretin with a gradual decline to baseline values (b). and DWI is a feasible imaging method enabling characterization and quantification of exocrine pancreatic function in healthy humans. For patients with suspected exocrine pancreatic disease, this noninvasive imaging technique may prove especially relevant and could provide new insight into alterations in pancreatic structure and functional characteristics of exocrine pancreatic disease. Studies on patients with exocrine pancreatic disease, however, are needed to establish the potential of this imaging method as a diagnostic tool. In the present MRI study, we quantitatively assessed pancreatic exocrine secretion among healthy subjects. Mean pancreatic secretion rates after secretin were estimated at 10.7 ml=min, 8.1 ml=min, and 6.4 ml=min during the first 5, 9, and 13 minutes, respectively, which are in line with previous MRI studies reporting mean pancreatic secretion rates of 6.8 ml=min (0 14 minutes postsecretin) (7), 8.1 ml=min (0 7 minutes) (10), and 7.4 ml=min (0 7 minutes) (9) among healthy subjects using similar MRI protocols and postprocessing methods. Interestingly, Sanyal et al (15) has also reported similar pancreatic secretion rates of 5.7 ml=min (0 10 minutes) after having administered a negative oral contrast agent which ideally eliminates the contribution of fluid in the bowel before secretin on calculations of fluid content after secretin. As opposed to the observed nonlinear Table 1 Correlation Between Pancreatic ADC Values and Pancreatic Secretion Secreted volumes at different timepoints postsecretin 1 min 5 min 9 min 13 min Mean ADC-values r P a r P a r P a r P a Pancreatic head Before secretin 0.15 0.54 0.54 0.01 0.52 0.02 0.48 0.03 1 min after secretin 0.21 0.38 0.60 0.005 0.54 0.01 0.49 0.03 5 min after secretin 0.22 0.35 0.52 0.02 0.46 0.04 0.53 0.02 9 min after secretin 0.40 0.08 0.62 0.004 0.53 0.02 0.52 0.02 13 min after 0.24 0.32 0.61 0.004 0.47 0.04 0.56 0.001 secretin Pancreatic body Before secretin 0.28 0.23 0.42 0.07 0.33 0.16 0.31 0.19 1 min after secretin 0.49 0.03 0.47 0.04 0.39 0.09 0.44 0.05 5 min after secretin 0.48 0.03 0.60 0.005 0.49 0.03 0.49 0.03 9 min after secretin 0.46 0.04 0.46 0.04 0.36 0.12 0.37 0.11 13 min after secretin 0.55 0.01 0.43 0.06 0.38 0.10 0.40 0.08 Significant values (P > 0.05) are given in bold. *Pearson bivariate correlation test. Figure 4. Correlations between imaging findings and clinical characteristics. Scatterplot depicting the positive correlation between peak bicarbonate concentration and secreted pancreatic volume at 1 minute postsecretin (r ¼ 0.64, P ¼ 0.04; n ¼ 10).

Exocrine Pancreatic Function Assessed by MRI 453 secretion rate with a gradual decline after 5 minutes postsecretin in the present study, Bali et al (7) reported fairly linear secretion rates during 0 14 minutes postsecretin, although a plateau phase similar to our findings was observed at the end of the time period in 30% of their control group; this apparent discrepancy may be related to the slightly different methods applied for quantification of fluid-filled bowel content. DWI is a functional imaging technique in which diffusion of water molecules in the tissue is measured. Pancreatic exocrine secretion involves transport of sodium and bicarbonate ions from the blood into the ductal lumen creating an osmotic gradient that induces movement of water molecules into the pancreatic duct (11). This physiological process, which is enhanced by secretin stimulation, could theoretically lead to alterations in pancreatic diffusion properties that could be detected and quantified by DWI. In the present study, DWI showed an early increase in pancreatic ADC value postsecretin with a gradual decrease in ADC values on consecutive images back to baseline values (Fig. 3b). This finding is in accordance with a previous study reporting peak pancreatic ADC values at 1.5 4 minutes postsecretin in healthy controls and later peak ADC values (at 4 8 minutes) in chronic pancreatitis (11). Mean pancreatic ADC value of 1.29 10 3 mm 2 =s before secretin in our study is quite similar to the reported pancreatic head=body=tail ADC values of 1.13=1.05=0.94 10 3 mm 2 =s in healthy subjects (16), and to the reported ADC values of 1.2 10 3 mm 2 =s in both chronic pancreatitis and controls (11). However, mean relative increase in ADC value postsecretin in our study (using b-values of 50=800 sec=mm 2 ) was only 15% as opposed to 75% in a previous study (using b-values of 0=400 sec=mm 2 ) (11). Since applying lower b-values in principle increases perfusion and decreases the diffusion effects resulting in relatively higher ADC values (16), this observed difference may be due to the different b-values applied. The very pronounced positive correlation observed in this study between secreted pancreatic volumes and absolute ADC values before and after secretin stimulation is interesting, indicating that the diffusion properties of pancreatic tissue may be related to its secretory capacity. However, since ADC values do not have a direct physical or physiological correlate, interpreting the physiological implication of this finding is difficult. Future studies of secretin-stimulated DWI in patients with pancreatic exocrine dysfunction are warranted to explore whether alterations in pancreatic diffusion properties in this patient group may prove clinically relevant. Bicarbonate concentration in duodenal fluid is considered one of the most reliable biochemical markers of pancreatic exocrine function (13). We found that peak bicarbonate concentration was positively correlated with secreted pancreatic volumes at 1 minute postsecretin with the same tendency also at 5, 9, and 13 minutes postsecretin. This finding supports the potential usefulness of s-mrcp in assessing exocrine pancreatic function. Larger samples, however, are needed to confirm this finding. In this study, calculated pancreatic size is comparable to previous reports on pancreatic imaging in healthy individuals. Mean pancreatic size of 86 ml in adults based on MRI in this study is in line with reported pancreatic size of 101 ml based on MRI (17) and 84 ml (2) and 70 80 ml in adults (aged 20 70 years) (18) from computed tomography (CT) studies. Our finding of pancreas divisum in 5% (1=20) of the subjects is also in line with reported prevalence of 5% 10% of this anomaly in subjects with no former history of pancreatitis (19). Our study has some limitations. The study sample (n ¼ 20) is relatively small, in particular the subgroup that underwent gastroscopy (n ¼ 10). It is, however, difficult to recruit healthy controls to studies involving gastroscopy. Since our findings were statistically significant in this small sample, we believe our method is quite promising. Thus, we are currently extending our imaging studies to include patients with various subtypes of exocrine pancreatic disease in which gastroscopy is a part of the standard diagnostic work-up. In conclusion, secretin-stimulated MRCP and DWI can characterize and quantify exocrine pancreatic function in healthy controls. This may represent a relevant and promising noninvasive diagnostic method in patients with exocrine pancreatic dysfunction. REFERENCES 1. Vesterhus M, Haldorsen IS, Raeder H, Molven A, Njolstad PR. Reduced pancreatic volume in hepatocyte nuclear factor 1A-qmaturity-onset diabetes of the young. J Clin Endocrinol Metab 2008;93:3505 3509. 2. Ræder H, Johansson S, Holm PI, et al. Mutations in the CEL VNTR cause a syndrome of diabetes and pancreatic exocrine dysfunction. Nat Genet 2006;38:54 62. 3. Sunnapwar A, Prasad SR, Menias CO, Shanbhogue AK, Katre R, Raut A. Nonalcoholic, nonbiliary pancreatitis: cross-sectional imaging spectrum. AJR Am J Roentgenol 2010;195:67 75. 4. Siegmund E, Lohr JM, Schuff-Werner P. [The diagnostic validity of non-invasive pancreatic function tests a meta-analysis.] Z Gastroenterol 2004;42:1117 1128. 5. Stevens T, Parsi MA. Update on endoscopic pancreatic function testing. World J Gastroenterol 2011;17:3957 3961. 6. Wallace MB. Imaging the pancreas: into the deep. Gastroenterology 2007;132:484 487. 7. Bali MA, Sztantics A, Metens T, et al. Quantification of pancreatic exocrine function with secretin-enhanced magnetic resonance cholangiopancreatography: normal values and short-term effects of pancreatic duct drainage procedures in chronic pancreatitis. Initial results. Eur Radiol 2005;15:2110 2121. 8. Schneider AR, Hammerstingl R, Heller M, et al. Does secretinstimulated MRCP predict exocrine pancreatic insufficiency?: A comparison with noninvasive exocrine pancreatic function tests. J Clin Gastroenterol 2006;40:851 855. 9. Gillams AR, Lees WR. Quantitative secretin MRCP (MRCPQ): results in 215 patients with known or suspected pancreatic pathology. Eur Radiol 2007;17:2984 2990. 10. Punwani S, Gillams AR, Lees WR. Non-invasive quantification of pancreatic exocrine function using secretin-stimulated MRCP. Eur Radiol 2003;13:273 276. 11. Erturk SM, Ichikawa T, Motosugi U, Sou H, Araki T. Diffusionweighted MR imaging in the evaluation of pancreatic exocrine function before and after secretin stimulation. Am J Gastroenterol 2006;101:133 136. 12. Haldorsen IS, Raeder H, Vesterhus M, Molven A, Njolstad PR. The role of pancreatic imaging in monogenic diabetes mellitus. Nat Rev Endocrinol 2012;8:148 159.

454 Wathle et al. 13. Jensen NM, Larsen S. A rapid, endoscopic exocrine pancreatic function test and the Lundh test: a comparative study. Pancreatology 2008;8:617 624. 14. Manfredi R, Perandini S, Mantovani W, Frulloni L, Faccioli N, Pozzi MR. Quantitative MRCP assessment of pancreatic exocrine reserve and its correlation with faecal elastase-1 in patients with chronic pancreatitis. Radiol Med 2012;117:282 292. 15. Sanyal R, Stevens T, Novak E, Veniero JC. Secretin-enhanced MRCP: review of technique and application with proposal for quantification of exocrine function. AJR Am J Roentgenol 2012;198:124 132. 16. Schoennagel BP, Habermann CR, Roesch M, et al. Diffusionweighted imaging of the healthy pancreas: apparent diffusion coefficient values of the normal head, body, and tail calculated from different sets of b-values. J Magn Reson Imaging 2011;34:861 865. 17. Williams AJ, Chau W, Callaway MP, Dayan CM. Magnetic resonance imaging: a reliable method for measuring pancreatic volume in type 1 diabetes. Diabet Med 2007;24:35 40. 18. Saisho Y, Butler AE, Meier JJ, et al. Pancreas volumes in humans from birth to age one hundred taking into account sex, obesity, and presence of type-2 diabetes. Clin Anat 2007;20:933 942. 19. DiMagno MJ, Wamsteker EJ. Pancreas divisum. Curr Gastroenterol Rep 2011;13:150 156.