Short pancreas: evaluation with multi-detector row CT.

PURPOSE

The purpose of this study was to analyze the length variations of the pancreas using computed tomography (CT) and establish a database for short pancreas.

METHODS

We retrospectively reviewed CT examinations of 228 adults and rated pancreatic lengths qualitatively on a scale of 1-3 using transverse images. 1, normal pancreas length; 2, mildly short pancreas; and 3, markedly short pancreas. The length of the pancreas from head to tail was also measured using the "curved line tool" through the midline of the organ on curved planar reconstructed (CPR) images. The pancreatic neck-tail length and the abdominal radius were measured on transverse images, and the ratio of pancreatic neck-tail length to abdominal radius was calculated to avoid the effect of body mass differences. All data were analyzed statistically.

RESULTS

The pancreas length was normal (group 1) in 180 (78.9%) patients, mildly short (group 2) in 38 (16.7%), and markedly short (group 3) in 10 (4.4%). The average pancreatic length on CPR evaluation was 207.5 ± 19.1 mm in group 1, 168.9 ± 8.5 mm in group 2, and 135.1 ± 10.7 mm in group 3. There were statistically significant differences between three groups. Although it was not statistically significant, percentage of diabetes mellitus was higher in group 3 (20%) than other two groups (13.2% in group 2, and 8.9% in group 1).

CONCLUSIONS

CT examination of the pancreas is an effective imaging method to classify the pancreatic length and to detect short pancreas. We suggest that pancreatic length variations should be reported on routine abdominal CT examinations.