Pancreatic cyst characterization: maximum axial diameter does not measure up

Background: Unidimensional size is commonly used to risk stratify pancreatic cysts (PCs) despite inconsistent performance. The current study aimed to determine if unidimensional size, demonstrated by maximum axial diameter (MAD), is an appropriate surrogate measurement for volume and surface area. Methods: Patients with cross-sectional imaging of PCs from 2012 to 2013 were identified. Cyst MAD, volume, and surface area were measured using quantitative imaging software. Non-pseudocystic PCs >1 cm were selected for inclusion to assess MAD correlation with volume and surface area. Cysts imaged twice >1 year apart were selected to evaluate volumetric growth rate. Results: In total, 195 cysts were included. Overall, MAD was strongly correlated with volume (r = 0.83) and surface area (r = 0.93). However, cysts 1-2 cm and 2-3 cm were weakly correlated with volume and surface area: r = 0.78, 0.57 and 0.82, 0.61, respectively. Cyst volumes and surface areas varied widely within unidimensional size groups with 51% and 40% of volumes and surface areas overlapping uni-dimensional size groups, respectively. Estimated changes in volume poorly predicted measured changes in volume with 42% of cysts having >100% absolute percent difference. Conclusions: Pancreatic cyst volume and surface area may be useful adjunct measurements to risk stratify patients and surveil cyst changes and deserves further study.

Automated summary

The study showed a strong correlation between unidimensional size and volume/surface area, but weaker correlations for cysts in the 1-2cm and 2-3cm range. Volume and surface areas varied widely within unidimensional size groups, and estimated volume changes poorly predicted actual changes in volume.