Published in:
01-03-2016
Measurement of pancreatic cystic lesions on magnetic resonance imaging: efficacy of standards in reducing inter-observer variability
Published in: Abdominal Radiology | Issue 3/2016
Login to get accessAbstract
Purpose
The purpose of this study is to assess inter-observer variability in the measurement of pancreatic cystic lesions with MRI and to determine the impact of measurement standards.
Materials and methods
In this IRB-approved, HIPAA-compliant study with waiver of informed consent, 144 MRI examinations, containing pancreatic cystic lesions measuring between 5 and 35 mm, were reviewed independently by two radiology attendings and two abdominal imaging fellows. Measurements were repeated by the same reviewers 12 weeks later after the introduction of measurement standards. Results were analyzed using within-subject standard deviation, intraclass correlation coefficient, and kappa.
Results
Prior to standardization, the within-subject standard deviation, showing measurement variability in each cyst, was 4.0 mm, which was reduced to 3.3 mm after introduction of measurement standards (p < 0.01). Overall inter-observer agreement, kappa, improved from 0.59 to 0.65 (p = 0.04). The frequency of all four reviewers agreeing on size category increased from 51% to 60%. The intraclass correlation coefficient increased from 0.81 to 0.86.
Conclusions
There is significant and frequent inter-observer variability in the measurement of pancreatic cystic lesions with MRI which could affect clinical management. Implementation of measurement standards reduces measurement variability and aids in preventing erroneous reporting of growth and potentially unwarranted changes in management.