Author of 1 Presentation
SS 12.8 - Inter-observer variability of radiologists for Cambridge Classification for Chronic Pancreatitis using CT and MRCP: results from a large multi-center study
Abstract
Purpose
To determine inter-observer variability among radiologists in assigning Cambridge Classification (CC) for chronic pancreatitis (CP) based on secretin-enhanced MRCP and contrast-enhanced CT.
Material and methods
This multi-institutional study included 39 subjects who were selected randomly without reference to their clinical data, from 422 participants enrolled in the PROCEED study (ClinicalTrials.gov ID NCT03099850) between June 2017 and August 2018. These subjects were categorized into three groups based on clinical data and review of imaging by local radiologist (LR) at the enrolling sites: controls (n = 8; CC of 0), suspected CP (n = 22; CC of 1 or 2) and definite CP (n = 9; CC = 3 or 4). Each study was scored by 3 radiologists, who were randomly selected from 5 blinded subspecialized radiologists. We compared the CC score by LRs and a consensus CC score by central radiologists (CRs). The weighted kappa statistic (wKappa) was used to estimate the inter-observer agreement.
Results
There was moderate agreement (63% and 67%, respectively) between CRs and LRs in both CT score (wKappa [95% CI] = 0.56 [0.34, 0.78]; p-value = 0.57) and MR score (wKappa [95% CI] = 0.68 [0.49, 0.86]; p-value = 0.72). The composite CT/MRCP score showed moderate agreement (wKappa [95% CI] = 0.62 [0.43, 0.81]; p-value = 0.80). For majority of subjects, the actual cohort assignment by the LR agreed with consensus composite CT/MRCP score by the CRs, with concordance ranging from 75% to 89%.
Conclusion
There is a high degree of concordance by radiologists for the assignment of CC using MRCP and CT.
Video-on-demand
Author of 1 Presentation
SS 12.8 - Inter-observer variability of radiologists for Cambridge Classification for Chronic Pancreatitis using CT and MRCP: results from a large multi-center study (ID 882)
Abstract
Purpose
To determine inter-observer variability among radiologists in assigning Cambridge Classification (CC) for chronic pancreatitis (CP) based on secretin-enhanced MRCP and contrast-enhanced CT.
Material and methods
This multi-institutional study included 39 subjects who were selected randomly without reference to their clinical data, from 422 participants enrolled in the PROCEED study (ClinicalTrials.gov ID NCT03099850) between June 2017 and August 2018. These subjects were categorized into three groups based on clinical data and review of imaging by local radiologist (LR) at the enrolling sites: controls (n = 8; CC of 0), suspected CP (n = 22; CC of 1 or 2) and definite CP (n = 9; CC = 3 or 4). Each study was scored by 3 radiologists, who were randomly selected from 5 blinded subspecialized radiologists. We compared the CC score by LRs and a consensus CC score by central radiologists (CRs). The weighted kappa statistic (wKappa) was used to estimate the inter-observer agreement.
Results
There was moderate agreement (63% and 67%, respectively) between CRs and LRs in both CT score (wKappa [95% CI] = 0.56 [0.34, 0.78]; p-value = 0.57) and MR score (wKappa [95% CI] = 0.68 [0.49, 0.86]; p-value = 0.72). The composite CT/MRCP score showed moderate agreement (wKappa [95% CI] = 0.62 [0.43, 0.81]; p-value = 0.80). For majority of subjects, the actual cohort assignment by the LR agreed with consensus composite CT/MRCP score by the CRs, with concordance ranging from 75% to 89%.
Conclusion
There is a high degree of concordance by radiologists for the assignment of CC using MRCP and CT.