Purpose or Case Report: In the PICU and NICU, radiographs are frequently obtained and preliminarily interpreted by intensivists several hours before a board certified pediatric radiologist provides an official interpretation. These preliminary interpretations often result in a change in clinical management. We quantified the discrepancy rates between intensivists at different levels of training and pediatric radiologists to determine areas for intensivist improvement via an educational initiative.
Methods & Materials: This IRB approved prospective study was conducted at 2 institutions and included 3 NICUs and 2 PICUs. Intensivists (attendings, fellows, nurse practitioners, and physician assistants) and 3 CAQ pediatric radiologists interpreted chest and abdominal films from August–December 2016 using online REDCap (Research Electronic Data Capture) forms. Each intensivist’s report was compared to the corresponding radiologist’s report. Discrepancies (including missed, misinterpreted, and overcalled findings) were placed into 1 of 6 groups and deemed clinically significant (requiring a management change) or insignificant (requiring no management change). Significant discrepancies were followed up to determine if management changes occurred.
Results: There were 506 reports that were reviewed (339 PICU and 167 NICU). While NICU and PICU discrepancy rates were 43% and 44% respectively (figure 1), clinically significant discrepancy rates in the PICU were lower than those of the NICU (5.9% and 19.8% respectively) and rates of discrepancies leading to management changes in the PICU were also lower than those in the NICU (3.2% and 12% respectively). PICU attendings had less total discrepancies, clinically significant discrepancies, and discrepancies leading to management changes than other PICU intensivists (figure 2). Discrepancy rates by NICU attendings were similar to those of other NICU intensivists. When stratified by patient age, discrepancy rates did not differ. The most common discrepancy resulting in management change was line/tube position followed by lung findings and air leaks (figure 3).
Conclusions: While significant discrepancies in interpretation exist between PICU/NICU intensivists and pediatric radiologists, many of these are clinically insignificant and do not lead to management changes. Discrepancies are more frequent in the NICU than the PICU. The most common discrepancy leading to management change was line/tube position followed by lung findings and air leaks.