Discrepancies in Radiograph Interpretation Between Pediatric Radiologists and Pediatric Intensivists in the Pediatric or Neonatal Intensive Care Unit

Adam Z Fink; Terry L Levin; Einat Blumfield; Sheri L Nemerofsky; Mark C Liszewski; Kandi George; Suhas Nafday; Mark Shlomovich; K Allen Eddington; Eliza H Myers; H Michael Ushay

doi:10.1016/j.jacr.2017.12.007

Discrepancies in Radiograph Interpretation Between Pediatric Radiologists and Pediatric Intensivists in the Pediatric or Neonatal Intensive Care Unit

J Am Coll Radiol. 2018 Nov;15(11):1580-1586. doi: 10.1016/j.jacr.2017.12.007. Epub 2018 Mar 2.

Authors

Affiliations

¹ Department of Radiology, Division of Pediatric Radiology, Children's Hospital of Montefiore Medical Center, Bronx, New York.
² Department of Radiology, Division of Pediatric Radiology, Children's Hospital of Montefiore Medical Center, Bronx, New York. Electronic address: tlevin@montefiore.org.
³ Department of Pediatrics, Division of Neonatology Children's Hospital of Montefiore Medical Center, Bronx, New York.
⁴ Department of Pediatrics, Children's Hospital of Montefiore Medical Center, Bronx, New York.
⁵ Department of Radiology, Division of Neonatology, Jacobi Medical Center, Bronx, New York.

PMID: 29501502
DOI: 10.1016/j.jacr.2017.12.007

Abstract

Background: In pediatric intensive care units (PICUs) and neonatal intensive care units (NICUs), patient management decisions are sometimes based on preliminary interpretations of radiographs by pediatric intensivists (PIs) before a formal interpretation by a pediatric radiologist (PR).

Objective: To quantify and classify discrepancies in radiographic interpretation between PRs and PIs in the PICU and NICU.

Materials and methods: This institutional review board-approved multi-institutional prospective study included three PRs and PIs at two PICUs and three NICUs. Interpretations of chest and abdominal radiographs by PIs and PRs were recorded on online forms and compared. Discrepancies in interpretations were classified as "miss," "misinterpretation," or "overcall." The discrepancies were also categorized as "actionable" or "nonactionable" based on extrapolation of the ACR actionable reporting work group's list of actionable findings.

Results: In 960 radiographic interpretations, the total, nonactionable, and actionable discrepancy rates between PRs and PIs were 34.7%, 26.8%, and 7.9%, respectively. The most common actionable discrepancies were line or tube positions and identification and interpretation of parenchymal opacities in the lungs. Identification of air leaks in the PICU and differentiation of normal from abnormal bowel gas patterns in the NICU followed in frequency. Air leaks accounted for 1% of total discrepancies and 11% of actionable discrepancies. Most discrepancies were nonactionable and included retrocardiac atelectasis and mischaracterization of neonatal lung disease in the PICU and NICU, respectively.

Conclusion: Although the total discrepancy rate was high, most discrepancies were nonactionable. Actionable discrepancies were predominantly due to line and tube position, which should be an area of focused education.

Keywords: NICU; PICU; Radiograph interpretation; discrepancy; pediatric intensivist; pediatric radiologist.

Publication types

Multicenter Study

MeSH terms

Child
Child, Preschool
Clinical Competence*
Diagnostic Errors / statistics & numerical data*
Female
Hospitals, Pediatric
Humans
Infant
Infant, Newborn
Intensive Care Units, Neonatal
Intensive Care Units, Pediatric*
Male
Observer Variation
Pediatrics / standards*
Prospective Studies
Radiology / standards*