RETURN TO ABSTRACT LISTING


Objective:
Radiographs are needed to verify temporary feeding tube position in critically ill children, but repeated exams may raise concern about radiation. Therefore, we developed a reduced-radiation radiographic examination for targeted assessment of temporary feeding tubes. This examination decreases radiation but also decreases overall image quality. Clinical utility of this exam depends on how often the tube is visible, whether malpositioned tubes are identified, and how often other clinically relevant findings are interpreted. It is also important to understand how radiologists and clinicians interact with this reduced image quality examination.

Materials and Methods:
This quality improvement initiative was IRB-exempt. Reviewing feeding tube radiograph reports from April 2018 through March 2019, we assessed tube visibility, location, and presence of new or critical findings using descriptive statistics. On a random subset of 100 exams, two radiologists and two critical care clinicians assessed whether findings in the reports were clinically relevant, using kappa values to assess agreement. To assess clinical experience, we sent Likert surveys to radiologists and to clinicians in surgery and intensive care, then characterized group responses with descriptive statistics.

Results:
Of 3286 reports, tube was not visible on 31 examinations (0.9%). Tubes were malpositioned in the airway on 8 examinations (0.2%) or esophagus/gastroesophageal junction on 74 exams (2.3%). Concern for pneumothorax or pneumoperitoneum was mentioned in 5 examinations (0.2%). Other clinical findings warranting follow-up were described in 19 examinations (0.6%). Intensivists substantially agreed on clinically relevant findings (k = 0.64), while radiologists moderately agreed (k = 0.49). However, radiologists and clinicians together had only fair agreement (k = 0.37). Radiologists (n = 30) strongly agreed that reduced-radiation feeding tube examinations were easy to evaluate (80%) and dictate (77%). Clinicians (n = 115) strongly agreed that reports for these examinations were clear and easily used for managing patients (74%).

Conclusion:
Reducing radiation and image quality for a targeted assessment of feeding tube location alleviates patients from unnecessary ionizing radiation but does not significantly limit assessment of tube position. Other concerning imaging findings may still be identified, but education may be needed to ensure common definition of clinically relevant findings. Reduced image quality did not lead to difficulty in evaluating, dictating, or using report information in the context of this targeted clinical question.