E1591. Blind Spots in CT Diagnosis of Acute Pathology in the Abdomen and Pelvis: Lessons Glimpsed From Peer Learning
Authors
Kheng Lim;
Penn Medicine
Objective:
We recently contracted with an external teleradiology service to provide preliminary interpretation for emergent CT scans of the abdomen and pelvis performed in two hospitals as a buffer to increasing volume, and to lessen the burden on our radiology residents during their overnight call. The purpose of this retrospective study is to understand the blind spots in interpreting abdominal and pelvic CTs with the hope that the radiology community can collectively learn to look for these blind spots to help optimize patient care.
Materials and Methods:
Using our health system electronic medical records, we performed a retrospective review for a period of six months in the first half of 2022. The teleradiologists provided preliminary reports for CTs of the abdomen and pelvis performed emergently in two of our hospitals. The final reports were generated the following morning by our abdominal radiologists. Actionable discordance between preliminary and final reports was flagged by our abdominal radiologists in the final reports after telephone notification to the ordering clinicians. These cases were reviewed for subsequent impact on the patient care and clinical course.
Results:
There were 30 CTs of the abdomen and pelvis where the final reports were flagged as containing significant changes to the preliminary reports in these organ systems: gastrointestinal system (11 cases); cardiovascular system (6 cases); peritoneum (3 cases); kidneys and urinary tract (2 cases); osseous fractures (2 cases); gallbladder (1 case); spleen (1 case), and miscellaneous (4). Two studies resulted in emergent surgeries (SMA thrombectomy and right hemicolectomy) supporting final radiology interpretation. A case of sepsis from perforated sigmoid colon was managed conservatively. Three potential unnecessary appendectomies were avoided by final radiologist interpretation. Further intervention for misdiagnosis of active bleeding (3 cases) was avoided by the final radiology interpretation.
Conclusion:
Understanding radiologist blind spots in image interpretation is paramount in directing optimal downstream management of diseases. We highlight the areas of anatomy that fully qualified, board-certified radiologists may overlook in their daily routine of providing imaging diagnosis. Learning the blind spots from these cases can sharpen radiologist diagnostic accuracy, further solidifying the central role of radiologists in optimal patient management and outcome.
