Quality assurance (QA) case databases serve a critical role in training residents and improving practice, enabling radiologists and researchers to identify common diagnostic errors in cases such as acute cholecystitis using root cause analysis, according to a study published in the March issue of the American Journal of Roentgenology.
“It is imperative to use root cause analysis of radiologic errors for improvement of work practice and identifying teaching opportunities for residents, fellows, and faculty,” explained Olga R. Brook, MD, and co-authors from the department of radiology, Beth Israel Deaconess Medical Center and Harvard Medical School in Boston.
“The purpose of this study was to identify contributing factors in the misdiagnosis of acute cholecystitis that can be used for continuing education,” Brook and colleagues continued. With the radiology department submitting cases to the hospital’s quality assurance database since August 2004, the system had collected 1,678 abdominal imaging cases as of April 2010, a period over which the department had performed 93,663 abdominal CT and 34,996 abdominal ultrasound examinations of emergency and admitted patients.
The authors searched the database for acute cholecystitis misdiagnoses, classifying errors as overcalls, which overestimated severity, and undercalls, which underestimated severity. Cases were reviewed by a senior attending abdominal imaging radiologist with more than 30 years of experience and an abdominal imaging fellow.
The authors identified 14 misdiagnoses involving a question of acute cholecystitis, 11 of which were undercalled and three of which were overcalled. All overcalled cases were ultrasound exams and six undercalls were interpreted with CT, while five were viewed on ultrasound. All identified cases involving CT were undercalls.
The final diagnoses for overcalled patients included hepatitis, sepsis and chronic cholecystitis. According to the authors, gallbladder wall edema was present but none of these cases portrayed distension of the gallbladder.
For undercalled cases, CT was misleading because cholecystitis was not considered clinically. The authors identified the main contributing factors to misdiagnoses as lack of recognition of wall edema (six patients), lack of recognition of gallbladder distention (four patients), absence of gallbladder wall edema (one patient), lack of conclusion in the report (two patients) and hospitalization in the ICU (two patients).
Overcalled cases resulted in minor complications, while complications in undercalled patients included progression to gangrenous cholecystitis in four patients (one of whom died), gallbladder perforation in two patients, postoperative complications of bile leak in one patient and one case of acute renal failure.
The authors also observed clustering of cases each July, which coincided with incoming classes of residents and fellows, indicating the role of inexperience in misdiagnosis. Still, the authors said that the “finding was unexpected because all studies are read by an attending radiologist. Nevertheless, even the best radiologist will miss in 10 percent of cases.”
Brook and colleagues cited the small sample size and submission bias to the database as limitations to their analysis.
“In conclusion, acute cholecystitis can be a difficult imaging diagnosis,” Brook and co-authors said. “Understanding the diagnostic findings and common pitfalls, such as the importance of gallbladder distention, and knowledge of differential diagnoses of gallbladder wall edema and variability of presentation of acute cholecystitis in intensive care patients may improve diagnostic accuracy.”