Skip to main content
Fig. 4 | Insights into Imaging

Fig. 4

From: Errors, discrepancies and underlying bias in radiology with case examples: a pictorial review

Fig. 4

Faulty reasoning (type 2) error with anchoring and confirmation bias. A 10-year-old female patient was admitted to the ER after MVA. She had signs of severe head trauma and was intubated in the ER due to altered mental status. Following intubation, the absence of left lung ventilation was noticed, and a chest tube was placed in the left lung. Left-sided pneumothorax and misplaced intubation cannula extending to the bronchus intermedius were seen on the control chest radiograph (a). After stabilizing the vital signs, thorax and abdomen CTs were performed in the second hour of admission to assess the severity of the traumatic injury. Thorax CT showed “collapsed left lung with pneumothorax” (b, c) and “discontinuity of left main bronchus” (red arrow, c). The on-call radiologist evaluated those findings in favor of “bronchial rupture” and “fallen lung sign.” After making a preliminary report, the radiologist was called by a clinician, who informed him that patient’s intubation was traumatic and difficult, and asked him whether these findings could be secondary to “acute bronchial obstruction and subsequent collapse.” The radiologist rejected this possibility without thinking and stated that the findings he saw were compatible with bronchial rupture. Following the telephone call, the radiologist scanned the literature and read that bronchial rupture is usually seen with severe accompanying thoracic injuries and is associated with pneumomediastinum and pneumothorax. Although the radiologist knew that the patient did not have any other thoracic injury or pneumomediastinum, he sought exceptional cases that supported his pre-diagnosis, and he stuck with his first decision. The patient’s poor general condition did not allow surgical intervention or bronchoscopy. She was followed up with chest tube drainage and bronchial aspiration. On the second day after admission, portable chest radiography showed that the left lung ventilation was markedly normalized (d). When the findings were evaluated retrospectively, it was understood that bronchial rupture was unlikely, and the diagnosis was compatible with “acute lung collapse and pneumothorax ex vacuo secondary to the traumatic intubation.” Thus, the radiologist, under the influence of anchoring confirmation bias, had falsely attributed the findings he detected in favor of a false pathology

Back to article page