With many radiology cases, it is only in hindsight — with the knowledge that there is, in fact, an abnormality — that the abnormality can then be identified by comparison.1 Consider how the outcome of this case might have been different if the second radiologist had more thoughtfully reported the discrepancies in imaging interpretation.
Allegation
The radiologist failed to recognize a recurrent tumor on CT studies for a year, which led to delayed diagnosis, progression of the disease, more invasive surgery and additional medical treatment.
Case File
A 70-year-old male patient was diagnosed with hepatocellular carcinoma (HCC) in 2013 and underwent primary resection of the tumor. His treatment thereafter included monitoring every three months via CT imaging. The CT images were reviewed first by Radiologist 1, then by an oncologist who specialized in the treatment of HCC. CT images through October 2015 were reported by Radiologist 1 as stable with no evidence of residual, recurrent or metastatic HCC. This was confirmed by the oncologist. In January 2016, Radiologist 2 discovered a lesion adjacent to the post-surgical site. Realizing this lesion represented a recurrence of the patient’s cancer, and with the knowledge of its location, she traced the lesion back to January 2015, which was the first time she could see it. Radiologist 2 created an addendum to the patient’s record noting that the lesion was visible in January 2015 and had been increasing in size. Radiologist 2 then confronted Radiologist 1. Radiologist 1 felt terrible. He called the patient, admitted he had missed the recurrence starting in January and said he was very sorry. The patient then met with Radiologist 2, who told him the tumor was visible in the January 2015 images and that it should not have been missed by Radiologist 1. The patient had several surgeries and chemotherapy, from which he suffered substantial side effects. He filed a lawsuit against Radiologist 1 and the radiology practice.
Discussion
Defense experts who reviewed the 2015 images without knowledge of the recurrence either did not identify the abnormality or identified it starting with the October 2015 images. When allowed to view the images after being shown the lesion on the January 2016 images, they were all able to see a hint of something on the January 2015 images. However, these experts did not believe the standard of care required identification of a recurrence in January 2015 because they were only able to identify it with prior knowledge of its existence, which Radiologist 1 did not have. Some experts believed Radiologist 1 should have recognized a recurrence in October 2015, because even though the nodule was less than 1 cm at that point, it was in a location where a recurrence would be expected. Therefore, this was really a case about a three-month diagnosis delay, not a 12-month delay as alleged. Causation experts believed the extent of surgery and chemotherapy recommendations would have been the same if the recurrence had been diagnosed in October 2015. Therefore, the defense could argue that the three-month delay in diagnosis did not cause the patient any injury.
However, in considering whether this case should be taken to trial, the defense team had to consider the fact that both defendant radiologists had admitted liability to the patient and that Radiologist 2 had essentially provided an expert opinion supporting the plaintiff’s claims in the addendum she wrote. Despite the fact that there was a strong causation defense, these other factors suggested that reasonable settlement was the way to resolve the matter.
Medical Liability Risk Management Recommendations
Consider the following recommendations when apparent discrepancies in imaging interpretation are discovered:1,2,3,4
- Standardize the method of identifying discrepancies and adopt an action plan for responding to them.
- Do not judge a prior radiology impression based on a retrospective review.
- When a previously missed lesion is identified, note it, but avoid accusatory or aggressive language (e.g., “missed,” “error,” “mistake,” “overlooked,” “not appreciated,” “obviously present” or “should have been identified”).
- Choose succinct, neutral language that is non-judgmental (e.g., “on further review, the recurrence was present on the radiograph of October 15, 2011”).
References
1. Yousem DM. Malpractice: What the Thinking Radiologist Should Know. American College of Radiology. (accessed 10/1/2019)
2. American College of Radiology. "ACR Practice Parameter for Communication of Diagnostic Imaging Findings." 2014. (accessed 10/1/2019)
3. Berlin L. Radiologic Errors and Malpractice: A Blurry Distinction. AJR. 2007; 189:517–522.
4. Pennsylvania Patient Safety Advisory. Communication of Radiographic Discrepancies Between Radiology and Emergency Departments. Pa Patient Saf Advis. 2010 Mar;7(1):18-22.
