One of the two significant misses in this week's study was the left ventricular thrombus. There are actually two separate hypodense filling defects in the left ventricle (red arrows).
Cardiac thrombi are occasionally encountered incidentally when reviewing enhanced CT studies of the chest and abdomen (similar to pulmonary emboli). It's important to identify cardiac thrombus because of the risk of distal embolization, which can result in stroke, visceral infarction, or distal limb ischemia. Another location to evaluate for incidental thrombus if the left atrial appendage, indicated below (red arrow). Your search pattern should include these two locations.
The second significant miss in this case was left axillary adenopathy (yellow arrow). This may be important for staging and treatment depending on the type of tumor and other sites of disease. Adenopathy was also seen in the upper abdomen, which was described in the report for that study.
Discussion
The contributing factors for these particular misses are unknown, but I would assume that workload/volume, interruptions, and inadequate search pattern may have played a role. Radiology practices with high-volume expectation are a set-up for these types of misses. The findings are clear on the static images (and in a courtroom) but can be easily missed when scrolling quickly through an exam.
It's important to learn the speed at which you can safely review imaging studies without missing subtle findings. One of the strategies that can help you identify subtle findings more quickly and accurately is to see several examples. Another strategy is to develop a comprehensive search pattern and stick to it. When these strategies are coupled with timely feedback about misses, you can get to a point where you rarely experience perceptual misses. Many radiologists incorrectly assume that the 2-3% miss rate seen in high quality retrospective studies is an unavoidable part of the way that humans process information and make decisions. I do not believe that this is true and have seen many examples where radiologists interpret thousands of cases over a period of weeks to months without having any significant perceptual errors.
