Francis Flaherty, MD

Here's a little game some radiologists like to play (surely no one you know). It's called "watching the cancer grow." Here's how it's played: Take a patient with lung nodules that are being followed. Each time you read out their chest CT exam, you compare it only to the most immediate prior study, not the oldest exam.

You report, accurately enough, that the nodules are stable from the prior study and you recommend continued surveillance. At no point do you bother going back to the original study and observing that, when compared back to the original exam, one or more of the nodules are slowly growing over time and therefore extremely suspicious for malignancy (ie, cancer until proven otherwise).[[{"type":"media","view_mode":"media_crop","fid":"44375","attributes":{"alt":"Francis Flaherty, MD","class":"media-image media-image-right","id":"media_crop_1286219409600","media_crop_h":"0","media_crop_image_style":"-1","media_crop_instance":"4953","media_crop_rotate":"0","media_crop_scale_h":"0","media_crop_scale_w":"0","media_crop_w":"0","media_crop_x":"0","media_crop_y":"0","style":"height: 237px; width: 170px; border-width: 0px; border-style: solid; margin: 1px; float: right;","title":"Francis Flaherty, MD","typeof":"foaf:Image"}}]]

Here are some variations to the game: Maybe you note that a lesion may have "slightly enlarged" but that this may be due to technical factors such as "slice selection." You recommend continued surveillance. Maybe you actually acknowledge that the lesion is enlarging, but you don't dare use the C word or the M word (cancer and malignancy, in case you were wondering).

Maybe you recommend PET scan for further assessment of your enlarging nodule (logically inconsistent, as an enlarging nodule would remain highly suspicious for malignancy even if it is PET negative).

Why do you do these things? Maybe you are too busy to compare the study with all the old priors. It is tedious. As long as you keep recommending follow-up, you're covered, right? Eventually some other radiologist will actually compare to all the old exams and find the cancer.

Well, what happens if you are forced to acknowledge that a nodule has enlarged? Why don’t or can't you call your enlarging nodule cancer until proven otherwise? What’s with the obfuscation and hemming and hawing? Perhaps you don't even know. Perhaps you don't like making the call, preferring instead to kick the can down the road to the next rad.

Perhaps you've internalized the caricature of the radiologist sitting on a fence eating a waffle. Whatever the case, you’re not doing anyone any good, not the patient, not yourself. Get down off the fence. Step away from the waffle. Make the call.

Articles

Watching the Cancer Grow

You know what I'm talking about. Don't deny it. A doc comes by to review a CT. The scan shows a large nasty looking pancreatic mass with vascular encasement and liver mets. The doc says, "Hey, didn't this guy have an abdominal CT eight months ago? Was this thing there?" Warily, you open the prior exam. You scroll to the pancreas. There, practically jumping off the screen, is an obvious pancreatic mass, though smaller and without liver mets or vascular encasement. It looks as big as a beach ball to your eye. Someone couldn’t miss this, could they? You glance at the report. No mention of the mass. You glance at the doc. You think to yourself, "Doesn’t he see how obvious this thing is?" He doesn't appear to see it.

You scroll off the pancreas. You start to hem and haw, "Well, there was motion artifact, the study wasn’t a dedicated pancreas protocol, the bolus timing was suboptimal, it’s hard to see if there’s anything there, blah, blah, blah." The doc leaves. You feel dirty. You tell your partner about their miss. They feel like crap. They notify their carrier. Maybe they notify hospital risk management. They wait for the lawsuit that may or may not come.

What should you have done? Should you have told the referring doc that this thing should have been seen? What good would that do for anyone? Does your hospital have a policy on disclosure of medical errors? If it does, do most docs actually follow it? Could you have missed something this obvious yourself? If you’re being completely honest, the answer to at least the last question should be a yes.

Just think about the numbers. The average radiologist reads approximately 15,000 cases per year.

 According to multiple studies, radiologists will misinterpret approximately 3%–5% of unselected cases and 30% of cases with positive findings.

Doing the math (assuming the approximately 4% error rate in unselected cases), an average radiologist (but, like the children of Lake Wobegon, aren't we radiologists all above average?) will misinterpret about 600 cases per year, each one potentially representing a successful malpractice claim.[[{"type":"media","view_mode":"media_crop","fid":"40167","attributes":{"alt":"","class":"media-image media-image-right","id":"media_crop_2271272627676","media_crop_h":"0","media_crop_image_style":"-1","media_crop_instance":"4085","media_crop_rotate":"0","media_crop_scale_h":"0","media_crop_scale_w":"0","media_crop_w":"0","media_crop_x":"0","media_crop_y":"0","style":"height: 150px; width: 200px; float: right;","title":"Â©Dooder/Shutterstock.com","typeof":"foaf:Image"}}]]

So why aren't we being sued all the time? This is where luck enters the equation. Whether we are sued depends on numerous factors: Is the patient aware of the miss? Was the patient harmed by the miss? If it was a cancer miss, was the finding picked up on a follow up study within a short enough time interval so as to not have made a clinical difference? Is the patient (or the patient's family) litigious?

Does the mere misinterpretation of a case constitute malpractice? Well, no. Malpractice requires both harm to the patient and a breach of the standard of care by the radiologist. If there is no harm, then (as they say in basketball), there is no foul. What about the ‘bad miss’? Does missing an obvious (in retrospect) finding constitute a breach of the standard of care? This is less clear.

The courts have long acknowledged that radiologists cannot be held to a standard of perfection, and that a 'missed' or misinterpreted finding does not in and of itself constitute malpractice. As long as the radiologist is appropriately trained and exercises due diligence in the interpretation of the exam (not reading the study from 30 feet away after reading 500 cases and being up for 72 hours straight), a missed significant finding should be defensible in court according to this standard. But it’s not.

In reality, if it’s an 'obvious' or 'bad' miss, it is not defensible, and the plaintiff’s expert witnesses will testify that the miss was sufficiently egregious as to constitute malpractice. For this reason, cases involving an obvious miss of a significant finding will be settled long before they see the inside of a courtroom, a de facto admission of wrongdoing by the radiologist. But is committing an obvious miss truly malpractice? It shouldn’t be, not according to the law and not according to the science.

If you go by the research on radiology errors, numerous obvious misses are inevitable for any long-time practicing radiologist, and to deny making them is either dishonest or delusional. (Interestingly, during my training, I had several attendings that would frequently testify for the plaintiffs in malpractice cases. Lo and behold, these were the very attendings known to the residents to often miss obvious findings.)

So what is the moral of this story? Be well trained. Be diligent. Be careful. Be lucky. Did I mention ‘be lucky’? For those of you who think that the ‘bad miss’ is something that the 

radiologist does, think again. The fault, dear Radiologist, lies not in our films, but in ourselves. Let those among you who is without radiographic error cast the first Dictaphone.

1. Bhargavan M, Kaye AH, Forman HP, Sunshine JH. Workload of radiologists in United States in 2006-2007 and trends since 1991-1992. 

2. Garland LH. Studies on the accuracy of diagnostic procedures. 

3. Siegle RL, Baram EM, Reuter SR, Clarke EA, Lancaster JL, McMahan CA. Rates of disagreement in imaging interpretation in a group of community hospitals. 

4. Borgstede JP, Lewis RS, Bhargavan M, Sunshine JH. RADPEER quality assurance program: a multifacility study of interpretive disagreement rates. 

A miss in radiology is inevitable; you just have to be lucky.

Francis Flaherty, MD

Articles

Watching the Cancer Grow

The Radiologist’s Bad Miss