CT Scans May Be Too Efficient in Detecting Pulmonary Emboli
Posted Jul 25 2013 12:00am
I am fascinated by the idea that some forms of imaging in radiology are too good. One aspect of this "too good" phenomenon is the so-called incidentaloma about which I have previously blogged (see: How to Avoid the Risks of a CT Incidentaloma ; More (and Interesting) Discussion about Incidentalomas ). An incidentaloma is an "incidental" radiology finding that is reported and may prompt unnecessary intervention. A recent article address another form of "too good" radiology reporting relating to the detection of a pulmonary embolus (see: CT Scan May Be Too Good at Finding Lung Problems, Study Finds ), In this second type of "too good" reporting, the radiologist reports the presence of a pulmonary embolus to the clinician, the detection of which was the purpose of ordering the CT scan in the first place. However and due to the sensitivity of the CT scan, the lesion reported may be so small that it will usually resolve on its own without clinical intervention. Below is an excerpt from the article:
CT imaging is a very good way to find pulmonary embolisms — sudden blockages of an artery in the lung, which can be fatal if not treated quickly. But a new analysis finds that the procedure might be too good, revealing tiny embolisms that are harmless and leading to treatments that are unnecessary, expensive and potentially dangerous. Until the introduction of CT angiography in 1998, the primary test for pulmonary embolism was ventilation/perfusion scanning, in which inhaled or injected radioisotopes are used to create an image of air and blood flow in the lungs....But when CT angiography was introduced, doctors enthusiastically adopted it. It is now widely used instead of ventilation/perfusion and often employed to find the causes of various other lung disorders.Its versatility is only one reason for its popularity. Missing a pulmonary embolism can be a fatal mistake that doctors obviously want to avoid. A CT scan can reassure them and protect against claims of malpractice....CT scanning devices are expensive, and when a hospital buys one, officials want to use the machine as often as possible to justify the investment. “There’s a lot pressure to order the most advanced tests, no matter how small the problem,” said the lead author of [a new study], Dr. Renda Soylemez Wiener....According to the report, published on Tuesday in the journal BMJ, the number of diagnoses of pulmonary embolism was steady at 62.1 per 100,000 in the five years before the introduction of CT scanners. In the eight years after their appearance, pulmonary embolism diagnoses increased by 80 percent. Despite this, age-adjusted deaths from pulmonary embolism hardly changed at all. Treatment of the condition has not changed and is no more effective, but in-hospital deaths in diagnosed cases have decreased to 7.8 percent from 12.1 percent. In other words, the report’s authors said, CT scans are finding many pulmonary embolisms that are harmless and should never have been treated. According to [one of the study authors], 80 million scans are performed every year in the United States, about a quarter of them unnecessary. “The real concern is that the treatment is so dangerous,” Dr. Wiener said. “In the case of these very small emboli, the treatment may be more dangerous than the pulmonary embolism itself.” The standard treatment — blood-thinning medication — is a leading cause of drug-related death, and patients may end up on lifelong treatment to prevent recurrent embolisms.
So the CT technology places physicians between a rock and a hard place. The radiologist must reveal the presence of the embolus, albeit small, and the clinician receiving the report may feel that he or she must treat the reported lesion on medico-legal grounds. It seems to me that there is some middle ground here and I am sure that such a scenario occurs countless times in many hospitals. The radiologist only need report that the visualized embolus is small and that the radiology literature indicates that most such small lesions need not be treated. Life-long anticoagulation therapy is more risky for the patient than the small embolus itself. The clinician receiving such a report mirrors this opinion in his own clinical notes. He or she indicates that an embolus identified in the radiology report is small and the medical literature supports not treating it. Physicians on both sides, diagnostic and therapeutic, are thus protected if the patient's pulmonary emboli recur and most patients come out ahead.