There is more rejoicing in the chambers of malpractice lawyers over one missed diagnosis than over ninety-nine dangerous, unnecessary and expensive investigations. It is therefore unsurprising that doctors are particularly anxious not to miss pulmonary embolus, a clot in the pulmonary arteries, which has been called a silent killer. There are several factors that predispose to PE, as it is known, among them age, a recent operation or having sat for a long time in a confined space. A friend of mine collapsed with a PE at home one day, and it was then that his polycythaemia rubra vera, a chronic overproduction of red cells, was diagnosed.
If PE can kill you, and if there is an effective treatment, it might seem that an improvement in the ability to detect it is to be welcomed. But, as is often the case in medicine, matters are a little more complex than they seem. As a recent article in the British Medical Journal put it:
If all pulmonary emboli caused important harm or death if untreated, finding more small clots would be an unqualified advance. However, there is evidence that some small clots do not need treatment…
And while new technology allows more small clots to be detected that would not have been detected by older methods, it does not give any guidance as to which of them should be treated. Since treatment (with anticoagulants) is itself not without risk, it is possible that the increased detection of small clots does more harm than good.
In the 8 years after the introduction of a new technique called multidetector CT pulmonary angiography, the rate of pulmonary embolus in the U.S. rose by 80 percent, from 62.1 to 112.3 per 100,000 adults. Part of the problem was that the new equipment, being very expensive to install, had to be used to justify its purchase. The result was many more PEs.
Treatment made no great advances in those years, and the overall rate of death from PE in the United States remained more of less constant. However, the fatality rate of detected PEs fell greatly, from 12.3 to 7.8 percent, suggesting that most of the PEs that were now being detected that would not previously have been detected were harmless.
The patients in whom such emboli are found tend to be treated nevertheless, on the precautionary principle. But in the largest study of patients with small PEs, the risk of recurrent embolus was only 0.7 percent while the risk of major bleeding caused by anticoagulation was 5.3 percent. In other words, the treatment seemed more hazardous than the disease (a common occurrence in the history of medicine), though this study did not answer the question of long-term mortality. Perhaps if the patients were followed up long enough, some advantage to treatment would reveal itself; but so far it has not.
Overdiagnosis of PE causes patients needless anxiety because they are told that they have a potentially fatal condition. If they are treated with anticoagulants, they spend possibly a liftetime worrying about the dangers of bleeding if they are injured. Moreover, some patients actually will be harmed by anticoagulation.
The latest technology tends to be used if available, and patients demand it because they think it will necessarily and inevitably benefit them. Often it will, but only if used with discretion. Patients are sometimes scanned uphill and down dale, or thoughtlessly put into what a physician friend of mine calls “the answering machine.” The sad fact is, however, that for the foreseeable future physicians will have to exercise an indefinable quality known as judgment, which one hopes grows with experience.