Michael Brant-Zawadzki, MD


October 19, 2006

As I join the ranks of the silvering boomers, I find myself less easily embarrassed and caring less about whom I offend. This extends to my observations about us physicians. Having that excuse, I will opine that one of the major differences between a good physician and a great one is that a good physician knows what to do and when to do it; a great physician knows what not to do and/or when not to do it. Unfortunately, one of the major factors in what ails US healthcare is that we have too many good physicians but not enough great ones.

Today I was asked to perform a procedure on an unfortunate patient with widely metastatic ovarian cancer, one with still a good functional status. Despite her systemic anticoagulation and subsequent inferior vena cava filter placement, she continued to show evidence of recurrent pulmonary emboli. A CT scan revealed a large thrombus at her superior vena cava and right atrial junction. Her team of good physicians, fearing imminent life-threatening pulmonary embolus, requested local thrombolytic infusion at the site of the clot, despite presence of bright red blood in her rectum (partly due to her very low platelet count and anticoagulation). I have no idea whether such a procedure would lower the risk for fatal pulmonary embolus, increase it, or make no difference. There is no evidence base for this. There is an evidence base for her stage of metastatic cancer and its prognosis.

Part of our culture as physicians is providing hope, even in the most dire circumstances. Our problem comes in differentiating the fine line between giving hope and selling it. A recent article in the New York Times entitled just that ("Selling Hope," by Alex Berensen, October 1, 2006), discussed a certain chemotherapeutic agent which is approved for shrinking tumors in patients with late-stage breast cancer ­­ approved by the FDA as a last-stage treatment because it is proven to shrink tumors, although prognosis is not altered by its administration. This newer agent costs $4200 per dose. A similar generic chemotherapeutic agent was shown to have less tumor-shrinking capability but proved no worse in survival statistics and costs $150 per dose. Yet insurance companies, employers, and physicians fear that not using the more expensive agent would subject them to the wrath of desperate patients and their families. The company's manufacturer recently applied to the FDA for approval to use the agent, which is 28 times more expensive in early-stage disease. The company's own chief medical officer, however, argued at the FDA hearing that clinical trials should not have to be performed for this approval because such testing would probably not prove that his company's drug was different from the conventional agent.

The New York Times also recently discussed the stent epidemic in Elyria, Ohio ("Heart Procedure is Off the Charts in an Ohio City," by Reed Ableson, August 18, 2006). The rate of coronary stent angioplasty seems to be threefold that in Cleveland. The Elyria cardiologists are quite good at what they do, and can point to some excellent outcomes in terms of technical success and, presumably, in-hospital morbidity and mortality or length of stay. But what about longer-term patient benefit? When it comes to treating coronary artery blockage, there are no definitive studies showing which therapeutic approach most benefits patients in the long term. In fact, the one proven benefit of coronary revascularization is increased longevity after surgically bypassing left main disease. All other coronary revascularization strategies struggle with debates around documentation of long-term benefit. A review analyzing data from New York State hospital registries, published in The New England Journal of Medicine in 2005, showed that mortality rates were actually lower among patients with multivessel coronary artery disease who underwent coronary artery bypass grafting vs among similar patients who underwent percutaneous coronary intervention. Yet the latter are booming in numbers, and threaten to severely restrict the number of coronary surgical revascularization procedures done, perhaps decreasing the skill of the operators or even dissuading medical students from training in that specialty. We have lots of good interventional cardiologists, in Elyria and elsewhere, happy to extol the virtues of minimally invasive percutaneous intervention to patients worried about their chest pain and warned daily on TV about the risk for cardiac death.

The wonders of modern medicine, and the fancy tools, are seductive. Indeed, the wishes of the consumer are playing a greater and greater role in the practice of medicine, partly because the consumer has been shielded from the true costs of healthcare given our insurance structure. Clearly, the patients perceive that they have gotten value from their physicians; whether perception is reality remains a debate among various analysts. In other cases, even when the debate is settled, expensive aggressive procedures are chosen over acceptable traditional care. Look at the remarkable rate of cesarean deliveries in our country, including the elective ones scheduled by modern women for convenience rather than medical need. It's not just interventions that are "sold" to patient-consumers. Diagnostic tools also are a source of questionably valid healthcare expenses. The role of screening serum PSA in its ability to affect disease-specific mortality from prostate cancer has not been proven, yet conventional medicine recommends it for all men over 50. The physician and patient psychology affecting the choice of PSA screening has been explored in a recent JAMA article (JAMA. 2002;119:663-667). The authors wrote: "A physician is positively reinforced for recommending screening, regardless of the test result, because a negative result makes a patient grateful for reassurance and a positive result makes a patient grateful for early detection. A patient who is impotent and incontinent after a decision for curative treatment may attribute his survival to surgery and be grateful for having his cancer cured."

A recent article (J Am Coll Radiol. 2005;2:411-414) pointed out the dramatic increase in healthcare expenditures for diagnostic imaging, particularly coincident with the proliferation of clinician in-office high-end scanners. Its author, Dr. Thomas Fletcher, discusses the markedly increased rate of referral for CT and MRI studies by physicians who own the equipment and can refer their own patients to it, as compared to independent facilities owned by hospitals and radiologists. Do such referring physicians unconsciously order potentially unnecessary examinations to help meet expenses? It is hard to know.

The negative economic impact from the dramatic rise in use of high-end imaging procedures by self-referring clinicians has been recognized by Congress, with recent legislation about to be enacted aimed at cutting reimbursement for diagnostic imaging across the board.

As radiologists, we see and benefit from the trends towards selling physician services uncritically. Spine surgery procedures are done 10 times as frequently per capita in the United States as in Scandinavian countries, 6 times as frequently as done in England. The combination of a demanding patient, one who may gain secondarily from aggressive therapy through the workman's compensation and disability insurance compensation; many willing doctors with very good technical skills for which they are compensated; and lack of evidence base as to the most beneficial treatment for chronic back problems is certainly one cause. Five-year outcomes of surgery for degenerative back problems equal those of conservatively treated patients. Radiologists as well as surgeons benefit from the associated imaging and aggressive management of such patients.

Another example is that of carotid endarterectomy. Although patients with symptomatic carotid artery occlusive disease clearly benefit from surgical intervention, the case for asymptomatic carotid occlusive disease is still debatable. Of several randomized trials done, only 3 document marginal improvement in stroke prevention by operation on carotid stenosis of more than 60%. These studies disagree with each other on whether greater degrees of stenosis provide stronger indication for intervention in asymptomatic patients, and differ from each other in terms of patient selection and methodology for measurements of stenosis as well as outcomes. One of the first studies to suggest benefit, when analyzed, indicates that one would need to do 52 carotid endarterectomies to prevent 1 minor stroke in the asymptomatic patient (no difference in major stroke rates ­­ those producing lasting deficits beyond 30 days ­­ was documented between the 2 approaches). Recent studies of statins in heart disease patients show that such drugs reduce the risk for stroke by 50%! We are about to embark on an explosion of endovascular stent angioplasty for asymptomatic carotid disease. The manufacturers of the devices, the operators with strong technical skills who can do the procedure as safely as a surgeon, and the patients themselves (scared regarding future potential of incapacitating stroke) see "hope" in the scenario of the minimally invasive approach. Do interventionalists suggest to patients with asymptomatic carotid stenosis that aspirin and statins may do as much good as their manual expertise? If told, would the patients accept the conservative approach? It's been extremely difficult to convince the NIH to fund a randomized study to get the answers, as it would be expensive. More expensive than money spent on still-unproven ­­ but very sexy ­­ technological intervention?

Although "selling hope" may be too strong a phrase for some, consider also the fact that the major share of the healthcare dollar is spent in the last year of life. Nevertheless, life spans are being extended through the tremendous scientific advances of modern medicine. Somehow, the chaotic process we have in place, despite its considerable financial costs, does produce some beneficial results.

In the end, whether the apple keeps the doctor away or provides too early an exit from the earthly Garden of Eden has yet to be determined.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: