Failure as teacher.

Drs Singh, Graeb, Fung, and Teal (1) have published a case report in this edition of the American Journal of Neuroradiology. Read it. It is a gripping story. In brief, a woman suffered a myocardial infarction. One week later she had a stroke. The middle cerebral artery embolus was successfully lysed with intraarterial urokinase. Then her heart ruptured and she died. Why publish such a case report in the AJNR? Confession cleanses the psyche; that is well known, but the AJNR is not in the business of providing for the psychotherapeutic needs of our neuroradiologic brethren. Medical journals, especially those that are clinically oriented, have a mission to foster the exchange of ideas and information of use in understanding disease and treating patients. What are the values and utilities of case reports in achieving that mission? Are they just padding for academic CVs, as some cynically believe, or do they have a true function in helping to serve patients in need? There are, in fact, several categories of case reports that truly are significant. Case reports deal with rare events, events sufficiently unusual that a series of similar cases cannot be accumulated. Rare events are not perforce insignificant; a single event can point to a concept that has important implications for more common situations. A case report can document a rare event that illuminates a new facet of a more common condition (eg, Hecht et al [2]). Some case reports alert members of the medical community to emerging new conditions that they might soon encounter. AIDS, Lyme disease, and the MELAS syndrome are just a few of the conditions that have been initially reported and described in recent years. We can expect to encounter many new maladies in the years to come, and many of the introductions will be via case reports. Some case reports alert colleagues to potential complications, which they can then avoid. Personally, I am glad Dr Singh and colleagues have already sensitized me to the risk of cardiac rupture complicating cerebral intraarterial thrombolytic therapy, rather than waiting to develop a larger series of similar cases before publication. Aware of this possible complication, I can now work to avoid it as I select patients for thrombolysis. Much as we detest them, complications are intrinsic to the practice of medicine. Important as they are, we tend to lump complications generically into a single, contemptible group of untoward events that lead to undesired outcomes. If we classify complications more precisely into groups or types, we can gain insights into their etiologies, and from there we can work to change the patterns that lead toward problems. In Forgive and Remember: Managing Medical Failure (3), a study of decision making in a surgical service of a teaching hospital, sociologist Charles L. Bosk divides medical errors into several types. Technical errors are those in which an operator’s skills fall short of those required by a task. All physicians make technical errors. As physicians gain experience, their frequency of technical errors will decrease, but technical errors can occur at any time, even to prudent physicians. Judgment errors are the result of selecting an incorrect treatment strategy. Curiously, the incidence of judgment errors can actually increase with experience, since more critical decisions tend to be made by more senior practitioners, and when those big decisions go wrong, they go wrong in a big way. Emotional as medical decision making tends to be (4), we evaluate the quality of decisions based on their results, rather than on the underlying scientific logic and reasoning. Bosk characterizes normative errors as those that result from a failure to discharge a role obligation. Normative error would include failure to disclose important events fully to colleagues, or even lying. Whereas all physicians make technical errors and judgment errors, most physicians go through an entire career without ever making a single normative error. Because they so severely undermine the conventions that underpin the social system we depend