Tuesday, July 12, 2005

Surgical Journal Ripostes

With apologies to Orac, who spends the majority of his time to putting out respectable research, it often seems that the number of worthwhile papers published in surgical journals is infinitesimally small in comparison to the huge total number published each month. As my journals roll in to the inbox every several days, it is sometimes difficult to wade through them to eliminate the obvious ("surgery is good for appendicitis"), the ridiculous ("we can do this easy operation now with a robot!"), and the B.S. ("in our series of 2,000 patients, we have never had a complication").

Sometimes, when papers are presented at meetings, the journal will also publish the discussion that follows. Those papers are always interesting to me, as the discussants often air views about research that perhaps are not widely known. Frequently, the "big hitters" in a particular field will weigh in with their opinion, which may or may not be particularly relevant.

However, what I love to find in a journal is an "invited critique" of a paper, which gives a respected academic physician the ability to summarize the results of the paper and put it into some perspective. No one, and I mean absolutely no one, does this better than Jack Pickleman, the now retired general surgeon from Loyola University in Chicago. He is concise, incisive, and has a deadly command of the weapons of wit and sarcasm. Most importantly, he wastes no time in declaring something B.S. that is clearly B.S. For example, in this month's Archives of Surgery (subscription required, so no link available) he critiques a paper describing yet another incarnation of scoring systems for cirrhotics:

In prior times, the surgeon on rounds would stand at the bedside and observe a lemon-yellow patient awaiting an urgent operation and proclaim, "He's gonna die." Although this judgement was wholly subjective, it was rarely incorrect. Subsequently, systems such as the Child and CTP classifications appeared, but these also suffered from subjectivity. The problem with all such predictors of death was that they did exactly that but provided the surgeon with little guidance in the way of therapeutic interventions by which to thwart the anticipated outcome.....The present study makes a strong case to use the MELD classification to categorize risk in these patients. Until, however, such classifications cannot only assess risk but provide therapeutic recommendations to decrease that risk, I fear that surgeons will now stand at the bedside and proclaim, "His MELD score is ___; he's gonna die."
Dr. Pickleman is basically stating what the vast majority of physicians would say --- judgement matters more than scoring systems for disease processes --- but somehow makes this point so sharply that it is more memorable than the paper he is critiquing.