Is a radiologist with a ruler a radiologist in trouble? I don't think so, but perhaps we can cool it on all these measurements.
During my residency, in the last days of the Age of Films (despite what you may hear, comfortably after the Stone Age), I heard someone quip: "A radiologist with a ruler is a radiologist in trouble."
I wasn't sure what to make of this; Was it that a good rad should be capable of making or excluding diagnoses without relying too heavily on measurements? Or was the original speaker a lazybones who didn't want to include measurements in his reports, preferring to disparage those who did?
As I moved on through fellowship, it did seem to me that a lot of unnecessary measuring was going on in our field. Why, for instance, was it important to include measurements of kidneys, spleen, gonads, common bile duct, etc. in ultrasound reports if they were normal and the clinician hadn't specifically asked for it? And, assuming for whatever reason that it was important in ultrasound, why was it not important to give measurements for the exact same structures in other modalities?
Were the numbers to prove that, between the sonographer and the radiologist, the measured organs had been sufficiently scrutinized? I saw enough non-subtle findings on follow-up CT and MR studies to know that the scrutiny was, shall we say, less-than. Not to mention the regular occurrence of other disparities - like a 9.3 cm spleen from ultrasound turning out to be 14.5 cm when seen by CT later the same day. Perhaps rendered cynical by such events, I theorized that somebody had noticed how much shorter ultrasound reports were than their counterparts from other modalities…and moved to bulk them up with some numerals.
Even when actual pathology was involved, I grew disenchanted with routine reporting of measurements as the years went by. Nobody ever seemed to measure things quite the same way, for instance. I would get a study to follow up a 3.4 cm aortic aneurysm, measure it at 3.1 cm, and then retrospectively remeasure the prior exam to find that it was, by my reckoning, 3.1 cm then, too. Or I would measure 3.9 cm, and retrospectively measure 3.9 cm on the prior. Dictating such reports, that there was no change without essentially besmirching the accuracy of colleagues, was a tightrope-walk.
Were radiologists following up my studies, then, disagreeing with the numbers I reported? If everybody was stuck remeasuring previous exams, what was the point of the previous radiologist dictating his numbers in the first place? To say nothing of the surgeons, etc., going on to measure things themselves and disagreeing with the lot of us.
Much of the time, even as I carefully take measurements and dictate them, I wonder what I'm doing other than cluttering up my reports with supposed evidence that, yes, I actually looked at the study and thought about it in the process. Does it matter to the clinician, for instance, whether the abscess is 5.5 cm or 4.9 cm? Even if it's a follow-up study, am I really giving useful information when I say anything more than that the abscess, following percutaneous drain-placement the previous day, is much smaller?
Part of my sense of non-utility comes from the intermittent stream of communications from clinicians who want specific measurements that I have not given. My favorite is when the ER guys want to know what the appendix measured. Evidently it is insufficient that my impression was "Acute appendicitis, without complications" and the body of my report described "thick-walled, fluid-filled appendix surrounded by moderate inflammatory changes." What is the ER doc going to do differently with an 8 mm appy than he would have done at 11 mm? Not call the surgeon? So part of me really wants to lean towards measuring less, knowing that a clinician will ring me up with a request for numbers if s/he really needs them.
But I go ahead and report the measurements, sucker for radiological peer pressure that I am, because everybody else does it and I was trained that way. And I know that, even if lots of these measurements seem pointless to me, there are legitimate points of decision in medicine which do hinge on certain threshold values: How big an angiomyolipoma should be before it poses enough of a hemorrhage-risk to merit elective removal. How wide the scapholunate interval is allowed to be before ligamentous pathology is implied. How big the ureteral stone is, lest removal be warranted. For every one of these I know, there may be dozens I do not; the data I am tempted to omit may be useful to some other subspecialist out there.
So no, I'm not agreeing that a radiologist with a ruler (or digital facsimile thereof) is necessarily in trouble. I'm just suggesting that maybe it's okay for us to cool our calipers a little.