About Me

My photo
I'm a radiologist and writing helps me make sense of the world.

"My method is to take the utmost trouble to find the right thing to say, and then to say it with the utmost levity" -George Bernard Shaw

Thursday 8 October 2015

More Rules of Radiology

Another 20-odd Über-truths, to be read in conjunction with The Rules of Radiology. Remember Rule #1 : You must obey the rules...


Rule #41 / / Radiologists don't wear suede shoes
Or brogues, open toed shoes or sandals. You'll know why once you grasp the concept of secondhand barium.

Rule #42 / / Never investigate on the day of discharge
Such tests always turn up something unexpected. It'll prove ultimately benign but will take a week to sort out.

Rule #43 / / Only give clinicians 15 seconds
If they can't cut to the chase, help them. We haven't got all day. The longer the preamble, the lower the pre-test probability.

Rule #44 / / You can never know too much anatomy
The more anatomy you know, the better the radiologist you become. It's the only thing you learn at medical school that won't have changed by the time you retire.

Rule #45 / / Be careful with "limited" or "quick" studies
Guaranteed you'll miss the cancer. Better off doing the full Monty or nothing at all.

Rule #46 / / Image quality is up to the radiologist
Image noise, coverage, adequacy of position and so on are all dictated by what you are willing accept, not by what was produced.

Rule #47 / / It is never a chordoma
Rare presentation of a common disease is commoner than common presentation of a rare disease. That classical rare bone lesion will turn out to be just another bone met.

Rule #48 / / Reporting rooms should be pitch black
If you report with the lights on, you are letting the entire profession down. Reporting in a dark room is for logical visual reasons but it also handily disorientates those who trespass.

Rule #49 / / Don't touch the screens
Not with greasy fingers, never ever with pens. The dirtier the screen you'll tolerate, the sloppier a radiologist you are.

Rule #50 / / Error is inherent to radiology
Get used to it. The images are subjective. Every cancer starts really, really small. The human body is really, really complex. See Rule #51.

Rule #51 / / Savour mistakes
And near misses too. Even the best radiologists can make the worst mistakes - you cannot know it all. But everyone can learn.  See Rules #38 and #50

Rule #52 / / Don't rush a report
Clinicians bizarrely expect it instantly. The more complex the scan, the sicker the patient, the closer they crowd you. Dispel them with 'You can have the wrong report now or the correct report in 15 minutes'

Rule #53 / / Never wake a patient up
If they are sound asleep, they don't need a scan right now. Sleep is good for ill people. However, probably wise to double check they are actually asleep and not moribund.

Rule #54 / / Know about esoterica
If you hear hoof beats and see stripes, it could be a zebra. There are so many rare conditions that it is common to have at least one of them in the hospital. See Rule #39.

Rule #55 / / Doctors aren't porters
If you let them just 'drop off a form' and leave without another word then you are doing them, you and the patient a disservice.

Rule #56 / / Don't shoot the messenger
It isn't the poor newbie medical houseman's fault that the 'surgeons want a CT before they see the patient'. Instead of ripping them a new orifice, save your ire for the original miscreant. See Rules #36 and #37

Rule #57 / / Question 'protocols'
If the major reason for a scan is due to 'the protocol', ask to see it 'for my education'. The protocol in question usually either doesn't exist or states exactly the contrary.

Rule #58 / / Don't be a hairdresser
Hairdressers never say 'you don't need a hair cut'. Question the motives of those who are paid per scan, especially when they recommend expensive additional studies.

Rule #59 / / Prognostication is not an indication
If someone is so ill that they are not fit for a haircut then a scan isn't going to change anything. It is wasting everyone's time. See Rules #15 and #64.

Rule #60 / / Beware of Mr Twitchy
If a patient can't stay still, abandon it straight away. Get them back another time. Otherwise you get asked to interpret a twitchogram. And that never ends well.

Rule #61 / / Always look at the scout image or localisers
That 10cm RCC isn't on the sag T2. And that basal lung cancer isn't on the volume dataset. Ignore them at your medico-legal peril.

Rule #62 / / Dictate considerately
If a secretary 6 rooms away can clearly transcribe your every word, you should probably speak a little more quietly. You are also probably irritating your colleagues beyond belief.

Rule #63 / / Don't scan instead of talking
Resist pressure to omit clinical discussion. Talking is cheap, quick and can avoid scans altogether. It also has relatively few serious side effects.

Rule #64 / / Never scan the dying
It is highly distressing, undignified and tantamount to assault. No amount of diagnostic electromagnetic waves will stop Mother Nature. See Rule #59.

Rule #65 / / Don't be a smart arse
Scans often only become a waste of time after you've done them. It's easy to be wise with hindsight: only a facile radiologist does this.

Rule #66 / / False positives are errors too
Over-calling is as equal a sin to under-calling. If you can’t report a CXR without asking for a CT, you need to take a long hard look at yourself. See Rules #29, #50 and #51.

Rule #67 / / Vetting requests is worthwhile
If you let other people do it for you (or don't do it at all), you cannot complain about unjustified scans or scrambled clinical reasoning. Man up and JFDI.

Rule #68 / / Be a holistic radiologist
Look at the whole image not just your area of interest. You may be the cleverest spinal radiologist since the Earth cooled but missing a 7cm AAA on MRI L-spine is never clever .



  1. Very interesting , invaluable rules, especially when it comes from some one whol himself applies to his work.
    Thanks Paul,was a pleasure working with you.

    1. Thanks Hala, always lovely to hear from you, especially such kind words too!

  2. By your picture, i would not expect so much wisdom. You really seem young!

    1. Or maybe I should try glasses

    2. Very kind. I'm 49. Not old enough to be wise but old enough to know better!