h2g2 The Hitchhiker's Guide to the Galaxy: Earth Edition

You may have thought that NaJoPoMo would be an opportunity to find out interesting things about our lives. No So. I thought I'd try and explain one of the more interesting projects that I'm working on.

I am getting interested in how many strokes don't show up on brain scans. It's tempting to assume that brain scans, pretty pictures taken on big shiny machines will give you an answers and cure your problems.

We (doctors) first defined a stroke before there were such things as scans, when the only way to define a stroke, was from looking at the patient, examining them, and hearing their story. When we got CT scans some of the strokes showed up, but most didn't. The big ones showed up, and the bleeds showed up, but the most common strokes, the ischaemic ones didn't show up on the scans. We did the scans, they were useful because they meant that you could see who had a brain tumour or a bleed. But just because someone had a normal scan didn't mean that they didn't have a stroke.

Now we have some lovely modern MRI scanners, which show most strokes up. They have fantastic sequence called DWI which is focuses so that the stroke shows up a bright white colour. Sometimes they can see very small strokes, just a few centimetres across, about 80% of strokes show up on the MRI scans, the newer MRIs show even more strokes.

Some people think that this means that the 20% of people with normal scans, don't have strokes at all, and they shouldn't be treated as if they do. I disagree with that, I think if we do that we'll false reassure some people with strokes that nothing is wrong, and they'll miss out on treatment.

All of our knowledge of strokes comes from studies where they made a clinical diagnosis, so if we changed to an scan-based diagnosis that would mean that we could be certain that all our knowledge was still accurate. No, we have to keep with the classification we have , for now. Maybe one day (when we get the super-powerful scanners that at the moment can only scan rats), we'll be able to say 'your scan's normal, but you haven't had a stroke'.

There has recently been an information campaign on Swedish TV about strokes, and it mentioned three things to check:

1. Can the person smile, showing teeth?

2. Can the person hold up both arms in the air for more than 10 seconds?

3. Can the person repeat a complete sentence?

If the person fails to do any of the above, call 112.

Obviously I wouldn't know what to look for in scan pictures. I am getting good at looking at and understanding ECG readouts though. Having had a fair few over the last decade I can see when things look wrong.

Good luck with the PhD

But, I think I'd fail each of those three tests, most of the time!

Diagnosis is, and probably always has been one of the most problamatic things in medical treatment... Well, obviously you can't treat someone unless you can diagnose them, and diagnosis has to be right, or you'll be treating the wrong thing...
Not just strokes either, which don't necessarily show up on scans, or don't necessarily demonstrate with sufficient symptoms for patents to present to their physition... When my Father had his heart attack, last year, he pointed out to the doctors, that if the symptoms he had, when it happened, was a heart attack, then actually it must have been his second, as they occured twenty or thirty years earlier pretty much the same, only a bit less severe, and after all the ECG and investigations this time round, it did indeed show some evidence of an earlier heart attack, albeit sometwhat smaller in severity
I'm terrible though, If I get a set of really generic symptoms and a sort of general ill type thing, I never go to the doctors, as I know the symptoms are just too vague for any sensible diagnosis, and simularly if I identify whats wrong with me, and its a mechanical fix of some sort to rectafy it, I deal with it myself...

I guess with stroke, though, the actual symptoms of the patient must vary, depending on which part of the brain, and to what extent/how much of that area is affected, and I guess in some areas of the brain, and if only a small area of tissue, there might not be any too obvious presenting symptoms...