ekg math

[pbruss]

there is some studies that show an abnormal t wave axsis and qt interval in a STEMI. i went through the protal and found 35 normal ekgs and 20 stemi ekgs and ploted the relationship between the 2 (see attached). the data sugests that there is a linear relationship between the 2 and that relationship is altered in a STEMI which makes sencs as the normal flow of electricity through the myocardium is altered by cell damage.  interestingly if you take the equation for the normals qt=(0.0736*t axsis)+358.49 and plug the normal t axsis values in you do not get a number greater then 365. and if you us the same equation and put in the stemi t wave axsis the reult is always greater then 365. my goal is to do more values to confirm the mathmatical relationship so in theory all you have to do is plug in the computr read otu for the t wave axsis and tell if someone is having a stemi or not. let me know what you think

Attachments:

Presentation1.pptx (208.7 KB)

[tyson]

that is def interesting.

how many studies have been done on this?

is this equation valid once you start adding in fascicular blocks, bundle branch blocks, electrolyte abnormalities, previous MI's etc....?

[pbruss]

so no stdues on this that i can find. for the data here i did not use eks with fasicular or bundle branch blocks for the normal or stemi group. however i do think the same reltationship should hold true for these blocks but would need a different mathmatical equation. the theory here is that the ekg is just measuring the electrical activity in the heart which if you think about it is just a vector like in physicis  and the computer caluclates that vector for you. the damage to the heart muscle affects that vector and i am trying to explain that mathmatically so all you have to do is look at the numbers the ekg give you and tell if this person is having a heart attack or not. theoretically it does not matter what underlying block or electrolyte abnormality the person has because the infarction would cause a focal problem in the heart muscle that would alter the mathmatical relationship with these vectors. im working on more data nd will keep you updated.

[slevittMD]

Definitely sounds interesting and would be something if an equation can pop out a number that accurately predicts STEMI vs. not. I don't have an EKG in front of me to reference, but what are you using for the x and y axis? I'm assuming the y axis is the QT interval, but not sure where the x axis information is coming from? Is t-wave axis something typically printed on the EKG? Have you tried this with something like dewinters or aVR elevation to see if it holds true there?

[pbruss]

all of these numbers came right off the ekg. in the top right corner the computer gives you the qt, qtc, and axsisi for the p,r and t waves.

[Bjs04f]

All of this is very cool, but after another bullshit decline by cards, I fear we may get more out of just getting cardiology behind what we do know, as they still dont have basic EKG down. Most recent was a wellens type two huge change from prior with cp that was crushing and had resolved in 80 yo guy. Dude also has pathologic Q waves in V1-3, so I go its wellens and hes got what looks like q wave MI.....not impressed heparin and floor

[slevittMD]

I think we should get trained in cath and start doing them ourselves. I bet we have better outcomes without having to worry about selling to cardiology.

[pbruss]

i tried to do that. technocally you can if the hospital credentials you but i found the hospital administration is reluctant if you are not a cardiologist. unfortunatly US interventional fellowship will only take IM trained residents. Europe will take er trained docs in their felowships but you need to live in their country for a year before being eligable and the US does not always recognise european trained interventionalists.