Monday, March 16, 2009

It's never too late to normalise your glucose (so long as you are still alive)

I've just plotted the risk of being dead in the near future vs post admission glucose for coronary patients in this study. Haven't seen the full text so I've no idea whether there was an acute drop in LDL cholesterol in the survival patients, to keep cardiologists happy, but I guess not or they would have mentioned it. In fact, if you had to guess, you could do worse than to guess that an acute drop in LDL would come accompanied by a black cloak and scythe.

Anyway here are data from the abstract in graph form. Enjoy.

Although it is possible the hyperglycaemia is the problem per se, it is equally possible that it simply reflects underlying insulin resistance. But I doubt anyone is going to ask for an IV glucose infusion plus some diazoxide post MI to find out!



Mark said...

Yeah, I remember those diazoxide studies. The first study found it effective in weight loss. The second used a completely different panel and compared diabetic or prediabetic patients with normals. The first panel was probably fired.


ItsTheWooo said...

I'm more inclined to believe glucose control deteriorates when patients are close to death. Working in a nursing home, I know that one of the hallmark signs a resident is declining is that their blood glucose becomes uncontrollable. This does not cause death in so much as it precedes it, possibly because dying creates a hypercatabolic state, stress hormones like cortisol pour sugar out from breakdown of tissue, which the dying tissues can no longer use for energy as, well, they are being broken down and dying.

Peter said...

They gave insulin to 39% of their patients and this lowered glucose but didn't normalise it in all patients, no full text so no info re the insulin protocol, or the potassium protocol, which may have some bearing (structured water etc). It would be criminal to give insulin without potassium homeostasis to anyone, let alone post AMI.

But the post admission glucose, with or without insulin, appears to be the predictor. That doesn't sound like its a pre death failure of everything. I'd be more suspicious of either NFkappaB mediated pro inflammatory issues or direct toxicity.

There's a helpful histogram chart here at the bottom of the page.