Wow, yes. This is really strange.
Are you sure than none of them are in mmol/L (roughly a factor of two difference)?
I also had some variability:
Maybe not. Here is my data from Quest on Lp(a)
And here is my data from LabCorp:
without being on any cholesterol related med:
(Have a few other data points around that level from LabCorp)
—
And here is from when being on a half dose of Repatha:
From the last two data points there was enough data for my cardiologist and me to ask ourselves if it might be worth trying a full dose of Repatha instead of the half dose. (A half dose lowered my Lp(a) 25%). This is an experiment we are doing right now.
But what if the only Lp(a) blood test we took was the Quest Diagnostics one from October 2 or 6 2023 where the levels were ~55? Those were BOTH from a period when I had not been on PCSK9i or any cholesterol med for many months.
Either I had found some magical way to lower my Lp(a) via diet or exercise or supplemented that there were no know Lp(a) benefits from… or much more likely those two tests has testing errors.
But if we only had one (or even both of those data points) and thought they was my baseline (a) we would be less concern and I may not have pushed for taking PCSK9i and (b) when we then looked at the Quest datapoints from when I was on PCSK9i in November and December at levels around 75 or so, we could have concluded based on data that PCSK9i increased my Lp(a) level by 35-40%….
Clearly when we see all the data points it’s almost certainly that I’m around 85-100 when on no medication, and it goes down by perhaps 25% to 70ish whenever I do decreases level PCSK9i.
This is just one example of why we should rely on more than one data point when gathering data to make decisions on.
There is always an error bar around any blood test, so you have to think about any one data point as being within a confidence interval that generally can be quite wide.
