Links Mentioned in Video
- Online calculator to correct total phenytoin for low albumin: http://clincalc.com/Phenytoin/Correction.aspx
- Online calculator to correct total phenytoin for concurrent valproic acid: http://clincalc.com/Phenytoin/WithValproicAcid.aspx
Video Transcript
I’d like to briefly discuss the difference between free phenytoin, total phenytoin, and clinically why it matters.
I’ve drawn 9 albumin molecules represented as these black circles, and then we have these phenytoin molecules that I’ve represented as red circles. Now, phenytoin is a highly protein-bound molecule, and about 90% is going to be bound to protein, and about 10% is considered “free”. The reason that the free portion matters is that the free is the biologically active portion. Because it’s not bound to protein, it is able to cross the blood brain barrier, and exert its pharmacologic effect.
As you can imagine, clinically, we’re more interested in the free portion because it really represents the patient’s true status of phenytoin. But historically, what was done, is labs would draw a total level. So if you look at the diagram above, you can see that we have a total of 10 phenytoin molecules, but the free number is 10% of that, which is 1.
So again, historically what was done is total phenytoin levels were drawn, it was assumed that about 10% was free (biologically active), and this correction factor was made. The problem, though, was that this free fraction of 10% is not always valid in all patients. Some patients don’t have as much albumin as other, especially critically ill patients. So as you can see, I’ve removed one of the albumin molecules, and now, we don’t have 10% of a free fraction. We still have 10 total molecules, but now instead of 1 free molecule, we have 2 free molecules. Again, nothing has changed in terms of the total amount of phenytoin — we still have 10 molecules, but we’ve double the amount of free (biologically active) phenytoin from 1 to 2. You can imagine that in very critically ill patients who have very low albumin levels, the free fraction can be even worse — something that needs to be accounted for.
So as you can see here, we have 3 free phenytoin molecules, and we still have the same total amount of phenytoin. A correction factor definitely needs to take place. And this correction factor is called the Winter-Tozer equation. The equation is available online (http://clincalc.com/Phenytoin/Correction.aspx), but essentially what you need to know about it is that it takes a total phenytoin level and estimates what a free phenytoin level might be given a certain albumin level for a patient.
So that’s one problem with total phenytoin levels. The other problem is that because phenytoin is highly protein bound, other molecules, like valproic acid, that are also highly protein bound, can compete with the binding site and actually “kick off” phenytoin molecules and make them turn into free phenytoin because they’ve competed for the binding site.
So, as you can see above, this valproic acid molecule has “kicked off” phenytoin. Now we have 4 free phenytoin molecules. We still have the same total number of 10, but now we have a higher free fraction. There are equations that can estimate what a free phenytoin level would be with a certain valproic acid level, but these equations aren’t great. Clinically, the best practice is to try and get a free phenytoin level. Unfortunately, not all labs can do free phenytoin levels on demand. Sometimes they are send-out labs, and it can take a while to get back. In that case, doing a total phenytoin level with a correction factor might be important. But again, the free phenytoin level is definitely the standard of therapy.
So the next time you see a total phenytoin level, think to yourself two different things. One, what is the patient’s albumin status? Does a correction need to take place? And two, concurrent medications? So is something like valproic acid, that competes with the binding site, can dramatically alter that total phenytoin level to the point where it’s essentially inaccurate unless a correction factor takes place.