Showing posts with label dialysis. Show all posts
Showing posts with label dialysis. Show all posts

Monday, October 2, 2017

Lithium Toxicity: To Dialyze or to Monitor?

(Hypothetical) Case scenario: Patient comes in, accidental overdose of lithium, HD stable, physical exam unremarkable, labs pristine, lithium level of 2.x. Poison control called. ED interventions performed flawlessly. Nephro consulted in the ED and recommends HD if level is above (insert arbitrary number here). On recheck several hours later, pt looks GREAT clinically but the lithium value has increased closer to (insert arbitrary number here). 

Now, I'm an intensivist and make my living off of placing vascular catheters but only if there's appropriate data behind it. It is time to look up the data. No, I won't even pretend to know everything. 

Hey, look! There's a Cochrane review about it!
Full (free) article: http://onlinelibrary.wiley.com/doi/10.1002/14651858.CD007951.pub2/epdf



Here are a couple things that caught my attention regarding this review: 
"No randomized controlled trials of hemodialysis therapy for lithium poisoning were identified."
This is definitely reasonable. I mean, imagine the complexity of consenting these patients who are on lithium already for an unfortunate reason and have overdosed for one reason or another where their capacity may be questioned. IRB permission, sigh.  

Hence, after a very thorough review of all the data including the more recent (2015) Extracorporeal Treatments in Poisoning (EXTRIP) workgroup systematic review, the author concludes that: 
"Although the use of hemodialysis to enhance the elimination of lithium in patients with lithium poisoning is logical, there is no research from randomized controlled trials on its benefits and harms in patients with lithium poisoning. Most patients with lithium poisoning recover fully, but the available data do not provide a reliable way to predict which patients will have a good or poor outcome. Until higher-quality evidence can be developed, the decision to use hemodialysis in addition to standard therapy with intravenous fluids will continue to be based on clinical judgment. Hemodialysis treatment should ideally be given as part of a randomized controlled trial."
This conclusion is to be expected, we like data, any good data, and we just don't have it here. 

The fine investigators in the EXTRIP workgroup published this fine piece of work which gave way to the table listed below. 
Full (free) article:http://docs.wixstatic.com/ugd/363318_f36816033c994d129d7663213a31ea71.pdf



This makes a lot more sense. Choosing an arbitrary number has no data behind it. Fortunately, my patient has great renal function, a stable mental status, and her heart is lovable. Let's hold off on HD. At least that's what I'm going to do, for now. 

In the words of one of my favorite attending physicians during my fellowship, sometimes you have to just "do something". I'm not going to listen to him this time. 

A big hat tip to the researchers involved because I just dislike doing research. 
-EJ