Saturday, 16 December 2017

Perioperative Management of INR

Eventually one finds that some sort of small surgery is needed, for those of us on warfarin (perhaps after a large surgery such as a heart valve replacement) the management of INR around that is important. In my case a colonoscopy was required.



This post is about my experiences in the perioperative management of my INR around my colonoscopy and thus is here also as a guide for those who (like me) self manage.

Starting point

My Gastroenterologist requested my INR to be managed down to ≤ 1.5 pre the 'procedure', which formed the basis for this "experiment" in validating my INR management and the predictive usefulness of my simple data model for INR.

Summary position


  • With good data you can manage your INR down to a suitable level for a small procedure probably without need for heparin (but be prepared for needing it)
  • I have again verified my model in an actual experiment but with better data gathering than before
  • I was perhaps too conservative in my management strategy (but who knows)
  • the outcomes were all good
  • I took a Heparin shot in the middle of the 'recovery of INR' phase (for "just in case"), pehaps it was too conservative but what the hell ... 


Premise

I manage my own INR (as you'll find on this blog for instance here). In making this a bit more predictable (rather than just gut feel heuristic) I have developed a data model on how my INR behaves. Previous surgeries (some yars back now) have given me some data to work with, and years of managing myself has allowed me to collect and analyze response due to various "oops" events such as "I missed a dose".  This post is based on that and my reading and understanding of the literature. In particular the following article is a useful advice on exactly this topic:



found in full here. I urge you to read it.

From that article I found the following points of significance which guided my strategy:

Summary
The perioperative management of patients on long-term warfarin therapy poses particular problems. This situation is exacerbated by the absence of randomised trials. The strategy used is based on the assessment of each patient's thromboembolic and bleeding risks. These determine the need for withholding warfarin and switching to heparin. Most patients having minor procedures can continue to take warfarin, provided that they are closely monitored and local measures are used to ensure adequate haemostasis
My primary take outs are highlighted but as always, read the above carefully and see it in context.

Risks of temporarily withholding warfarin 
The risks are difficult to quantify due to the lack of randomised trials examining this issue. They vary according to the indication for the warfarin therapy.
The article goes on to iterate some major risk groups:



Which in my case is the simple fact that I have a modern bileaflet mechanical aortic valve ... which puts me in about the lowest risk category ... the same may be true for you too.

Next:

Do the benefits of anticoagulation outweigh the risks?
The approach to the management of anticoagulation in patients with prosthetic valves undergoing non-cardiac surgery remains controversial. The need for perioperative anticoagulation in patients with mechanical heart valves has been questioned in a recent review. The authors argue that for every 10 000 patients with mechanical heart valves who are given perioperative intravenous heparin, three thromboembolic events are prevented at the cost of 300 major postoperative bleeding episodes

Think about those numbers the prevention of three thrombo events vs 300 major bleeds. Note "The authors" refers to the study being cited by that article. That study (referenced in their bibliography of course) is "Kearon C, Hirsh J. Management of anticoagulation before and after elective surgery. N Engl J Med 1997;336:1506-11." and is found here.

Now as a background, one of my friends (about the same age as me, but not on warfarin) had a colonoscopy the week before me. He had a life threatening bleed (and was taken from his home by ambulance to a local hospital then transferred to a more major hospital when they couldn't stop the bleeding.

I have also had a friend die of a GI bleed ... Clotting has a major and important role in survival.

Lastly the literature is full of examples of people who have had low INR for extended periods that have come to no harm ...


Shit like this takes a long time to form, who knows how low for how long they were. So maybe we are just too conservative? Probably that's a good research question too.


So I took my management of this from that perspective;
 IE my risk of a clot (low) vs the (higher) potential for a bleed.

Management Strategy

I decided to monitor my INR daily and to cease warfarin a day earlier than my model predicted I would need to cease to achieve the target that the Gastroenterologist had given me. I wanted to do this because I wanted to make sure that bleeding wasn't a factor.

My process was this:

  • measure INR daily (in the AM, usually about 8am)
  • my usual dose time is 7pm
  • I recorded INR and the dose
  • the graph below has INR on the LHS Y axis and dose in mg on the right axis
  • the bars represent actual data and the lines part of my model
  • I ceased warfarin on Sunday (I take in the PM)
  • my colonoscopy was Wednesday around noon

So, lets look at the data.


As it happens my model was quite close in predicting my INR reduction (represented as Pinr 3 period moving average), which fell to below 1.5 by tuesday (as predicted by the model) as well as indicating the rise in INR (Pinr 4 period moving average).

This meant that I went into my procedure with my INR of 1.2 (measured in the AM) for a 2pm procedure. This is consistent with my goals of reduced bleeding complications but a little conservative (it was my first try) and the target could perhaps have been achieved by ceasing a day closer to the procedure.

I decided to not recommence warfarin on the evening of the procedure but wait for the next day. Again it is on this point I feel that the collected data shows that  I was a bit too conservative. I believe that I could have resumed warfarin that evening and the data shows this to be correct. Making for a reduction in days outside the therapeutic window of two days less time.
NB: from 9 days to 7 days (so far).

Due to my conservative restart my INR continued to fall on Friday (down to 1.1) and so in prudence I went to my local hospital seeking a heparin shot. I (happily) met a very good young Doctor to whom presented my INR data in support of my request for the heparin shot. He reviewed my data and was quite supportive of my approach. You see, if you're organised and detail oriented you can get support from the medical community ... even if you fall outside the box.

So I could thus have perhaps restarted a bit more aggressively, but then I was concerned about the possibility of a bleed in the period of time directly following the procedure. Perhaps more conservative but consistent with the wish to avoid a GI bleed.

The strategy I employed was (using my model) to get my warfarin levels back up to a level which would resume my anticoagulation as before the process began. I did so in a manner to avoid any over shoot "spikes" in INR (due to over dosing) while allowing the INR to climb in a manner which is reasonable. My model showed me that by day 5 (of resumption of ACT) after the  I would be coming within "safe" territory, and by day 6 I would be within my desired therapeutic range (2.0 ~ 3.0).

Knowing this "delay in INR restart" I could safely shave another day off by resuming directly in the PM of the evening after the procedure, and perhaps with a larger bolus dose to start me off. This would result in an overall reduction of 1 days outside the therapeutic range making it 6 days in all.

Discussion

This process has enabled me to feel much more comfortable in managing my INR and has given me added validation that my (albeit simple) model of INR behaviour in my system works. Certainly there are parameters which need better tuning, but in terms of Pareto Principle we have 80% of the benefits here already.

Based on my model I feel that I could have reduced my window of withholding my warfarin by ceasing later and resuming later. This would have still had my INR at less than 1.5 for the procedure and raising to INR >2 sooner. By ceasing as soon as I did it allowed my INR to drop further than needed and take longer to then recover the desired levels.

However is this result bad? No, I don't think so. For the need of coagulation in my GI tract is not known, meaning it may be that I benefited from the extra time to heal. So for the sake of a single heparin shot two days after my procedure (on Friday, restoring AC levels required and perhaps also washing out / allow the destruction of any early stage thrombosis) I was able to give good coagulation to the site of injury (some polyps were removed) and act in a prudent manner.

Lastly my model has given me something important; Quantative  INR recovery estimates. This is a good source of confidence. We rely on "knowing" the future to feel less anxious about things. Knowing nothing is what you do when you gamble, knowing something can take much of the anxiety out of things and helps plan.

Much of the operation of my model is not clarified here in this blog post. That is deliberate, for until I decide what to do with my intellectual property I'd like to keep the cards close to my chest. The accuracy of the model after reaching INR 1.5 seems to be wanting. I believe this is because there are "lags in the system" which means that the INR response to the warfarin may be delayed. I normally factor in a 4 day rolling average to the model to predict the rise. As you can see it is close to that, but not perfect ... meaning there are some more research questions arising from this research.

I hope this has been of benefit to some of you.

If you are a valver, and want to work with me on your INR management please feel free to comment with your email address. Comments are moderated and I will not publish a comment with an email address.

I won't charge you anything for my assistance but I will require you work in a rigorous manner in collecting data, and taking your warfarin. I will also need a base data set of some months of weekly INR readings to build your data model from. Without all of that ... it just can't happen.

Last but not least:

Acknowledgements

I would like to take the opportunity to thank my Gastroenterologist for her great team and well executed event. Everyone was professional, caring and understanding. Asked good questions and gently demanded good answers. Everything went well from the admission through to the recovery.

Thanks Olga (you know who you are).

Also I would like to thank Michele for her kind donation of some XS strips which has saved me about $100 in this instance (cos INR tests aren't free) and will in all likelihood give me another 6 months of tests too. {HatTip}

Thanks Michele, much appreciated!

Warning

This post contains what to the best of my knowledge is the current best practice. It is good advice if you don't have any other issues that are unknown to me. As long as you measure and follow the INR levels in here I believe that you'll be fine. Indeed if I were doing it again (which unless I die soon I will be) I'll probably just restart warfarin directly after the procedure and trust my knowledge of the INR fall and go off earlier.

BUT: Don't fuck with this stuff if you don't have a clue. You may get hurt, and that hurt may be permanent. Instead get a clue and start by having a read my other posts on INR management (use the INR tag in the tag list) keeping records and being on top of your own health. Reach out to me for assistance if you wish.


2 comments:

  1. Great material and data! Thanks for your effort and for sharing it.

    ReplyDelete
  2. Have a dental procedure coming up, and read your article, will suspend the warf. for 2 days before procedure, and resume same day in the evening; dentist told me it was ok to have 2.5, but i ll try to go close to 1.6 - 2; just to make it less bloody sort of speaking :) , thank for the data model,

    ReplyDelete