Question on probability of later problems due to coumadin

[ks1490]

Hi,

I would be interested in people's opinion on one line that stood out for me in the University of Cologne study that is mentioned in the recent "St. Jude or On-X" thread. The quote is: "Late adverse event rates were low, most notably thromboembolism (0.9%/pt-yr after AVR; 1.6%/pt-yr after MVR) and thrombosis." If the chances of a thromboembolic event is about 1% per year, doesn't that mean that after 20-30 years, your chance of some nasty event is now 20-30%? I'm wondering how people felt about that and how it may have figured into your decision to go tissue vs. mechanical. If the mortality rate for a re-op in the next 15 years is only a few percent, doesn't that outweigh the higher percentage risk of a coumadin-related event?

Thanks for your thoughts.

 

[Kate]

Hi!
I ultimately chose a biological valve and so don't know as much about coumadin as some others who will likely comment soon. However, this is my understanding - the risk of a bleeding event on Coumadin is roughly 1 - 2% percent per year. However, it is important to keep in mind two things. First, the vast majority of these bleeding events do not result in death or permanent disability- still upsetting, but can't really be compared to the fatality rates for repeat surgeries. Second (and this can be good or bad depending on how you look at it) these statistics include everyone including those who don't manage their Coumadin well and those who are very old (and are at much higher risk of bleeding events). So, if you are younger and are careful, your risk is probably much lower. Of course, the reality is that (hopefully) we will all be old one day, so this statistic will eventually not be in your favor.

These statistics were certainly one (although not the only) reason I went biological. Hope this is helpful! Kate

 

[geebee]

The 1%-2%/year chance of a bleeding event is not cumulative. 20 Years from now, the chance is still 1%-2%/year. I have been on coumadin for over 25 years and my chance is still the 1% or 2%.
Therefore, the risk is greater for problems during a reop than a per year chance of a bleeding event. However, in all faireness, we do live with the chance each year. For me it is a chance I am willing to take to avoid another surgery and I have been through reops and know how difficult they are.

 

[twinmaker]

I had a major bleeding event two years after my mechanical mitral valve replacement. An ovarian cyst ruptured, my PT was way off, and I started bleeding internally into my abdomen. I required emergency surgery and almost died during the surgery. During the first two years of being on Coumadin, it was hard to keep me in range and the testing was not what it is today. I feel somewhat responsible because I had bruises all over me, and I mean all over. I just didn't think it was anything to be concerned about. I was out of town at the time but went to a doctor to have my Protime tested and the bleeding episode happened that evening before I got the results back. I must say I also waited a good three to four hours before going to the ER...not a smart thing to do. I woke with what I thought were gas pains and walked the floor until I realized that things weren't getting better and my abdomen had started to blow up like a balloon. I looked like I was pregnant. Anyway, that was 23 years ago, and I haven't had any more bleeding events. All this to say that sometimes you just happen to be in that 1-2% and it happened to me while I was young and had only been on Coumadin for a relatively short time. I don't regret for a minute being on Coumadin and having a mechanical valve. It's lasted a good long time and I haven't had to have a third OHS for which I'm thankful. LINDA

 

[ALCapshaw2]

Some Doctors seem to believe that the bleeding risk is cumulative, and this may have been true in the early days of Coumadin management.

I've forgotten exactly when the INR testing protocol was developed (90's? maybe), but THAT has been a MAJOR breakthrough in Coumadin Management.

Prior to using INR as the standard, Prothrombin Time was measured directly with a blood draw. The PROBLEM was that the reagents used to evalutate the results had a HIGH LEVEL of VARIATION, making the measurements themselves questionable. The end result was that BLEEDING and STROKE incidents were MUCH more common 10 to 30 years ago than they are today.

The development of finger testing is another Major breakthrough that will continue to improve management as more and more people opt for Home Testing.

The HOPE for lower anticoagulation levels and possibly even aspirin therapy alone does make the On-X, ATS, and St. Jude Regent (all third Generation) valves appealing...

'AL Capshaw'

 

[Andyrdj]

Simple probability

Simple probability

Ok - to work things out from a probability of 2% per year, we turn it on its head and work out the probabilty of something not happening, and subtract that from 1. This gives us the probability of any non zero number of events - i.e. probability of (1 event or 2 events or 3 events....)

Let's do this for 20 years - and bear in mind that this is a "before I start" estimate.

So: prob of no events for 20 years =0.98 x 0.98...... (20 times), or 0.98 to the power of 20 = 0.6676,

i.e a 66.76% chance that you will be event free for 20 years

Therefore there is a 33.24% chance of at least one event (could be 2, could be 3, 4 or 20) within 20 years.

However, bear in mind also that each year is a fresh start. It's the same as rolling a dice 100 times and never getting a six - the probability of a six on the next throw (if indeed you don't have a duff dice!) is still 1/6.

So if you've been "lucky" as in Gina's case, (for 25 years the maths gives us 60.34% chance of freedom from any events) there's no mystic force which is about to make your luck end soon.

You are still playing the chance game on even odds with everyone really - although I would suggest a degree of personal stability and good INR management is likely to contribute to it in Gina's case.

I did the Maths again on other examples, and you have to go for 35 years before your odds drop to less than even for being free from events. Remember, that's a calculation you make for the coming 35 year period.

I hope no-one takes these results the wrong way - all I've done is plug in a little maths.

Try the same for 1% and you get a better result - 81.79% free after 20 years, 70.34% free after 35 years, and you take around 70 years for the odds to go below 50%.

Give me accurate figures, and I'll be interested to calculate this some more.

 

[Karlynn]

Been on the rat poison for over 14 years, and so far no bleeding even - knocking wood now.

The statistics for bleeding while on Coumadin are positive, in my opinion. But I think we spend a lot of time looking at statistics for new, or "scary" things when we don't really pay attention to statistics for the every day, mundane things. If we did, we'd probably be surprised by them. Most of us drive cars, but has anyone looked at what the statistics of a having a major car accident are, before we get in the car and drive to the grocery store, or take a 6 hour road trip? Or what about bike riding, or playing baseball? Or what are they for aspirin therapy? My point is, all of what we do can have statistics assigned to it. You can't get much lower than 1 - 2% per year. But there are people that end up on the wrong side of the statistics, those numbers do represent people.

I once heard that a woman over 40 who has never been married, statistically has a better chance of getting killed by a terrorist than getting married.

 

[Ross]

I once heard that a woman over 40 who has never been married, statistically has a better chance of getting killed by a terrorist than getting married.

Bring her to me!!!

Andy your not going to need to worry about anything if you keep trying to put statistical calculations behind everything. Your heads gonna explode long before heart surgery ever happens.

 

[PJmomrunner]

However, bear in mind also that each year is a fresh start.

Given the fresh start every year, wouldn't the cumulative probability of having a bleeding event be:

(.98 to the 20 + .98 to the 19 + .98 to the 18...+ .98) = 16.99653115%

 

[Karlynn]

Given the fresh start every year, wouldn't the cumulative probability of having a bleeding event be:

(.98 to the 20 + .98 to the 19 + .98 to the 18...+ .98) = 16.99653115%

????
I'm so confused!

If the chance is 1 - 2% per year and that each year is a "fresh start", doesn't that mean that there is no cumulative #? Doesn't this mean that the chance is 1 - 2% for the first year, and 1 - 2% for the 15th year?

 

[catwoman]

We should ask Al Lodwick.

 

[Andyrdj]

Int Maffs Brilliant?

Int Maffs Brilliant?

OK, the "fresh start" has everyone confused - though it is mathematically correct.

Assuming the 2% probablity, and using ^ to mean "to the power of"...

This analysis applies when each year's event is independant of the previous years - rather like subsequent throws of a dice.
Before you start an N year cycle, the prob of no events in N years is

P= (0.98^N)

for 20 years, P=0.98^20 = 0.6676 = 66.76%

But if you're four years in and still OK, you're asking "what is the probability of freedom from failure in that same 20 year period given that I'm 4 years in and no adverse event?"

This is the same as asking "what is the probability of the remaining 16 years of the period I first thought of being free?" Because remember that each year is being assumed independant of the previous year

In that case, the probability of being ok for the remainder of the period is 0.98^16 = 0.7238 = 72.38%

But, if you were to ask the probability of being ok for a new twenty year period starting from now, then it would be back to the 66.76% quoted earlier.

(See extra example at bottom for more info)

Remember, probability is about what you don't know yet, and it's crucial to be careful about what question you're asking. Sorry if I confused anyone - this sort of stuff is second nature to me, I've always loved maths. No head exploding events so far, Ross!

Now, I'd welcome Al Lodwick's input on this. I know my Maths is right, but I'm not sure if the assumptions behind it are correct.

It might be
- that the risk for each given year increases with time (getting older and frailer), or

- it might be that the risk can be minimised after a few years with experience, or

- It might be that when you've gone a long time and still been ok, then you're justified in regarding yourself as a particularly low risk patient. After all, if you throw a dice enough times and it never comes up six, you begin to quite reasonably suspect it might be rigged!


Another example:
If I have a non-rigged, guarnteed fair dice and I roll no sixes in 99 throws, what's the probability of a "non-six throw" on the 100th roll?

answer: It's 5/6, as before.

But if you asked before you started "what are the odds of 100 throws and no sixes", then it's (5/6)^100 (about 0.00000001). So, before you started, you'd think you had no chance, but if someone's throw 99 with no sixes and they say "want to bet on me making the 100th", then it's a 5/6 probability that they will - good odds in my book

DON'T confuse this sort of thing with Russian Roulette, where the odds change with each shot and get progressively worse!



To end with another piece of Statistics (which is actually true)
Did you know that there's a strong statistical correlation between people who don't understand statistics and people who believe in the paranormal?

 

[Karlynn]

I hate math, but as someone who views things from a practical standpoint my point is this: If Coumadin were a drug that began to have increasingly greater risks the longer a person was on it, then mechanical valves would not be used in young patients for the hope that it would be their last surgery. The reason mechanical valves are used is to avoid future surgery, due to the risks that additional surgeries may present. If the person receiving a mechanical was then subjected to greater increasing risks with the accumulation of time on the drug, then mechanicals would not be used in the way they are used today and all doctors would be recommending that the patient take their chances with additional surgeries.

 

[Andyrdj]

well...

well...

I take your point, Karlynn - though it's worth checking out that recent post which indicated that for the young population as a whole, survival stats were about the same for Mech vs Bio with Multiple re-ops.

Of course, I haven't performed an analysis on whatever similar adverse risk factors exist for a Bio patient. Would be interesting as the mathematics aren't independant year by year - there is indeed cumulative wear year by year, possibility of the occasional valve expiration, and then with scenarios of interrupted periods of increased risk due to it being a "bumpy ride" around the time of surgeries.

Quite fascinating! Any mathematically inclined members who understand the Maths term "Markov Chains" might like to have a little chat at some time, see if we can't model this.....

Anyway, regarding the "increasing risk" thing - I was sort of thinking in the realm of maybe 2% per year increasing to 3 or 4% for old age, as stroke risks generally do increase with age anyway - although I have no idea what the actual stats are.

Perhaps the risk stays steady, or perhaps it's a relatively small increase/decrease which doesn't tweak the probabilites much.

So when you say

If the person receiving a mechanical was then subjected to greater increasing risks with the accumulation of time on the drug, then mechanicals would not be used in the way they are used today

I might be pedantic and alter that to "Whatever increased risk with time exists has not historically been enough to offset the risk of multiple surgeries"

Not that I wish to increase your worry! Like I said, you shouldn't regard your past good fortune as being a finite pool of luck, so you have everything to look forward to in the future.

 

[Karlynn]

If the person receiving a mechanical was then subjected to greater increasing risks with the accumulation of time on the drug, then mechanicals would not be used in the way they are used today

I might be pedantic and alter that to "Whatever increased risk with time exists has not historically been enough to offset the risk of multiple surgeries"
.

Now I feel like I'm in a debate with my daughter. Don't take that as an insult! She's just very bright and will have a counter point to my point, ever time, even if she agrees with me! (She's going to law school next year - a wise use of her "talent" )

So my counter-counter point to your counter point ( ) would be that we have members here hedging their bets by hoping (assuming) future technology will allow repeat procedures to be less risky - and with that, there is little or no historal data to backup as-yet unknown or unperfected procedure. They just have a desire for medical technology to stay the course and continue to improve at a fast rate and hopefully achieve success in the area they wish to be successful, at the time they will need it. We have some members choosing On-X valves with the hopes that, at some point in time, a more "benign" drug will be available to prevent clotting with that valve. A promising area, but nothing with any historical data yet.

I would also point out that when you refer to "repeat surgeries" the historical data of repeat valve surgeries can't have much more historical data than Coumadin usage. After all, we are not yet to the 50th anniversary of valve replacement surgery itself. We'd need to take a look at when people started having multiple valve surgeries. Many people weren't surviving their first surgeries back then - ask RCB.

This is not to say that any of these choices are unwise, or wrong. But what I am saying is that if we are talking historical data, there's more backing up Coumadin, than some of the other areas that our members choose to place their future health in. You, yourself, are always posting new information here, in preparation for the choice that you will someday have to make. These new procedures won't have as much historical data as warfarin, but I'm sure that will not be your sole criteria.

My main point was that if most of us were to stop and pursue the statistical risks of the things we do regularly (drive cars, bicycle etc), we would find that many have higher risk factors than taking Coumadin....and to be fair, have higher risk factors than repeat surgeries.

(Point of Personal Info - I love a good debate - so Andy - thanks for the fun. If this were my daughter and I discussing something - my husband would have walked out of the room at the start! )

 

[Ross]

If this were my daughter and I discussing something - my husband would have walked out of the room at the start! )

That be a smart man!

 

[MikeHeim]

I hate math, but as someone who views things from a practical standpoint my point is this: If Coumadin were a drug that began to have increasingly greater risks the longer a person was on it, then mechanical valves would not be used in young patients for the hope that it would be their last surgery. The reason mechanical valves are used is to avoid future surgery, due to the risks that additional surgeries may present. If the person receiving a mechanical was then subjected to greater increasing risks with the accumulation of time on the drug, then mechanicals would not be used in the way they are used today and all doctors would be recommending that the patient take their chances with additional surgeries.

At the risk of being accused of redundancy, the math is annoying, but the concept is easy. The chances of being in a car accident are roughly 1/10000 every time you get in your car (please don't ask me to site the source). Assuming you get into your car twice per day, statistically you should get into an accident every 13.6 years. The same goes with Coumadin. The chance you will have a "bleeding incident" is 1/100 per year (from above). Therefore, if you are on Coumadin for 100 years, you should end up with one bleeding incident.

 

[Karlynn]

Mmmm, don't think that quite works out that way. Technically that's the way you could put it, but it's kind of like saying those who play the lottery every time have a better chance of winning than those who are playing for the first time.

Pop. of US ~299,000,000, # of drivers in US ~196,000,000
# of auto accidents for 2004 - 6.3 million
3.1% having been involved in an auto accident in a year.
(This of course doesn't take into consideration those people involved in more than one accident.) - but you can see why I stated that Coumadin is less of a risk, statistically, for me than driving my car. Not to mention that I live in the Chicago area, so I think my statistics are even more wide!!!!

http://www-nrd.nhtsa.dot.gov/pdf/nrd-30/NCSA/TSFAnn/TSF2004.pdf
http://www.car-accidents.net/car-accident-stats.html

(And just so you all know I'm not a nut case, I have this site because I was helping my neighbor do some research on teen drivers in the US. Oh, and incidentally - slightly more MALES involved in accidents than females. )

I should also mention, that with this talk of statistics, we aren't discussing other relevant factors such as aging, or additional drug usage, or the like which would change your personal statistical #.

 

[Ross]

(And just so you all know I'm not a nut case, I have this site because I was helping my neighbor do some research on teen drivers in the US. Oh, and incidentally - slightly more MALES involved in accidents than females. )

I can tell you why too. Were busy checking out you babes in bikinis and we become distracted.

 

[JohnnyV_46]

I for one appreciate Andys dedication to seeing both sides. He doesn't even have one yet but I know he's made up his mind. People have to know that their is life with and without Coumadin. Why do they make a tissue valve? I don't know the answer to that but there has to be a legitimate reason they spent the millions of dollars in research. I'm glad I chose the Tissue valve and I hope I still am 15 years from now. I'm also glad that everyone else is doing so well on Coumadin.