This blog has too many mini-blogs on too many subjects. I've separated them off into (for now) three separate blogs:
1. Insulin Pumps - Bubble Trouble
http://insulinpumps-bubbletrouble.blogspot.com
2. Regressions - Calcium Scoring and Heart Attacks
http://regressions-calciumscoring.blogspot.com
3. Regressions - US Climate Change?
http://regressionsusclimatechange.blogspot.com
Go to one of these for the latest on the subject.
Tuesday, September 14, 2010
Friday, May 28, 2010
More perspectives on insulin pump bubble problems.
1. From what I've read on forums about bubbles in insulin pump tubing, and from what I've heard from other pump-users with bubble troubles, physicians almost always say, "This is new to me. I haven't seen this problem before." A friend who's a relatively young physician said that physicians are almost obligated to say this. If they acknowledge they are aware there's a problem, then it can held to be their responsibility if something goes wrong and one of their patients is admitted to the hospital because of sky-high BG due to bubbles.
Also, it would be a LOT of work for a physician to go up against Medtronic, and few physicians would want to use their time fighting against large corporations. They want to follow their protocols; and they are less likely to be sued if they follow the conventional wisdom. And if nothing in the FDA approval literature says to watch out for bubbles, then bubbles don't exist.
2. From Medtronic's perspective, they're certainly going to deny there's a problem and blame bubbles on "patient error." Since we live in southern California, we offered to go to the Medtronic facility in the San Fernando Valley and discuss the situation with them. They said that would be fine with them, as long as we signed a nondisclosure agreement saying we would never ever divulge ANYTHING they told us, with the penalty for disclosure being major liability in a lawsuit. Needless to say, we haven't taken them up on their offer.
In a way, a bubble in the tubing is a perfect poison. There's no trace of it left after it goes into the body; and if someone ends up in the emergency room with high BG readings it's impossible to prove a bubble was the problem. The only way to prove bubbles cause problems is for a patient to deliberately allow a long (6" or longer) bubble to go through the tubing and into their body, take a video as the bubble moves, and then go to the emergency room with a BG higher than 500. In the emergency room, the patient, of course, would be likely to be referred to the psych ward for doing that to themselves.
"What about the FDA?" you ask. There is a system for reporting problems with medical devices and drugs. Here's the link:
https://www.accessdata.fda.gov/scripts/medwatch/medwatch-online.htm
Lots of luck getting a meaningful response. The FDA is unlikely to get involved unless there's been real harm to a patient and, again, it's almost impossible to prove the harm was due to a bubble.
For now, all I know that people with bubble problems can do is to check their tubing 2-3 times a day, and prime out the bubbles before they reach the body. See my other postings for lots of detail.
Also, it would be a LOT of work for a physician to go up against Medtronic, and few physicians would want to use their time fighting against large corporations. They want to follow their protocols; and they are less likely to be sued if they follow the conventional wisdom. And if nothing in the FDA approval literature says to watch out for bubbles, then bubbles don't exist.
2. From Medtronic's perspective, they're certainly going to deny there's a problem and blame bubbles on "patient error." Since we live in southern California, we offered to go to the Medtronic facility in the San Fernando Valley and discuss the situation with them. They said that would be fine with them, as long as we signed a nondisclosure agreement saying we would never ever divulge ANYTHING they told us, with the penalty for disclosure being major liability in a lawsuit. Needless to say, we haven't taken them up on their offer.
In a way, a bubble in the tubing is a perfect poison. There's no trace of it left after it goes into the body; and if someone ends up in the emergency room with high BG readings it's impossible to prove a bubble was the problem. The only way to prove bubbles cause problems is for a patient to deliberately allow a long (6" or longer) bubble to go through the tubing and into their body, take a video as the bubble moves, and then go to the emergency room with a BG higher than 500. In the emergency room, the patient, of course, would be likely to be referred to the psych ward for doing that to themselves.
"What about the FDA?" you ask. There is a system for reporting problems with medical devices and drugs. Here's the link:
https://www.accessdata.fda.gov/scripts/medwatch/medwatch-online.htm
Lots of luck getting a meaningful response. The FDA is unlikely to get involved unless there's been real harm to a patient and, again, it's almost impossible to prove the harm was due to a bubble.
For now, all I know that people with bubble problems can do is to check their tubing 2-3 times a day, and prime out the bubbles before they reach the body. See my other postings for lots of detail.
Saturday, March 14, 2009
Do insulin pump bubbles cause medical emergencies?
We recently got an e-mail from another person who's had problems with insulin pump bubbles.
If bubbles in the pump cause a few strange high blood glucose readings (say in the 200s or low 300s), that's probably not a big deal - almost every insulin-dependent diabetic has these from time to time when they mis-guess the number of carbs in a meal. But a number of 420 and a trip to the emergency room is a big deal, both health-wise and bottom-line-wise for patients and insurance companies, and - we can hope - for federal regulators who are supposed to watch out for problems with medical devices.
Here is what the person said (personal identifying information has been removed):
--------------------------------------------
Hello,
I just found your information about air bubbles in the Medtronic Insulin pump.
It was as you had read my mind!!
I was put on the pump in December ....
My problems were first found Jan.... I noticed a long air bubble in my tubing as I was about to bolus for a meal.
No telling how many more had gone through without my knowledge!
I even had to go to the E.R. one night with a 420 reading!
While there...I found a 2" air gap in my tubing line!
Since that time...it's been a pain, checking...purging....wasting insulin.
I called and talked to the help line also... actually talked to Christine...and yes she's a jewel!! Very nice helpful person.
Switched infusions, reservoirs with Medtronics...changed insulins...finally got a new pump. Still have bubbles...none as large as before...but yes, I do still have to check constantly to see if there is a air gap in the line. I actually have had 5" bubbles!!
Talk about a big problem if I hadn't checked!!
I'm sending your blog and your posting in the forum to the Diabetes Clinic that I go to...Just to let them know it's NOT just me!
One thing that was offered to me was to fill the reservoir ahead of time...let it sit (hoping the fizzie bubbles will all come to the top) before changing my sets. I'm like you...I think that bubbles come in from the o-ring. When I tap the bottle to remove bubbles...there seem to be more coming from the o-ring, and that doesn't make sense.
I hope to hear back from you, hopefully with information on this problem!!
If bubbles in the pump cause a few strange high blood glucose readings (say in the 200s or low 300s), that's probably not a big deal - almost every insulin-dependent diabetic has these from time to time when they mis-guess the number of carbs in a meal. But a number of 420 and a trip to the emergency room is a big deal, both health-wise and bottom-line-wise for patients and insurance companies, and - we can hope - for federal regulators who are supposed to watch out for problems with medical devices.
Here is what the person said (personal identifying information has been removed):
--------------------------------------------
Hello,
I just found your information about air bubbles in the Medtronic Insulin pump.
It was as you had read my mind!!
I was put on the pump in December ....
My problems were first found Jan.... I noticed a long air bubble in my tubing as I was about to bolus for a meal.
No telling how many more had gone through without my knowledge!
I even had to go to the E.R. one night with a 420 reading!
While there...I found a 2" air gap in my tubing line!
Since that time...it's been a pain, checking...purging....wasting insulin.
I called and talked to the help line also... actually talked to Christine...and yes she's a jewel!! Very nice helpful person.
Switched infusions, reservoirs with Medtronics...changed insulins...finally got a new pump. Still have bubbles...none as large as before...but yes, I do still have to check constantly to see if there is a air gap in the line. I actually have had 5" bubbles!!
Talk about a big problem if I hadn't checked!!
I'm sending your blog and your posting in the forum to the Diabetes Clinic that I go to...Just to let them know it's NOT just me!
One thing that was offered to me was to fill the reservoir ahead of time...let it sit (hoping the fizzie bubbles will all come to the top) before changing my sets. I'm like you...I think that bubbles come in from the o-ring. When I tap the bottle to remove bubbles...there seem to be more coming from the o-ring, and that doesn't make sense.
I hope to hear back from you, hopefully with information on this problem!!
Friday, March 13, 2009
Insulin Pump Bubbles continued
OOPS! I accidentally put the material on insulin pump bubble problems into the section on regressions. Sorry about that. But more on the bubble subject anyway.
Since we sent the letter to Medtronic about the bubble problems, we got a letter back from Shirajul Karim, the "Sr. Dr. of Quality" dated January 16, 2009. Here's the important sentence in his letter:
---------------------------------------
"While our experience with our products is not consistent with the observations and testing described in your letter, we will carefully and fully investigate your claims."
---------------------------------------
And here's the letter we sent back to him. He hasn't replied, nor is there any evidence they made any attempt to "carefully and fully investigate" our findings. Nada.
----------------------------------------
Shirajul Karim, Senior Director of Quality
Medtronic Minimed
18000 Devonshire Street
Northridge, CA 91325
Dear Mr. Karim:
Thank you for your letter of January 16 acknowledging our letter of December 24. Let us first note that, despite the bubble problems we (and others) have reported, we are pleased with our experience with the pump. The fine control over insulin delivery rates it offers is a vast improvement over that available with injections.
It is surprising that ongoing problems with bubbles and air pockets have not previously been brought to the attention of Medtronic Minimed Quality Control. The issue certainly has been addressed to your HelpLine personnel by other people in addition to us. Here’s a link that illustrates the point:
http://tudiabetes.com/group/minimedparadigmusers/forum/topics/583967:Topic:147765
And here’s another link where “Dave” gives a sure-fire method to avoid bubbles – we have followed all of his suggestions, with no difference in bubble formation:
http://tudiabetes.com/forum/topics/583967:Topic:256260
As our December 24 letter shows, the bubbles develop only after and only if the insertion set is in FFFF’s body, take several days to grow, and their growth appears to involve leakage of air at the o-ring-reservoir interface (at least, that’s where the large bubbles are found before they break loose and enter the tubing).
It obviously would be simple for Medtronic Minimed to survey 100 or so patients who use the pump to see if they have bubbles in their reservoirs (e.g., by asking callers to the HelpLine to look at their pump reservoirs). However, the failure of Medtronic Minimed to have a substantive response to our letter in more than a month, the content of your letter, and the protocols used by the HelpLine personnel indicate that Medtronic Minimed has already made up its mind that the only problem is “patient error.” Presumably the solution will be for an outside agency (e.g., the FDA) or health plan (e.g., Kaiser Permanente) to investigate bubble formation in reservoirs and the possible risks to their patients. As we noted previously, the largest bubble we’ve “primed out” was equivalent to more than 2 Units of insulin that would not have been delivered, enough to increase my BG reading by about 150 mg/dL.
In the meantime, we would be interested to know if Medtronic Minimed comes up with any ideas for how the bubbles develop. As you may know, our first pump has been tested and found to be working flawlessly – as we said in our letter we expected to be the case. Please let us know if we can give you any additional information.
Since we sent the letter to Medtronic about the bubble problems, we got a letter back from Shirajul Karim, the "Sr. Dr. of Quality" dated January 16, 2009. Here's the important sentence in his letter:
---------------------------------------
"While our experience with our products is not consistent with the observations and testing described in your letter, we will carefully and fully investigate your claims."
---------------------------------------
And here's the letter we sent back to him. He hasn't replied, nor is there any evidence they made any attempt to "carefully and fully investigate" our findings. Nada.
----------------------------------------
Shirajul Karim, Senior Director of Quality
Medtronic Minimed
18000 Devonshire Street
Northridge, CA 91325
Dear Mr. Karim:
Thank you for your letter of January 16 acknowledging our letter of December 24. Let us first note that, despite the bubble problems we (and others) have reported, we are pleased with our experience with the pump. The fine control over insulin delivery rates it offers is a vast improvement over that available with injections.
It is surprising that ongoing problems with bubbles and air pockets have not previously been brought to the attention of Medtronic Minimed Quality Control. The issue certainly has been addressed to your HelpLine personnel by other people in addition to us. Here’s a link that illustrates the point:
http://tudiabetes.com/group/minimedparadigmusers/forum/topics/583967:Topic:147765
And here’s another link where “Dave” gives a sure-fire method to avoid bubbles – we have followed all of his suggestions, with no difference in bubble formation:
http://tudiabetes.com/forum/topics/583967:Topic:256260
As our December 24 letter shows, the bubbles develop only after and only if the insertion set is in FFFF’s body, take several days to grow, and their growth appears to involve leakage of air at the o-ring-reservoir interface (at least, that’s where the large bubbles are found before they break loose and enter the tubing).
It obviously would be simple for Medtronic Minimed to survey 100 or so patients who use the pump to see if they have bubbles in their reservoirs (e.g., by asking callers to the HelpLine to look at their pump reservoirs). However, the failure of Medtronic Minimed to have a substantive response to our letter in more than a month, the content of your letter, and the protocols used by the HelpLine personnel indicate that Medtronic Minimed has already made up its mind that the only problem is “patient error.” Presumably the solution will be for an outside agency (e.g., the FDA) or health plan (e.g., Kaiser Permanente) to investigate bubble formation in reservoirs and the possible risks to their patients. As we noted previously, the largest bubble we’ve “primed out” was equivalent to more than 2 Units of insulin that would not have been delivered, enough to increase my BG reading by about 150 mg/dL.
In the meantime, we would be interested to know if Medtronic Minimed comes up with any ideas for how the bubbles develop. As you may know, our first pump has been tested and found to be working flawlessly – as we said in our letter we expected to be the case. Please let us know if we can give you any additional information.
Tuesday, January 13, 2009
Insulin Pump Bubbles
Note: Our names, address, etc. have been replaced by capital letters to preserve privacy. Feel free to send us an e-mail at wands65mail-blogspot@yahoo.com, and we’ll be glad to reply.
----------------------------------------------
XXX YYYY Street
San Diego, CA 92ZZZ
Phone: AAA-BBB-CCCC
E-mail:
wands65mail-QQQ@yahoo.com
December 24, 2008
Medtronic:
Quality Control, Customer Support Center
Christopher J. O'Connell, Senior Vice President and President, Diabetes
The purpose of this letter is to ensure that you are aware of what we believe to be a significant safety issue with the Paradigm pump system involving formation of air bubbles in the pump reservoir and propagation of such bubbles through the tubing into patients' bodies. Such bubbles have the potential of resulting in serious or even catastrophic health effects on patients. This letter is - unfortunately - quite long, because we have tried to document it as completely as possible with the resources available to us, and because of the potential seriousness of the problem.
Contents of this letter:
1. Disclaimer/qualification
2. Executive summary
3. Our background
4. Our observations
5. Our interactions with the Medtronic HelpLine
6. Experiences of others
7. Our experiments
8. Chemistry 1 exercises
9. Our conclusions
10. Draft URGENT WARNING statement
1. Disclaimer/qualification
The information and conclusions in this letter represent our good-faith attempt to understand and describe the problems with our Paradigm 522 pump. We do not have access to the laboratory equipment that would be required to determine the precise origin of the problem. Every statement of fact in this letter should be understood to include an initial clause, "In our experience/judgment,...."
2. Executive summary
Air bubbles can develop at the interface between the o-ring and the reservoir body of the Paradigm pump (we'll term this the O-R interface). These bubbles can develop no matter how carefully the patient removes air bubbles during the infusion set insertion procedure; they appear to develop because of leakage of air into the reservoir around the o-ring as the pump operates. These bubbles can grow to become of significant size (>2 mm diameter), break loose, migrate to the top of the pump, into the tubing, and into the body of the patient, causing a significant (>1 Unit) loss of insulin flow into the patient's body. In turn, this loss can cause high blood sugar readings (>300 mg/dL); such spurious readings could also cause patients and their physicians to increase insulin doses with the possibility of resulting severe hypoglycemia and its severe consequences.
On or - preferably - before January 12, 2009, we ask Medtronic:
a. To convince us that our analysis of the situation is faulty. We regard ourselves as reasonable people, and would be delighted to be shown that something we (and other Paradigm pump users - see Section 6 below) are doing is wrong.
or
b. To resolve the problem and take the steps required to ensure that air bubbles resulting from action of the pump do not pose a significant danger to patients using the pump.
or
c. To contact every user of a Paradigm pump and every physician who prescribes Paradigm pumps to inform them of the possibility of development of such bubbles, the danger these bubbles present, ways of checking for bubbles in the reservoir and, and ways of checking for and removing air pockets in the infusion tubing. A draft of such an URGENT WARNING statement is given as Section 10 of this letter.
or
d. To take other appropriate action to address this problem.
3. Our Background
DDDD's bachelor's degree is in chemistry from the California Institute of Technology; his Ph.D. is also in chemistry from Stanford University; and he did postdoctoral work in biochemical neurology at the Columbia College of Physicians and Surgeons. He was a chemistry professor for 25 years at the University of California, EEEE, and is now an emeritus professor of chemistry. He has worked with and taught students about scientific instruments throughout his professional career.
FFFF was diagnosed with Type 1 diabetes in 2007 at the age of 63(!!). She has been using a Paradigm 522 pump since June, 2008. FFFF eats a very controlled diet, with ~90-105 g carbs per day. She uses ~20 Units of Novolog insulin per day.
4. Our observations
Approximately the first of November, 2008, FFFF began to have increased problems with high (sometimes >300 mg/dL) and erratic blood glucose (BG) readings. We eventually learned that the situation could be corrected by a new infusion set insertion, and thought perhaps the cannula had been inserted into a muscle (??) from which the insulin was not as well absorbed.
However, one time we noticed that there were large (>2 mm diameter) bubbles in the old pump reservoir when the infusion set was changed. We reasoned that if such a bubble worked its way into the tubing and into FFFF's body that a significant shortfall of insulin delivery could result.
We are scrupulous in eliminating bubbles during the infusion set insertion. We inspect both the reservoir and the tubing using a magnifying lens and (currently) a flashlight. We see absolutely no bubbles when the insertion set process is complete - the bubbles develop afterward, and the bubbles of significant size develop at the O-R interface.
Typically, a bubble will first be noticed by us when it's ~0.5-1 mm in diameter. Over the next day or so, additional bubbles develop, some bubbles stay the same size, and others become larger while they remain at the O-R interface, eventually becoming as large as 2-3 mm in diameter. When such a bubble works its way into the tubing, we currently "prime the bubble out" (see Section 10).
When we first began notice bubbles in the reservoir at the O-R interface, we saw only a few - or even no - bubbles at each infusion set change. However, as time has passed, the frequency of bubble formation has increased, so that we now see 5-15 bubbles at the O-R interface every time the infusion set is changed.
5. Our Interactions with Medtronic
(Dates are approximate, and some conversations are summarized or their substance mixed together - Medtronic has recordings of the conversations and can transcribe them, but we doubt that is important.)
First, a compliment: we talked mostly with Christine on the HelpLine. She was wonderful to work with: interested, empathetic, and obviously trying to do the best she could for us with the information available to her. Can you clone her to handle all HelpLine calls?
We called the Medtronic HelpLine, and the following was recommended:
a. Keep the insulin vial that is currently being used at room temperature rather than in the refrigerator. Note: as best we can tell, this instruction is NOT in the User's Guide for the pump.
b. Try de-gassing the insulin vial by repeatedly withdrawing air and creating a partial vacuum before filling the reservoir. We tried this procedure, found that it did not prevent subsequent formation of bubbles at the O-R interface, and it actually made it difficult to fill the reservoir because of the vacuum. In fact, this procedure tends to cause air bubbles to come into the reservoir around the o-ring. Aha! Air can leak around the o-rings. Note: as best we can tell, this instruction is NOT in the User's Guide for the pump.
c. Move the reservoir plunger up and down several times before filling the reservoir with insulin to better distribute o-ring lubricant. Note: as best we can tell, this instruction is NOT in the User's Guide for the pump.
d. Try different lot numbers for the reservoirs, and different batches of insulin. So far we're used at least 3 different lots numbers for reservoirs and 2 different insulin batches.
Following any or all of these recommendations had no effect. Air bubbles continued to develop at the O-R interface, to grow, and sometimes to break loose and enter the tubing. The longest air pocket we've seen in the tubing so far was about 6" (!!!) long, corresponding to a shortfall of more than 2 Units of insulin that would not have been delivered.
We were told that Medtronic believes that development of air bubbles in the reservoir is due to faulty patient technique, not to design or manufacturing flaws. We went over our insertion technique with the HelpLine, and they were unable to identify any deficiencies.
After several other conversations, a Medtronic contract nurse was assigned to come observe our technique on Tuesday, December 16. Also, Christine told us that replacing the pump was a possibility if the problems became more severe.
On Saturday, December 13, we did a regular infusion set replacement. We had seen one large bubble (~2 mm diameter) at the O-R interface that had remained unchanged for several days. However, one can see only a fraction of the circumference of the reservoir through the window in the pump. When we removed the reservoir from the pump, we observed 8 bubbles at the O-R interface, ranging in diameter between approximately 1 mm and 2.5 mm. Any of these could have broken loose and entered the tubing without us being aware of that happening unless we inspected the tubing.
We called the Medtronic HelpLine, talked to Shari (Christine was not working on that day), and explained our concerns. Shari said that, since a nurse was scheduled to come on the following Tuesday, there was no possibility of replacing the pump before that time, and that her decision to that effect reflected Medtronic policy. She further said that Medtronic believes all cases they have seen of bubbles developing in the reservoir are due to operator/patient error. Period.
The nurse on contract to Medtronic came on Tuesday, December 16 and observed our technique for insertion of an infusion set. Another compliment: the nurse (Sarah Johnson) was helpful, empathetic, and clearly cared about the patients she deals with. Ms. Johnson observed our technique for infusion set insertion, and found no defects in our procedure.
6. Experiences of others
We thought that we had a unique problem, or that something was wrong with our technique until we did a simple internet search and found that others have had similar problems. See, for example,
http://www.diabetesforums.com/forum/pumping-insulin/15923-air-bubbles-reservoir-tubing.html
Here's a posting:
"The pump is a minimed 722. The past two sets, when I was under 40 units remaining, I discovered air bubbles both in the reservoir and in the tube. When I filled the reservoir I was careful both times to make sure no bubbles were present so I don't know how they got introduced into the system. These bubbles caused my sugar to go >300 with corrections not working since I was not really getting the correct dosage (I started a new set immediately when I noticed this)."
(Searching in www.diabetesforum.com for, say, "pump bubbles" gives many discussions about problems with bubbles in the reservoir and tubing.)
http://www.diabetesdaily.com/forum/insulin-pumps/20871-air-bubble
Here's a posting:
"I got up this morning, checked my BG and it read 160. It should have been lower, but oh well. I programmed my pump for a 40carb breakfast and ate. After 1.5 hours I wanted to head to the gym so I checked my BG again—300! %$#$%! I started a correction bolus and while it was delivering I noticed a slight color change in a section of my infusion set tube. I looked closed and saw that it was moving! Turns out, it was an air bubble about 2 inches long. Have any of you had this problem? I have been having trouble with my morning boluses and now I’m beginning to wonder if I’m getting air bubbles in my line while I sleep."
http://tudiabetes.com/forum/topic/show?id=583967%3ATopic%3A109266
Here's a posting:
"I recently started using the Minimed Paradigm pump. I love using the pump it has made everything so much more convenient but I am having problems with bubbles. I have made sure the insulin is room temperature and when I fill the cartridge I make sure there are no bubbles. However, a couple of times I have had soaring blood sugars and after trying to work out what else it could be have changed my cartridge and found it full of bubbles. Has anyone else had this problem and what can I do?thanks"
There's even a youtube video on getting rid of bubbles!
http://www.youtube.com/watch?v=OwbelFCNGz4
The general rule of the internet is that for every such posting there are many, many more people with a similar problem.
In our opinion, it is difficult to believe that Medtronic is not aware of such forum discussions and problems people have with bubbles in the reservoir. One presumes that Medtronic monitors postings to such forums. Medtronic's policy to attribute air bubbles exclusively to operator/patient error ("Blame the patient") is, at best, irresponsible.
7. Our experiments
a. We filled normal disposable syringes with insulin, both with room-temperature insulin and with insulin taken directly from the refrigerator, inserted the needles into a rubber stopper to prevent leakage, and observed the syringes for several days. Did any bubbles develop? Of course not - neither in the room-temperature insulin nor in the straight-from-the-frig insulin. Medtronic's recommendation to begin with the insulin at room temperature to prevent formation of large bubbles appears to be irrelevant. Note that normal syringe plungers have a rubber-like end on the piston, not an o-ring seal. More on that later.
b. We filled a reservoir with insulin straight from the frig, did a "prime" with the reservoir plunger (that is, we didn't put the reservoir in the pump), clamped off the tubing, and cut off the end of the tubing to remove the insertion needle. We then allowed this reservoir to sit around for several days, and put it in a shirt pocket for several hours several times to allow it to warm to close to body temperature. Here is what we found.
There were several small, insignificant (see Chemistry 1 exercise in Section 8 below) bubbles (0.2 to 0.6 mm in diameter) at apparently-random places inside the reservoir. NO bubbles developed at the O-R interface. There might have been even fewer and smaller bubbles if we had used room-temperature insulin. It doesn't matter; the bubbles we did observe were insignificant.
This observation also answered one of our technical questions about whether bubbles might be more likely to develop spontaneously at the O-R interface than at random places on the plastic reservoir body because of gas nucleation dynamics. They didn't.
This observation also confirms that our technique for filling the reservoir without bubbles works. The bubbles aren't introduced by operator/patient error - Medtronic can't "Blame the patient." The bubbles at the O-R interface occur after the bubble-free reservoir is in the pump, the pump is operating, and the insulin is going into the patient.
After additional discussion with Christine, we were sent a new pump. This gave us the opportunity for additional experiments.
c. The problem is NOT with the pump! After FFFF began using the new pump, we filled the reservoir for the old pump, and did everything except the insertion of the Quick-set cannula into her body. We allowed the pump to run according to its usual settings, again, with the insulin flowing into the air instead of into her body. NO significant bubbles formed no matter what we did - e.g., fixed primes with the end of the tubing open to the air or clamped off. We believe that no problems will be found with the pump when it is tested.
d. We used the same infusion set lot numbers and the same insulin vial that was used for experiment c. above with the new pump, with the cannula inserted into her body. Bubbles formed in the reservoir for the new pump, just as with the old pump, with the bubbles of significant size forming at the O-R interface.
e. We thought that perhaps the pump and reservoir could get warm enough under the bedcovers at night to cause bubbles to form. The temperature of the surface of the pump can be as high as 95°F (it's under the bedcovers and close to the body). Parenthetically, we have no idea if a temperature of 95°F causes insulin to degrade too rapidly; we haven't observed any problems in the typical three days a reservoir is used. We slept with the pump that was being replaced (again, pumping insulin into the air instead of into her body), and there was no significant formation of bubbles.
To summarize these three experiments:
i. The bubbles apparently do not form because of a malfunction of the pump. Bubbles of significant size do not form unless the pump is operating to pump insulin into her body. Apparently, bubbles form because of an interaction involving some combination of the pump, the reservoir, the Quick-set cannula system, and the patient's body.
ii. Our technique for filling the reservoir and removing bubbles appears to be flawless, since bubbles do not form in a reservoir that isn't pumping insulin into her body but otherwise is operating normally.
iii. It's hard to imagine that anything we do with the Quick-set insertion could affect bubble formation.
iv. We have no idea what mechanism is responsible for the bubble formation.
8. Chemistry 1 exercises
a. 1 Unit of insulin corresponds to 0.01 mL of insulin. Exercise: Show that the number of Units of insulin in a bubble of diameter X mm is about 0.05 X3.
So if the diameter of a bubble is less than 1 mm, the number of units in a bubble is less than 0.05 - insignificant.
On the other hand, since the number of Units goes as the cube of the bubble diameter, a bubble 2 mm in diameter corresponds to 0.4 Units - if such a bubble goes into a patient, that represents a significant insulin defect. And a bubble 3 mm in diameter corresponds to about 1.35 Units of insulin. An insulin dose that's missing that much insulin because air rather than insulin gets into the patient could cause large changes in BG readings.
b. The diameter of the tubing corresponds to 1 Unit of insulin per about 2½-3" of tubing. Exercise: Show that a 1/4" air pocket in the tubing is insignificant, but an air pocket 1/2" or more is a concern.
9. Our conclusions
A joke of sorts: Somebody suggested that perhaps the people who designed o-rings for NASA in the 1980s are now designing o-rings for Medtronic.
We have discussed this problem with two friends. One is an instrumentation expert who repairs and rebuilds scientific instruments. His reaction: "It's most likely a problem with a mismatch between the o-ring and the reservoir. The o-ring seal simply isn't tight enough."
The other is a NASA scientist, who noted that o-rings are normally used for permanent seals between two stationary surfaces, not to form a seal between moving parts. Using o-rings to form a seal between moving surfaces is a dicey proposition. Again, note that the interface between the piston and the reservoir body on a normal disposable syringe is a rubberized end on the piston, not o-rings; and there's no air leakage into a syringe as a syringe piston moves forward.
We continue to be mystified about the development of bubbles at the O-R interface, especially about the following two points:
a. Why isn't bubble formation reported by many more patients? Perhaps the reason bubbles are not, according to Medtronic, regularly reported as a problem is because many patients use far more insulin than FFFF (she uses ~20 Units per day). A loss of a Unit or so every once in a while because of an air pocket in the tubing wouldn't be as significant to someone who uses much more insulin. We don't know the answer to this question.
b. What's the reason for bubbles to develop at the O-R interface and grow in size? Since the piston presumably moves only forward and is essentially under "forward" pressure, it seems unreasonable for bubbles to develop in the reservoir that's under positive pressure. Development of bubbles would, of course, be expected if the piston sometimes moved backward rather than always forward.
We simply do not know the reason for the origin and growth over time of the bubbles at the O-R interface. Perhaps the design of the system is faulty. Perhaps temperature cycling or pressure cycling of the o-rings and piston is causing the piston to move back and forth enough to cause air leakage. Perhaps pressure variations in the body (systolic/diastolic??) are transmitted to the o-ring seal system? We can only make wild guesses.
An obvious solution is to abandon the o-ring-based plunger and move to a plunger with a rubberized end, as in a regular disposable syringe. Presumably Medtronic considered such a possibility when the pump was designed, but a tighter-fitting plunger might make it more difficult to sense occlusions.
10. Draft URGENT WARNING statement
If Medtronic cannot find an error in our procedure for inserting an infusion set or cannot fix the problem of air bubbles, we believe that a statement along the lines of the draft below should be sent to every patient who uses a Paradigm pump and to every physician that prescribes them. We believe that bubbles that develop in the pump reservoir pose a significant danger to patients.
Notice: The URGENT WARNING statement below is Copyright 2008 by DDDD and FFFF HHHH.
Any private individual may reproduce or transmit this statement to any person of his/her choice.
However, any reproduction of any portion of this statement or any of the recommendations in this statement in any form (plagiarized, adapted, modified, etc.) by any corporation or company, or any representative of any corporation or company, will be regarded as a severe infringement of copyright laws and will subject the violator to criminal penalties and civil damages, unless they receive express permission from DDDD and FFFF HHHH.
The purpose of this notification is to prevent a company (e.g., Medtronic) from issuing a weasel-word warning in which they essentially continue to "Blame the patient" for formation of air bubbles in the pump reservoir. We will, of course, be more than happy to give permission to Medtronic or other companies to issue warning statements that show where responsibility lies as well as give vital information to patients and their doctors.
Caveat: The statement which follows is a draft statement. We are not physicians; nor are we experts on insulin pump design and problems. This statement is based on our experience on how to minimize the danger caused by bubbles which develop in insulin pump reservoirs. Patients should consult their physicians about their treatment regimen and how to operate the insulin pump.
--------------------------------------------------------------------------------------------------------------------------------------
URGENT WARNING
It has been found that air bubbles of significant size can develop in insulin pump reservoirs, mainly at the interface between the o-ring and the body of the reservoir. Such bubbles can develop no matter how scrupulously the patient removes bubbles during the process for insertion of the infusion set.
These bubbles can result in air pockets moving into the infusion-set tubing and into the patient's body - air rather than insulin goes into the patient's body. This replacement of insulin by air can cause high and even dangerous blood glucose readings (>300 mg/dL).
At this time, no one knows the reason for the origin of these bubbles. While the problem is being studied, the following procedure is recommended to avoid air pockets moving into the patient's body.
1. At least four times a day (e.g., before each meal and at bedtime), inspect
(a) the pump reservoir for bubbles through its window on one side of the pump and
(b) the tubing for air pockets in the tubing.
If a bubble in the reservoir disappears, it is likely to show up later as an air pocket in the tubing.
Inspection is assisted if a magnifying glass is used, and if a flashlight is also used (holding the flashlight under the tubing helps to identify air pockets in the tubing). Roughly speaking, when insulin is in a section of the tubing, the entire cross-section of the tubing looks similar to frosted glass. When there is an air pocket, one can see what looks like an empty space in the middle of the cross-section of the tubing. Look carefully; air pockets in the tubing are difficult to see.
2. If a significant air pocket (length >1/4" or >6 mm) is seen, disconnect the "Quick-set" (or equivalent) from your body. See your health-care provider if you do not know how to do this.
3. "Prime the bubble out." That is, deliver a "Fixed prime" sufficient to cause the air pocket to move completely through and out of the tubing. One Unit of insulin corresponds to 2½" of tubing. For example, if the air pocket is 6" from the end of the tubing, you could deliver a "Fixed prime" of 5 Units and be reasonably sure the air pocket would be removed from the tubing. Consult with your health-care provider before the first time you do this.
VERY, VERY IMPORTANT: After you have delivered the "Fixed prime" as described above, deliver another "Fixed prime" that corresponds to the "Fixed prime" you normally deliver when you insert the infusion set (typically 0.3 Unit). The reason for this additional "Fixed prime" is to make sure you have returned the pump to its original settings. If you do not, the next time you do a "Fixed prime," you could deliver, say, 5 Units, which could be disastrous.
4. If you see such air bubbles at the interface between the o-ring and the reservoir body or air pockets in the tubing, call 1-800-XXX-XXXX [phone number to be added] to report the problem.
Also important: An air bubble in the reservoir less than about 1/16" (or 1 mm) in diameter is not significant, nor is an air pocket in the tubing shorter than about 1/4" (or 6 mm). Small bubbles in the reservoir and small bubbles in the tubing are completely normal and pose no danger. Do not be concerned about these, and don't bother calling to report these.
If you have followed this inspection procedure through several infusion set changes and have seen no significant bubbles in the reservoir, you can probably do these inspections less frequently than four times per day. But if you have any otherwise-inexplicable high blood glucose readings, you should suspect that air bubbles in the reservoir are the cause.
--------------------------------------------------------------------------------------------------------------------------------------
Thank you for your attention to this letter.
Sincerely,
DDDD J. HHHH, Ph.D. FFFF L. HHHH
----------------------------------------------
XXX YYYY Street
San Diego, CA 92ZZZ
Phone: AAA-BBB-CCCC
E-mail:
wands65mail-QQQ@yahoo.com
December 24, 2008
Medtronic:
Quality Control, Customer Support Center
Christopher J. O'Connell, Senior Vice President and President, Diabetes
The purpose of this letter is to ensure that you are aware of what we believe to be a significant safety issue with the Paradigm pump system involving formation of air bubbles in the pump reservoir and propagation of such bubbles through the tubing into patients' bodies. Such bubbles have the potential of resulting in serious or even catastrophic health effects on patients. This letter is - unfortunately - quite long, because we have tried to document it as completely as possible with the resources available to us, and because of the potential seriousness of the problem.
Contents of this letter:
1. Disclaimer/qualification
2. Executive summary
3. Our background
4. Our observations
5. Our interactions with the Medtronic HelpLine
6. Experiences of others
7. Our experiments
8. Chemistry 1 exercises
9. Our conclusions
10. Draft URGENT WARNING statement
1. Disclaimer/qualification
The information and conclusions in this letter represent our good-faith attempt to understand and describe the problems with our Paradigm 522 pump. We do not have access to the laboratory equipment that would be required to determine the precise origin of the problem. Every statement of fact in this letter should be understood to include an initial clause, "In our experience/judgment,...."
2. Executive summary
Air bubbles can develop at the interface between the o-ring and the reservoir body of the Paradigm pump (we'll term this the O-R interface). These bubbles can develop no matter how carefully the patient removes air bubbles during the infusion set insertion procedure; they appear to develop because of leakage of air into the reservoir around the o-ring as the pump operates. These bubbles can grow to become of significant size (>2 mm diameter), break loose, migrate to the top of the pump, into the tubing, and into the body of the patient, causing a significant (>1 Unit) loss of insulin flow into the patient's body. In turn, this loss can cause high blood sugar readings (>300 mg/dL); such spurious readings could also cause patients and their physicians to increase insulin doses with the possibility of resulting severe hypoglycemia and its severe consequences.
On or - preferably - before January 12, 2009, we ask Medtronic:
a. To convince us that our analysis of the situation is faulty. We regard ourselves as reasonable people, and would be delighted to be shown that something we (and other Paradigm pump users - see Section 6 below) are doing is wrong.
or
b. To resolve the problem and take the steps required to ensure that air bubbles resulting from action of the pump do not pose a significant danger to patients using the pump.
or
c. To contact every user of a Paradigm pump and every physician who prescribes Paradigm pumps to inform them of the possibility of development of such bubbles, the danger these bubbles present, ways of checking for bubbles in the reservoir and, and ways of checking for and removing air pockets in the infusion tubing. A draft of such an URGENT WARNING statement is given as Section 10 of this letter.
or
d. To take other appropriate action to address this problem.
3. Our Background
DDDD's bachelor's degree is in chemistry from the California Institute of Technology; his Ph.D. is also in chemistry from Stanford University; and he did postdoctoral work in biochemical neurology at the Columbia College of Physicians and Surgeons. He was a chemistry professor for 25 years at the University of California, EEEE, and is now an emeritus professor of chemistry. He has worked with and taught students about scientific instruments throughout his professional career.
FFFF was diagnosed with Type 1 diabetes in 2007 at the age of 63(!!). She has been using a Paradigm 522 pump since June, 2008. FFFF eats a very controlled diet, with ~90-105 g carbs per day. She uses ~20 Units of Novolog insulin per day.
4. Our observations
Approximately the first of November, 2008, FFFF began to have increased problems with high (sometimes >300 mg/dL) and erratic blood glucose (BG) readings. We eventually learned that the situation could be corrected by a new infusion set insertion, and thought perhaps the cannula had been inserted into a muscle (??) from which the insulin was not as well absorbed.
However, one time we noticed that there were large (>2 mm diameter) bubbles in the old pump reservoir when the infusion set was changed. We reasoned that if such a bubble worked its way into the tubing and into FFFF's body that a significant shortfall of insulin delivery could result.
We are scrupulous in eliminating bubbles during the infusion set insertion. We inspect both the reservoir and the tubing using a magnifying lens and (currently) a flashlight. We see absolutely no bubbles when the insertion set process is complete - the bubbles develop afterward, and the bubbles of significant size develop at the O-R interface.
Typically, a bubble will first be noticed by us when it's ~0.5-1 mm in diameter. Over the next day or so, additional bubbles develop, some bubbles stay the same size, and others become larger while they remain at the O-R interface, eventually becoming as large as 2-3 mm in diameter. When such a bubble works its way into the tubing, we currently "prime the bubble out" (see Section 10).
When we first began notice bubbles in the reservoir at the O-R interface, we saw only a few - or even no - bubbles at each infusion set change. However, as time has passed, the frequency of bubble formation has increased, so that we now see 5-15 bubbles at the O-R interface every time the infusion set is changed.
5. Our Interactions with Medtronic
(Dates are approximate, and some conversations are summarized or their substance mixed together - Medtronic has recordings of the conversations and can transcribe them, but we doubt that is important.)
First, a compliment: we talked mostly with Christine on the HelpLine. She was wonderful to work with: interested, empathetic, and obviously trying to do the best she could for us with the information available to her. Can you clone her to handle all HelpLine calls?
We called the Medtronic HelpLine, and the following was recommended:
a. Keep the insulin vial that is currently being used at room temperature rather than in the refrigerator. Note: as best we can tell, this instruction is NOT in the User's Guide for the pump.
b. Try de-gassing the insulin vial by repeatedly withdrawing air and creating a partial vacuum before filling the reservoir. We tried this procedure, found that it did not prevent subsequent formation of bubbles at the O-R interface, and it actually made it difficult to fill the reservoir because of the vacuum. In fact, this procedure tends to cause air bubbles to come into the reservoir around the o-ring. Aha! Air can leak around the o-rings. Note: as best we can tell, this instruction is NOT in the User's Guide for the pump.
c. Move the reservoir plunger up and down several times before filling the reservoir with insulin to better distribute o-ring lubricant. Note: as best we can tell, this instruction is NOT in the User's Guide for the pump.
d. Try different lot numbers for the reservoirs, and different batches of insulin. So far we're used at least 3 different lots numbers for reservoirs and 2 different insulin batches.
Following any or all of these recommendations had no effect. Air bubbles continued to develop at the O-R interface, to grow, and sometimes to break loose and enter the tubing. The longest air pocket we've seen in the tubing so far was about 6" (!!!) long, corresponding to a shortfall of more than 2 Units of insulin that would not have been delivered.
We were told that Medtronic believes that development of air bubbles in the reservoir is due to faulty patient technique, not to design or manufacturing flaws. We went over our insertion technique with the HelpLine, and they were unable to identify any deficiencies.
After several other conversations, a Medtronic contract nurse was assigned to come observe our technique on Tuesday, December 16. Also, Christine told us that replacing the pump was a possibility if the problems became more severe.
On Saturday, December 13, we did a regular infusion set replacement. We had seen one large bubble (~2 mm diameter) at the O-R interface that had remained unchanged for several days. However, one can see only a fraction of the circumference of the reservoir through the window in the pump. When we removed the reservoir from the pump, we observed 8 bubbles at the O-R interface, ranging in diameter between approximately 1 mm and 2.5 mm. Any of these could have broken loose and entered the tubing without us being aware of that happening unless we inspected the tubing.
We called the Medtronic HelpLine, talked to Shari (Christine was not working on that day), and explained our concerns. Shari said that, since a nurse was scheduled to come on the following Tuesday, there was no possibility of replacing the pump before that time, and that her decision to that effect reflected Medtronic policy. She further said that Medtronic believes all cases they have seen of bubbles developing in the reservoir are due to operator/patient error. Period.
The nurse on contract to Medtronic came on Tuesday, December 16 and observed our technique for insertion of an infusion set. Another compliment: the nurse (Sarah Johnson) was helpful, empathetic, and clearly cared about the patients she deals with. Ms. Johnson observed our technique for infusion set insertion, and found no defects in our procedure.
6. Experiences of others
We thought that we had a unique problem, or that something was wrong with our technique until we did a simple internet search and found that others have had similar problems. See, for example,
http://www.diabetesforums.com/forum/pumping-insulin/15923-air-bubbles-reservoir-tubing.html
Here's a posting:
"The pump is a minimed 722. The past two sets, when I was under 40 units remaining, I discovered air bubbles both in the reservoir and in the tube. When I filled the reservoir I was careful both times to make sure no bubbles were present so I don't know how they got introduced into the system. These bubbles caused my sugar to go >300 with corrections not working since I was not really getting the correct dosage (I started a new set immediately when I noticed this)."
(Searching in www.diabetesforum.com for, say, "pump bubbles" gives many discussions about problems with bubbles in the reservoir and tubing.)
http://www.diabetesdaily.com/forum/insulin-pumps/20871-air-bubble
Here's a posting:
"I got up this morning, checked my BG and it read 160. It should have been lower, but oh well. I programmed my pump for a 40carb breakfast and ate. After 1.5 hours I wanted to head to the gym so I checked my BG again—300! %$#$%! I started a correction bolus and while it was delivering I noticed a slight color change in a section of my infusion set tube. I looked closed and saw that it was moving! Turns out, it was an air bubble about 2 inches long. Have any of you had this problem? I have been having trouble with my morning boluses and now I’m beginning to wonder if I’m getting air bubbles in my line while I sleep."
http://tudiabetes.com/forum/topic/show?id=583967%3ATopic%3A109266
Here's a posting:
"I recently started using the Minimed Paradigm pump. I love using the pump it has made everything so much more convenient but I am having problems with bubbles. I have made sure the insulin is room temperature and when I fill the cartridge I make sure there are no bubbles. However, a couple of times I have had soaring blood sugars and after trying to work out what else it could be have changed my cartridge and found it full of bubbles. Has anyone else had this problem and what can I do?thanks"
There's even a youtube video on getting rid of bubbles!
http://www.youtube.com/watch?v=OwbelFCNGz4
The general rule of the internet is that for every such posting there are many, many more people with a similar problem.
In our opinion, it is difficult to believe that Medtronic is not aware of such forum discussions and problems people have with bubbles in the reservoir. One presumes that Medtronic monitors postings to such forums. Medtronic's policy to attribute air bubbles exclusively to operator/patient error ("Blame the patient") is, at best, irresponsible.
7. Our experiments
a. We filled normal disposable syringes with insulin, both with room-temperature insulin and with insulin taken directly from the refrigerator, inserted the needles into a rubber stopper to prevent leakage, and observed the syringes for several days. Did any bubbles develop? Of course not - neither in the room-temperature insulin nor in the straight-from-the-frig insulin. Medtronic's recommendation to begin with the insulin at room temperature to prevent formation of large bubbles appears to be irrelevant. Note that normal syringe plungers have a rubber-like end on the piston, not an o-ring seal. More on that later.
b. We filled a reservoir with insulin straight from the frig, did a "prime" with the reservoir plunger (that is, we didn't put the reservoir in the pump), clamped off the tubing, and cut off the end of the tubing to remove the insertion needle. We then allowed this reservoir to sit around for several days, and put it in a shirt pocket for several hours several times to allow it to warm to close to body temperature. Here is what we found.
There were several small, insignificant (see Chemistry 1 exercise in Section 8 below) bubbles (0.2 to 0.6 mm in diameter) at apparently-random places inside the reservoir. NO bubbles developed at the O-R interface. There might have been even fewer and smaller bubbles if we had used room-temperature insulin. It doesn't matter; the bubbles we did observe were insignificant.
This observation also answered one of our technical questions about whether bubbles might be more likely to develop spontaneously at the O-R interface than at random places on the plastic reservoir body because of gas nucleation dynamics. They didn't.
This observation also confirms that our technique for filling the reservoir without bubbles works. The bubbles aren't introduced by operator/patient error - Medtronic can't "Blame the patient." The bubbles at the O-R interface occur after the bubble-free reservoir is in the pump, the pump is operating, and the insulin is going into the patient.
After additional discussion with Christine, we were sent a new pump. This gave us the opportunity for additional experiments.
c. The problem is NOT with the pump! After FFFF began using the new pump, we filled the reservoir for the old pump, and did everything except the insertion of the Quick-set cannula into her body. We allowed the pump to run according to its usual settings, again, with the insulin flowing into the air instead of into her body. NO significant bubbles formed no matter what we did - e.g., fixed primes with the end of the tubing open to the air or clamped off. We believe that no problems will be found with the pump when it is tested.
d. We used the same infusion set lot numbers and the same insulin vial that was used for experiment c. above with the new pump, with the cannula inserted into her body. Bubbles formed in the reservoir for the new pump, just as with the old pump, with the bubbles of significant size forming at the O-R interface.
e. We thought that perhaps the pump and reservoir could get warm enough under the bedcovers at night to cause bubbles to form. The temperature of the surface of the pump can be as high as 95°F (it's under the bedcovers and close to the body). Parenthetically, we have no idea if a temperature of 95°F causes insulin to degrade too rapidly; we haven't observed any problems in the typical three days a reservoir is used. We slept with the pump that was being replaced (again, pumping insulin into the air instead of into her body), and there was no significant formation of bubbles.
To summarize these three experiments:
i. The bubbles apparently do not form because of a malfunction of the pump. Bubbles of significant size do not form unless the pump is operating to pump insulin into her body. Apparently, bubbles form because of an interaction involving some combination of the pump, the reservoir, the Quick-set cannula system, and the patient's body.
ii. Our technique for filling the reservoir and removing bubbles appears to be flawless, since bubbles do not form in a reservoir that isn't pumping insulin into her body but otherwise is operating normally.
iii. It's hard to imagine that anything we do with the Quick-set insertion could affect bubble formation.
iv. We have no idea what mechanism is responsible for the bubble formation.
8. Chemistry 1 exercises
a. 1 Unit of insulin corresponds to 0.01 mL of insulin. Exercise: Show that the number of Units of insulin in a bubble of diameter X mm is about 0.05 X3.
So if the diameter of a bubble is less than 1 mm, the number of units in a bubble is less than 0.05 - insignificant.
On the other hand, since the number of Units goes as the cube of the bubble diameter, a bubble 2 mm in diameter corresponds to 0.4 Units - if such a bubble goes into a patient, that represents a significant insulin defect. And a bubble 3 mm in diameter corresponds to about 1.35 Units of insulin. An insulin dose that's missing that much insulin because air rather than insulin gets into the patient could cause large changes in BG readings.
b. The diameter of the tubing corresponds to 1 Unit of insulin per about 2½-3" of tubing. Exercise: Show that a 1/4" air pocket in the tubing is insignificant, but an air pocket 1/2" or more is a concern.
9. Our conclusions
A joke of sorts: Somebody suggested that perhaps the people who designed o-rings for NASA in the 1980s are now designing o-rings for Medtronic.
We have discussed this problem with two friends. One is an instrumentation expert who repairs and rebuilds scientific instruments. His reaction: "It's most likely a problem with a mismatch between the o-ring and the reservoir. The o-ring seal simply isn't tight enough."
The other is a NASA scientist, who noted that o-rings are normally used for permanent seals between two stationary surfaces, not to form a seal between moving parts. Using o-rings to form a seal between moving surfaces is a dicey proposition. Again, note that the interface between the piston and the reservoir body on a normal disposable syringe is a rubberized end on the piston, not o-rings; and there's no air leakage into a syringe as a syringe piston moves forward.
We continue to be mystified about the development of bubbles at the O-R interface, especially about the following two points:
a. Why isn't bubble formation reported by many more patients? Perhaps the reason bubbles are not, according to Medtronic, regularly reported as a problem is because many patients use far more insulin than FFFF (she uses ~20 Units per day). A loss of a Unit or so every once in a while because of an air pocket in the tubing wouldn't be as significant to someone who uses much more insulin. We don't know the answer to this question.
b. What's the reason for bubbles to develop at the O-R interface and grow in size? Since the piston presumably moves only forward and is essentially under "forward" pressure, it seems unreasonable for bubbles to develop in the reservoir that's under positive pressure. Development of bubbles would, of course, be expected if the piston sometimes moved backward rather than always forward.
We simply do not know the reason for the origin and growth over time of the bubbles at the O-R interface. Perhaps the design of the system is faulty. Perhaps temperature cycling or pressure cycling of the o-rings and piston is causing the piston to move back and forth enough to cause air leakage. Perhaps pressure variations in the body (systolic/diastolic??) are transmitted to the o-ring seal system? We can only make wild guesses.
An obvious solution is to abandon the o-ring-based plunger and move to a plunger with a rubberized end, as in a regular disposable syringe. Presumably Medtronic considered such a possibility when the pump was designed, but a tighter-fitting plunger might make it more difficult to sense occlusions.
10. Draft URGENT WARNING statement
If Medtronic cannot find an error in our procedure for inserting an infusion set or cannot fix the problem of air bubbles, we believe that a statement along the lines of the draft below should be sent to every patient who uses a Paradigm pump and to every physician that prescribes them. We believe that bubbles that develop in the pump reservoir pose a significant danger to patients.
Notice: The URGENT WARNING statement below is Copyright 2008 by DDDD and FFFF HHHH.
Any private individual may reproduce or transmit this statement to any person of his/her choice.
However, any reproduction of any portion of this statement or any of the recommendations in this statement in any form (plagiarized, adapted, modified, etc.) by any corporation or company, or any representative of any corporation or company, will be regarded as a severe infringement of copyright laws and will subject the violator to criminal penalties and civil damages, unless they receive express permission from DDDD and FFFF HHHH.
The purpose of this notification is to prevent a company (e.g., Medtronic) from issuing a weasel-word warning in which they essentially continue to "Blame the patient" for formation of air bubbles in the pump reservoir. We will, of course, be more than happy to give permission to Medtronic or other companies to issue warning statements that show where responsibility lies as well as give vital information to patients and their doctors.
Caveat: The statement which follows is a draft statement. We are not physicians; nor are we experts on insulin pump design and problems. This statement is based on our experience on how to minimize the danger caused by bubbles which develop in insulin pump reservoirs. Patients should consult their physicians about their treatment regimen and how to operate the insulin pump.
--------------------------------------------------------------------------------------------------------------------------------------
URGENT WARNING
It has been found that air bubbles of significant size can develop in insulin pump reservoirs, mainly at the interface between the o-ring and the body of the reservoir. Such bubbles can develop no matter how scrupulously the patient removes bubbles during the process for insertion of the infusion set.
These bubbles can result in air pockets moving into the infusion-set tubing and into the patient's body - air rather than insulin goes into the patient's body. This replacement of insulin by air can cause high and even dangerous blood glucose readings (>300 mg/dL).
At this time, no one knows the reason for the origin of these bubbles. While the problem is being studied, the following procedure is recommended to avoid air pockets moving into the patient's body.
1. At least four times a day (e.g., before each meal and at bedtime), inspect
(a) the pump reservoir for bubbles through its window on one side of the pump and
(b) the tubing for air pockets in the tubing.
If a bubble in the reservoir disappears, it is likely to show up later as an air pocket in the tubing.
Inspection is assisted if a magnifying glass is used, and if a flashlight is also used (holding the flashlight under the tubing helps to identify air pockets in the tubing). Roughly speaking, when insulin is in a section of the tubing, the entire cross-section of the tubing looks similar to frosted glass. When there is an air pocket, one can see what looks like an empty space in the middle of the cross-section of the tubing. Look carefully; air pockets in the tubing are difficult to see.
2. If a significant air pocket (length >1/4" or >6 mm) is seen, disconnect the "Quick-set" (or equivalent) from your body. See your health-care provider if you do not know how to do this.
3. "Prime the bubble out." That is, deliver a "Fixed prime" sufficient to cause the air pocket to move completely through and out of the tubing. One Unit of insulin corresponds to 2½" of tubing. For example, if the air pocket is 6" from the end of the tubing, you could deliver a "Fixed prime" of 5 Units and be reasonably sure the air pocket would be removed from the tubing. Consult with your health-care provider before the first time you do this.
VERY, VERY IMPORTANT: After you have delivered the "Fixed prime" as described above, deliver another "Fixed prime" that corresponds to the "Fixed prime" you normally deliver when you insert the infusion set (typically 0.3 Unit). The reason for this additional "Fixed prime" is to make sure you have returned the pump to its original settings. If you do not, the next time you do a "Fixed prime," you could deliver, say, 5 Units, which could be disastrous.
4. If you see such air bubbles at the interface between the o-ring and the reservoir body or air pockets in the tubing, call 1-800-XXX-XXXX [phone number to be added] to report the problem.
Also important: An air bubble in the reservoir less than about 1/16" (or 1 mm) in diameter is not significant, nor is an air pocket in the tubing shorter than about 1/4" (or 6 mm). Small bubbles in the reservoir and small bubbles in the tubing are completely normal and pose no danger. Do not be concerned about these, and don't bother calling to report these.
If you have followed this inspection procedure through several infusion set changes and have seen no significant bubbles in the reservoir, you can probably do these inspections less frequently than four times per day. But if you have any otherwise-inexplicable high blood glucose readings, you should suspect that air bubbles in the reservoir are the cause.
--------------------------------------------------------------------------------------------------------------------------------------
Thank you for your attention to this letter.
Sincerely,
DDDD J. HHHH, Ph.D. FFFF L. HHHH
Wednesday, June 25, 2008
Tim Russert - Heart attacks, calcium scores, regressions, and life-saving lessons
Tim Russert, the moderator of NBC's Meet the Press program and head of its Washington, D.C. news bureau, died suddenly of a heart attack on June 13, 2007. It's a tragedy, especially for his family and friends. But it also offers an important - perhaps a life-saving - lesson for the rest of us. And it's also a good excuse to do a regression!
Russert's heart attack shouldn't have come as a surprise to anyone familiar with his medical condition - his heart was a land mine ready to go off for any reason, or for no reason. In fact, it's almost surprising that he hadn't already had a heart attack! More below.
First, the lesson:
Everybody should have a Calcium Scoring Test, men at about age 50 and women at about age 55. The results of the test should be used to guide lifestyle and medication decisions.
1. The information in this blog is based on news reports. I have no first-hand knowledge of any of the data. Russert's physicians and family have not released a full report - it's even possible there are ongoing financial negotiations about the apparently less-than-optimum care Russert received.
2. I have absolutely no financial connection to any company that has anything to do with Calcium Scoring Tests.
3. I'm a Ph.D. chemist/physical chemist/biochemist/retired chemistry professor, not an M.D. M.D.s are constrained by community standards of care and what is regarded as standard practice - and standard practices tend to change very slowly. If a physician suggested to his/her patients that a Calcium Score is more important than the patient's cholesterol level, the physician could be considered to be guilty of malpractice if the patient subsequently had a heart attack.
4. Whether the "baseline" test is done at age 45 or 50 or 55 (for men) probably isn't terribly important. For men with high risk factors (bad family history, high cholesterol, etc.), 45 is probably better. And similarly for women. The idea is to have the test before the Calcium Score (and plaque development) has become irreversible.
5. Very few physicians pay much attention to Calcium Scoring Tests. This is a situation where the patients have to lead the doctors.
6. Any time a physician tells you that you need to lower your LDL (bad cholesterol) to less than 75, say, "What about Tim Russert?" (This doesn't mean, of course, that cholesterol is unimportant heart-wise. What it does mean is that there are other important factors, one of the most important ones being the Calcium Score.)
7. For people who know nothing about Calcium Scoring Tests: The test is a CT-type X-ray that shows the degree of calcification in the coronary arteries. When plaque builds up in the arteries and sits around for a while, calcium builds up in the plaque. The amount of calcium is then a marker for the degree of atherosclerosis in the arteries, and is a reasonably good predictor of heart disease. There are several links at the end of this blog about the test. Facilities to give them are probably in every metropolitan area. Most insurance companies won't pay for them, because they're still considered experimental (at least, the insurance companies consider them to be experimental, even if the recent medical literature considers their usefulness an "answered that" question). As mentioned earlier, standard medical practice changes slowly, especially if there's money involved. The test is absolutely painless, except in the wallet.
8. What should you do after you have the Calcium Scoring Test and get the results? Talk with and try to educate your physician. If you have a high number (for example, higher than the median for your age), try to do something about it: Lose weight, improve your diet, exercise, and so on.
9. Will lifestyle changes stop the calcification? If a person has his/her first Calcium Scoring Test at a later age and the results are bad, will it be too late to do anything? These are questions to which there are no firm answers - the test is too new for enough patients to have been followed for long enough, and there simply aren't enough good data. What is for sure is that it's better to have bypass surgery or a stent insertion on an elective basis rather than in the middle of or after a heart attack.
Russert's heart attack shouldn't have come as a surprise to anyone familiar with his medical condition - his heart was a land mine ready to go off for any reason, or for no reason. In fact, it's almost surprising that he hadn't already had a heart attack! More below.
First, the lesson:
Everybody should have a Calcium Scoring Test, men at about age 50 and women at about age 55. The results of the test should be used to guide lifestyle and medication decisions.
1. The information in this blog is based on news reports. I have no first-hand knowledge of any of the data. Russert's physicians and family have not released a full report - it's even possible there are ongoing financial negotiations about the apparently less-than-optimum care Russert received.
2. I have absolutely no financial connection to any company that has anything to do with Calcium Scoring Tests.
3. I'm a Ph.D. chemist/physical chemist/biochemist/retired chemistry professor, not an M.D. M.D.s are constrained by community standards of care and what is regarded as standard practice - and standard practices tend to change very slowly. If a physician suggested to his/her patients that a Calcium Score is more important than the patient's cholesterol level, the physician could be considered to be guilty of malpractice if the patient subsequently had a heart attack.
4. Whether the "baseline" test is done at age 45 or 50 or 55 (for men) probably isn't terribly important. For men with high risk factors (bad family history, high cholesterol, etc.), 45 is probably better. And similarly for women. The idea is to have the test before the Calcium Score (and plaque development) has become irreversible.
5. Very few physicians pay much attention to Calcium Scoring Tests. This is a situation where the patients have to lead the doctors.
6. Any time a physician tells you that you need to lower your LDL (bad cholesterol) to less than 75, say, "What about Tim Russert?" (This doesn't mean, of course, that cholesterol is unimportant heart-wise. What it does mean is that there are other important factors, one of the most important ones being the Calcium Score.)
7. For people who know nothing about Calcium Scoring Tests: The test is a CT-type X-ray that shows the degree of calcification in the coronary arteries. When plaque builds up in the arteries and sits around for a while, calcium builds up in the plaque. The amount of calcium is then a marker for the degree of atherosclerosis in the arteries, and is a reasonably good predictor of heart disease. There are several links at the end of this blog about the test. Facilities to give them are probably in every metropolitan area. Most insurance companies won't pay for them, because they're still considered experimental (at least, the insurance companies consider them to be experimental, even if the recent medical literature considers their usefulness an "answered that" question). As mentioned earlier, standard medical practice changes slowly, especially if there's money involved. The test is absolutely painless, except in the wallet.
8. What should you do after you have the Calcium Scoring Test and get the results? Talk with and try to educate your physician. If you have a high number (for example, higher than the median for your age), try to do something about it: Lose weight, improve your diet, exercise, and so on.
9. Will lifestyle changes stop the calcification? If a person has his/her first Calcium Scoring Test at a later age and the results are bad, will it be too late to do anything? These are questions to which there are no firm answers - the test is too new for enough patients to have been followed for long enough, and there simply aren't enough good data. What is for sure is that it's better to have bypass surgery or a stent insertion on an elective basis rather than in the middle of or after a heart attack.
Now, more about Russert.
1. Russert had a Calcium Scoring Test in 1998 when he was 48, and had a Calcium Score of 210. This put him well above the 90th percentile for his age, and corresponded to a "moderate" heart attack risk. "Moderate risk" - ho, hum.
2. The important point is that Calcium Scores increase roughly 22% (typically, 15%-30%, depending on the individual) per year, corresponding to doubling every 3-5 years (see the Regression Lesson at the end). That is, suppose your score is 10 at age 45. If it doubles in 5 years your score will be 20 at age 50, 40 at age 55, and 80 at age 60, and so on. In Russert's case, a score of 210 at age 48 would correspond to a score of about 800 (doubling every 5 years) to 1800 at age 58 (doubling every 3 years) - most likely, about 1500. All of these are very high numbers, and are far beyond "moderate" risk levels. A recent Calcium Scoring Test very likely would have shown him to be at very high risk for a heart attack.
3. In fact, it's almost surprising that Russert hadn't already had a heart attack. His probable Calcium Score of 1500 or so at age 58 would put him in a very high-risk category, with about a 15% chance per year of having a heart attack (depending on the study one looks at, his chances per year might have been 5%, or they might have been 50% - there's no firm number yet, so we'll take 15% as a reasonable estimate). Let's suppose his risk at age 47 was 0% (meaning no chance of having a heart attack between age 47 and 48). Then we'll suppose his chances increased by 1.5% per year: a 1.5% chance between age 48 and 49, 3.0% between 49 and 50, up to a 15% chance between 57 and 58. The cumulative probability for him having a heart attack between ages 48 and 58 is about 60%. That is, it's not surprising that he had a heart attack by age 58; it's almost surprising that he didn't have one sooner. (See Cumulative Probability below for details.)
4. Unfortunately, his physician, like most physicians, clearly did not put much stock in the results of the Calcium Scoring Test, and instead put the emphasis on medication. The mantra these days is to lower cholesterol levels, mainly using statins such as Lipitor. (Statins are associated with muscle problems, liver problems, and mental problems; but that's a subject for another day.) This process was quite successful for Russert: his LDL (bad cholesterol) level was 68, which is an outstandingly good number. Also, his blood pressure was treated with medication, and was 120/80.
His physician was reported to have said that Russert's heart condition was well controlled with medication and exercise. This statement will not go down in medical history as a great moment, and is roughly equivalent to the old saying that "the operation was successful even though the patient died."
5. Russert's Calcium Score at age 48 should have been more than a red flag for him and his physician. It should have been red flashing lights, sirens, and flares. He should have been given very strong advice to change his lifestyle completely - or at least as much as humanly possible given his position. Weight loss, exercise, more relaxation, and so on should have been mandatory. And regular follow-ups, including Cardiolite or equivalent tests, could have well shown when intervention (bypass surgery or angioplasty with insertions of stents) would be required.
Whether lifestyle changes, follow-ups, and/or surgical intervention would have done the job and kept him alive longer is impossible to know. Russert may have had a genetic condition that pre-disposed him to a fatal heart attack. But if the information from the Calcium Scoring Test had been properly used, at least he and his physicians could have said they did the best with the cards he was dealt. The understatement of the day: They didn't.
Once again, here's the lesson:
Everybody should have a Calcium Scoring Test, men at about age 50 and women at about age 55. The results of the test should be used to guide lifestyle and medication decisions.
If you follow this lesson you'll be able to say you did the best you could with the cards you were dealt - even if you're hit by a bus as you leave the Calcium Scoring Test facility!
Here are a few links about Calcium Scoring Tests.
http://www.mayoclinic.com/health/heart-disease/HB00015
http://www.heartsavers.md/index.html
http://www.americanheart.org/presenter.jhtml?identifier=10000015&q=calcium+scoring&x=24&y=6
http://brighamrad.harvard.edu/patients/education/ct/ctguideheart.htm
There are lots more, and many, many articles saying that Calcium Scoring Tests are very useful - even though they haven't made it into standard medical practice.
Currently, a test costs about $500. However, substantial discounts are often available, for example, if a spouse or relative has had a test. Mine cost $200 because a relative had recently had a test. A suggestion: Get 5 or 6 people to go in together for testing, and ask the facility what they can do for you price-wise.
Of the six or so people I know who have had tests in about the last year, one had a high score, further evaluation, and eventual bypass surgery. Others with relatively high scores are trying to modify their lifestyles. I'm using the results of mine mainly to fight off physicians who say I should take statins to lower my cholesterol level (varying from year to year from about 180 to 220, but I have no significant risk factors for heart problems).
Regression Lesson for the Day
Here are the median (50th percentile) values for the Calcium Score for asymptomatic males as a function of age:
(The "Log" represents the base-10 logarithm of the calcium Score; base-e logarithms are more familiar to scientists but less familiar to everybody else.)
Age............... Calcium Score...................... Log (Calcium Score)
45..................... 3............................................ 0.48
50.................... 5............................................. 0.70
55................... 22............................................ 1.34
60................... 70............................................ 1.85
65 .................160............................................ 2.20
70................. 300............................................ 2.48
The Calcium Score clearly doesn't increase in a straight line (linearly) with age; in fact, it increases exponentially with age. Draw a graph of Calcium Score versus Age - that's nothing like a straight line. If you crank up your Regression Machine and regress Calcium Score against age, you'll find a slope of about 11 and - more important - a correlation coefficient (R-square) of 0.83, indicating a relationship, but not a good linear one, between Age and Calcium Score.
Let's do a better job. To a good approximation, the logarithm of the score increases in a straight-line fashion with age. Draw a graph of Log (Calcium Score) versus Age - not all that good a straight line, but close enough to be useful.
Crank up your Regression Machine and regress Log (Calcium Score) versus Age. For these data, you'll find:
Slope = 0.086 (standard error of 0.006 - a good fit)
R-square = 0.98 (a quite-good linear fit)
It turns out that - isn't this fascinating! - the slope of the regression line is the reciprocal of the number of years it takes the Calcium Score to increase ten-fold (in this case, about 12 years). With a little math, you can show that the number of years it takes the Calcium Score to double is 0.30 divided by the slope, in this case about 3.5 years.
You can do the math to show that in ten years the Calcium Score should increase by about a factor of 7. That's why we would estimate Russert's Calcium Score to be roughly 1500 in 2008, ten years after his first test in 1998.
A comment: This slope and doubling period is for the median. People at a higher (worse) percentile tend to increase more slowly, either because that's how the body works for them or because the people with higher numbers tend to die off.
Cumulative Probability
Let's suppose Russert's chance of having a heart attack between age 48 and 49 was 1.5%, increasing by 1.5% per year. That is, his chance of not having a heart attack between age 48 and 49 was 97%. We'll also suppose that his chances increases by 1.5% per year, to be close to 15% between ages 57 and 58 (the 15% figure is about what would have been expected, based on his projected Calcium Score). Here are the figures:
..................Chance of .........Chance of NOT ...........Cumulative Chance of NOT
Age ..........Heart Attack ...Having Heart Attack ..Having Had a Heart Attack
..................at that Age........at that Age...................by that Age
48 to 49......1.5%......................98.5%...................................98.5%
49 to 50......3.0%.....................97.0%.................................. 95.5%
50 to 51......4.5% .....................95.5%.................................. 91.2%
51 to 52 ......6.0% ....................94.0% ..................................85.7%
52 to 53 ......7.5% ....................92.5%................................... 79.3%
53 to 54 ......9.0% ....................91.0% ..................................72.2%
54 to 55 .....10.5% ...................89.5% ..................................64.6%
55 to 56 .....12.0% ...................88.0% ..................................56.9%
56 to 57 ......13.5% ...................87.5% ...................................49.2%
57 to 58 ......15.0% ...................85.0% ..................................41.8%
The cumulative chances are calculated by multiplying the number for the previous age by the chances of not having a heart attack that year (converted from a percentage to a decimal). For example, the 72.2% figure for ages 53 to 54 is the chance of not having a heart attack before that age (79.3%) times the chance of not having a heart attack in that year (91%). For Russert, the cumulative chance he hadn't had a heart attack by age 58 was only 42%, meaning that, statistically, there was about a 60% chance he would have already had a heart attack. By the time he was 56, he had at least a 50% chance of already having a heart attack and was living on borrowed time.
That's all for now.
Sunday, August 19, 2007
Projected US temperatures in 2050 - U.S. and/or global warming?
What will US temperatures be in the year 2050?
Many have predicted that ongoing global warming, due in large part to human activity, will produce a US meltdown by the year 2050, with much of Florida under water, rampant tropical diseases such as malaria, and the US essentially going to Hell temperature-wise. What’s actually most likely to happen?
The Bottom Line
In 2050, the US temperature is projected to be essentially the same as it was in the period 1995-2005.
Background
The US government publishes standardized temperatures for the US through the years, trying to compensate for heat effects of urbanization, etc. These temperatures are called “temperature anomalies” but we’ll call them “temperature differences“ - the word “anomalies“ has an implication of being unusual. These are some sort of average of temperature readings all across the continental US. The temperature differences are relative to the average for 1950-1980. That is, a temperature of 1 °C doesn‘t mean the average temperature was 1°Celsius; it means the temperature for that year was 1° Celsius hotter than the 1950-1980 average. These temperature differences can be found at the following web site:
http://data.giss.nasa.gov/gistemp/graphs/Fig.D.txt
Note: when going to this site, type it in exactly as shown. For example, fig.d.txt (lower case) gives an error message.
More information can be obtained by moving backward through the site; e.g.,
http://data.giss.nasa.gov/gistemp/graphs/
Note: The temperatures were recently updated because of a Y2K glitch, caught by an outsider, not by the government scientists who produced them. (Is anybody surprised?) It is not known how many additional errors are in their data and interpretations, because the source code and original data are not available to the public. But we’ll use these data as the best currently available.
Analysis
Suppose one wants to predict the temperature for some year. The absolutely simplest form of analysis imaginable is to do a linear regression of temperatures versus time
Y= a + b X, where Y is temperature and X is the year
with the year for which the temperature is being predicted set to X=0 in the regression equation. The intercept of the regression line is the temperature of the year of interest, and the slope is how rapidly the temperature is changing.
The regression analysis is equivalent to finding the best-fit straight line for a graph of temperature differences vs. year and then extrapolating the line to find the temperature for the year in which you’re interested. See Fig. D (“U.S. Temperature”) on the site
http://data.giss.nasa.gov/gistemp/graphs/
Here’s a simple example of such a regression calculation. Suppose in 1934 Al Gore’s grandfather had the temperature data available only for the years 1924-34, and decided to predict the temperature for 2004. The data would look like:
Actual ...........Year (for ..............Temperature
Year ..............regression) ..........(°Celsius)
1924 ...............-80......................... -0.74
1925 ...............-79 ...........................0.36
1926 ...............-78 ...........................0.04
1927 ................-77 ...........................0.15
1928 ...............-76 ............................0.07
1929 ...............-75 ...........................-0.58
1930 ..............-74 .............................0.16
1931 ...............-73 ............................1.08
1932 ...............-72 ...............................0
1933 ...............-71 ...........................0.68
1934 ...............-70 ...........................1.25
Put these numbers into an Excel spreadsheet, crank up the Tools\Data Analysis package, and regress “Temperature” against “Year (for regression),” and you get the following results - and lots more statistical information as well. Temperatures are given in both °Celsius and
°Fahrenheit, because a typical US citizen hasn‘t got a clue what a Celsius temperature is.
Predicted average US temperature difference in 2004: 9.1 °C (16.4 °F)
95% confidence interval: 1.2 °C to 17.1 °C (2.1 °F to 30.7 °F)
Slope (change in temperature per year): 0.12 °C/year (0.21 °F/year) - that is, the regression equation says the temperature increases 1 °F every 5 years
Significance level of F-statistic: 0.031 (without being too fancy about it, the significance level indicates a significant relationship at the 97% confidence level between temperature and year)
So Al Gore’s grandfather could have made a movie and gone out on the lecture circuit talking about how the US would be impossibly hot in the year 2004. The predicted temperature in 2004 would be about 9 °C (16 °F) hotter than the average during the 1924-1934 period (0.2 °C or 0.4 °F temperature difference) and perhaps as much as 17 °C (30 °F) hotter. A truly frightening possibility of a truly frightening calamity by 2004!! Back to reality: The actual 2004 US temperature difference was 0.44 °C (0.79 °F). OOPS!
This example was just to show how the regression process works - try it yourself to see if you get the same numbers. It’s obviously silly to extrapolate too far from the actual observations. For example, it would be downright stupid to extrapolate from the US temperatures in 2000-2006 to say the sky will be falling and Florida will disappear into the ocean by 2050. OOPS again! That’s what people do.
But let’s try a serious analysis. We want to predict temperatures in 2050, using temperatures through 2006; that is, 44 years in the future. So let’s see how good a job we can do at predicting temperatures 44 years in advance.
We’ll try predicting the temperature difference in 2004 using temperature data from 1880-1960. The reason for choosing 2004 is that it’s the last year we can look at the 5-year average for 2002-2006 as what we’re really trying to predict. There are enough year-to-year variations that it’s better to predict the 2004 temperature difference, and see how closely the predicted temperature compares with the 2002-2006 5-year average, not just with 2004 itself - although we can also make that comparison.
So we take the 1880-1960 data, set 1960 as year -44, 1959 as year -45, etc., and do the regression with the intercept being the predicted temperature difference for 2004. Here’s what we find:
Predicted average US temperature difference in 2004: 0.66 °C (1.18 °F)
95% confidence interval: 0.30 °C to 1.01 °C (0.55 °F to 1.82 °F)
Significance level of F-statistic: 0.0003 (significant at the 99.97% confidence level)
The actual temperature differences in 2002-2006:
2002 0.53 °C (0.95 °F)
2003 0.50 °C (0.90 °F)
2004 0.44 °C (0.79 °F)
2005 0.69 °C (1.24 °F)
2006 1.13 °C (2.03 °F)
And the actual 5-year average for 2002-2006: 0.66 °C (1.18 °F)
Obviously, the agreement to within 0.01° between the predicted 2004 temperature difference and the average of the 5 years around 2004 is sensationally good. In fact, much, much better than one would reasonably expect. And the agreement between the predicted and actual 2004 temperatures (within about 0.2 °C or 0.4 °F) isn’t shabby.
Now that the model for the temperature estimation has been validated way beyond any reasonable expectations, let’s address the multi-trillion-dollar question: What will the US temperature be in 2050?
What we’ll do is to take all the data we have (1880-2006), and do the linear regression for the 2050 temperature. (In the regression, 2006 is year -44, 2005 is year -45, down to 1880 is year -170). Here’s the result:
Predicted average US temperature difference in 2050 (the intercept): 0.58 °C (1.06 °F)
95% confidence interval: 0.35 °C to 0.82 °C (0.64 °F to 1.48 °F)
Change in temperature per year (the slope): 0.0048 °C/year (0.0087 °F/year) - the regression predicts the temperature will increase 1 °C about every 200 years or 1 °F every 115 years.
Significance level of F-statistic: 0.00001
Let’s compare this result to some recent temperatures. The average temperature in the period 1995-2005 was 0.53 °C ( 0.95 °F).
In other words, the linear regression predicts that the US temperature difference in 2050 (0.58 °C or 1.06 °F) will be essentially the same as it was in the years 1995-2005 (0.53 °C or 0.95 °F). A bit warmer than the average since 1880, but not exactly a big deal.
A few quibbles, caveats, etc.
1. There are as many ways of fitting data to equations as there are statisticians. The easiest way to begin is with the simple linear equation. That’s always the first place to start, and most of the time does about as well as much fancier equations. Although, of course, there are data for which a linear equation either doesn’t work or can be improved upon.
Anybody can use any equation, and see if that equation produces a significantly better fit than the simple linear equation. There are straightforward statistical tests to see if a fancier equation improves the fit. We’ve tried a few without much success, but anybody is welcome to try any other equation.
2. Whether it will be successful to do a straight-line extrapolation of previous temperatures for 44 years into the future is an open question. The extrapolation worked sensationally successfully 44 years in the “future” to 2004. What will happen going forward to 2050 is unknown. Perhaps a gigantic meteor will hit the Earth, even more gigantic volcanoes will erupt, or the cataclysmic tipping point events predicted by the global warming crowd will come to pass. Who knows?
3. The prediction here is only for US temperatures. Somebody might want to try the equivalent exercise for temperatures in other parts of the world.
4. The advantage of the linear extrapolation of temperatures is that the data and analysis are completely transparent - anybody can reproduce the calculation or look up the data. But ask any climatologist who has a climate model to show you the source code - lots of luck getting it. It’s a closely-held secret whether or not the code contains a line
IF ((T2050 - T2006) < 2.5) THEN (((T2050 - T2006) = 2.5) AND CALL AL GORE)
In modern science, data are considered proprietary trade secrets, even when the data are gathered on government contracts and grants.
The general rule is that you should take with a very large grain of salt any projection showing major effects on temperature due to human activity when the projection is done by somebody whose grant funding depends on getting results showing that human activity causes temperature changes. Too many scientists are more than happy to sing for their suppers whatever tune strikes the fancy of the person paying for the supper.
5. And a question to which we’ve never heard a very good answer: What will be the temperature effect of injecting, say, 1,000 kg of carbon dioxide into the atmosphere? 0.0000001 °C or 0.0000000000001 °C or what? Does anybody have an answer and, more important, will they let other people check their calculations?
Subscribe to:
Posts (Atom)