Monday, December 30, 2013

Three weeks and three days on the 530 G

So far: have worn six quick sets. The fifth one I put in on a Monday morning, bg started rising about three hours later and I bolused a lot trying to get it down; also I skipped lunch. It went down to about 240 before turning around and heading back up over 300; at that point I tested urine ketones, and they were very large so I pulled the site. The cannula had folded on itself completely. So, that explained that! I put in a sixth one and it worked okay, although when I pulled it there was blood in the cannula.

So far: have worn two silhouettes. The first one (at a 45 degree angle in my arm) had good absorption but it felt horrible and when I pulled it, the area looked like it had been worn raw. The second one (20 degree angle in my lower back) also had good absorption but about two days after I put it in it started bleeding again. I pulled it. The area has a fairly superficial bruise and it's tender and swollen and it hurts hurts hurts.

I put in my first sure T yesterday. I'm sure it went into muscle, but it feels okay-ish. Sometimes I feel it and it feels like I've got a needle in me, which I do, so that makes sense.

I met with the pump trainer on Friday and discussed sites. He thinks my best bet is silhouettes, to avoid muscle. I'm not so sure.

I really feel that wearing infusion sets has so far big a much more problematic and painful experience than people had led me to expect. Ouch ouch ouch!

I am finally wearing a sensor that's lasting. It's now on day 8, and performing well.

Saturday, December 21, 2013

Two weeks on the 530 G

Overall: Wish I had never gotten it. Sometimes feel okay or even cool with it, much of the time feel frustrated, scared, itchy, and tender.

My first infusion set (6mm quick set on my butt) had no delivery alarms and had to be pulled. My second infusion set (6mm quick set on my butt) was okay. My third infusion set (6mm quick set on my thigh) was okay although really tender, but when I pulled it, it bled so much that there was a puddle of blood as big as my hand on the floor. It was scary. My fourth infusion set (6mm quick set on my butt) was okay. My fifth infusion set (13mm silhouette on my arm) is still in my arm right now and it's pretty tender but working okay.

My first and second sensors I wrote about already. The third did okay through the fifth day but on the sixth day the ISIG:bg ratio went bad quickly resulting in false lows and finally in SENSOR FAILED shortly before the six day period was up. The fourth sensor has been performing with terrific accuracy (and is now five days in) but some WEAK SENSOR and SENSOR LOSTs have occurred.

My bgs have been worse than usual but not dramatically so.

Overall though I just don't like pumping. I'm not sure how much of it is the newness and how much of it isn't. I don't know whether or not to return it. I'm disappointed that it doesn't have a lot of features I think it should have (including features I expected because the Guardian has 'em as well as features found on other pumps and features that aren't found on other pumps that I just think it ought to have), and frustrated or annoyed by many of the features that it does have. 

Thursday, December 12, 2013

5th day on the 530 G

In the night I woke to find that the pump was reading Lost Sensor. I reconnected, but because I went back to sleep without calibrating, I woke up later to no data from the medT CGM almost all night. Even though I calibrated in the morning, I again got a "Meter BG now" and no data during a class I taught this afternoon. So there's another hour's data missing.

I thought I understood that the minimed CGM would give me gaps if I went for 12 hours without calibrating (unlike with the Dexcom) but I did n't remember or expect all these gaps. This is hardly CONTINUOUS glucose monitoring.

Pump worked okay and bg was okay-ish most of the day.

Wednesday, December 11, 2013

4th day on the 530G

The CGM performed better over the course of the day. Nothing failed. I spent most of the day outside my bg targets- the night high, the day low. Not low enough to trigger the low suspend. Saw my endocrinologist. Discussed site issues and so forth. He was encouraging.

Tuesday, December 10, 2013

3rd day on the 530G

The evening blood sugar after the supper with the combo bolus was a lot like if I had had shots- blood sugar was fine during supper but rose later. It didn't rise a lot though. I bolused a correction and for a snack at 9 PM.

At 4:45 AM I woke up to a Dexcom reading of 308, but happily enough my meter only read 256. I got a No Delivery alarm while trying to correct, and I took off the tubing from the site to really look at it. It still looked fine. I reconnected and told it to resume, then programmed in the basal that I wanted, and I went back to bed.

At 8 I woke up to a dropping blood sugar- Dexcom read 175. I decided to bolus for breakfast a little bit ahead of time, but a little over a unit into the ~9 unit bolus, I got another No Delivery alarm, and this time when I hit resume and tried to bolus again, I got a No Delivery alarm before it had bolused anything at all.

So I called the hotline. The person on the phone had me look at the stuff which still looked fine. She had me disconnect from the infusion site and try priming 5 units. The 5 units dripped out without triggering a No Delivery alarm. So she asked me to take out the site to see if maybe the tubing had kinked. I took out the site and looked. It was straight but there was blood in about 1/3-1/2 of the cannula. She said that was probably the cause of the No Delivery.

I cleaned off and put in a new site not far from the old one. I have 18" tubing and I don't think that would be adequate for very many locations. I had to stop myself from laughing at the lady on the phone's advise regarding sites on the abdomen: at least 3" from the belly button, and 2" from each other, and all around including up and down from the belly button. But that is not possible on this body of mine. My abdomen is roughly 11" wide. From the bottom of my rib cage to my hip bones is 3" if I'm standing up.  So yeah.

Anyways. I put in a new site. Being on the phone and doing the new site took me enough time that there wasn't any time left for breakfast, which meant my basals were way off for the workday because I have really different insulin requirements on days when I don't eat breakfast. I was cranky about it.
 I got home to two boxes and a letter from medtronic... and that is a start to day 4.

Monday, December 09, 2013

Second day on the 530 G

I went to bed around 10 pm last night with my Dexcom G4 going and the pump component of the 530G, but no enlite because the senserter broke. I had about a unit on board but I thought I had a lot of food in my belly still digesting, and my bg was 100 (yes really with two zeroes).
I woke up at 5:30 to a Dexcom that had run its battery fairly low going "LOW" since roughly midnight. If I'd had an enlite sensor on, of course, that would have been the perfect night to see how well the 530G could perform, but of course, I didn't.
I ate a small "let's get this low treated" breakfast and then went to the synagogue for morning prayers. While I was there the high alarm on my Dexcom went off (oops guess I overtreated that low) and I realized that I couldn't bolus without beeping unless I changed the alert types to vibrate instead of beep, which would require scrolling through the menu which would also beep. And I couldn't step out because I was needed for the prayer quorum. Sorry for beeping, everybody!
I had a second breakfast after prayers and my blood sugar dropped steadily until I ate lunch, leveled off, and then started dropping again. I ran a temp basal of 0.05 u/hr for an hour to try to head it off but that didn't seem to do much.
After work I set the basal for the 10 am to 3 pm period lower, from 0.325 to 0.275 u/hr (I have the midnight to 10 am basal at 0.450 and the 3 pm to midnight at 0.375 u/hr)
It had just come up (an hour after I ate about five fig newtons) when it was supper time, and I decided to do a dual wave bolus because supper was rice and lentils, a very favorite supper of mine that tends to cause late highs. Wouldn't you know it, a little after I put the pump down, still in the delivery of the bolus, I got a "No Delivery" alarm (my second one so far- and they were 36 hours apart FYI). I checked the tubing, which still looks fine to my inexperienced eyes, and hit "resume". But although that makes it resume basal, it had totally abandoned the bolus- both the remaining up front part, and the square wave part. So I looked at the bolus history, decided to guesstimate it, and that's what I did.

On another note regarding sites: the infusion site just now started feeling a little itchy and achey but is overall okay. I think I may have pulled on it too much. The place where I had a sensor is sore and the area where the overtape went- the extra square of tape- is still pink and tender. I think I need a different tape. Sigh! I'll try it as sent for one more sensor and if that one is also not so great, I'm going to try using Opsite Flexifix (which I don't love but which is OK on my skin).

Sunday, December 08, 2013

First 24 Hours on Medtronic 530G

Note: For new readers or those who've forgotten, I have never been on an insulin pump before this, but I have used three different CGMs prior to this system, I have worn I-Ports, and I'm more than comfortable setting pump settings for other people. And I've been using a jet injector for the last thirteen months. I use Novolin R.

My pump and sensors and infusion sets and manuals(!) and senserter and quick serter and reservoirs and charger and transmitter for the system came on Monday. I spent a few days talking to the trainer and starter and all those people going back on forth on whether or not I need training on the system and to wait and all that, and on when my 30 day return window counts down from.

Last night I decided to try wearing the system. Since I had already set up the pump with the settings I wanted, I put on a temp basal of 0.025 u/hr so as not to be giving myself basal in addition to the lantus, I took out the insulin from the fridge, I took a bath and shaved the areas where I intended to place the sensor and the infusion site, and then I assembled the stuff.

First: I loaded up the reservoir. I put in the full 1.8 mL of air into the vial of insulin, which was a mistake. It was too pressurized. I tried to draw out 140 units of insulin and to not have any bubble and I thought I had succeeded, but when I unscrewed the reservoir from the blue thing that connects the reservoir to the insulin vial, I could see a huge bubble. After attaching it to the connector with the tubing, I had some difficulty pushing the air out, but I think I managed that too. However, it was really difficult for me to tell when the bubbles were all out.

Putting the quick set into the quick serter was not as easy as expected- there was a bit of a trick to getting the set into the serter. And it hurt more than I expected going in. But- it's in. I put it on my left butt cheek. I took a short break before coming back to insert the sensor. The needle looked scary big and I have bad memories of sof-sensor insertions. The sensor went in pretty easily - I put it in my left abdomen. The sen serter hit the area around my abdomen with kind of the distraction thing going for it, and it didn't hurt much. I looked at the pictures of the taping up and tried to do it right, but wasn't sure I got it. I connected it with the transmitter.

The ISIG of the sensor took about an hour to really correspond with my Dexcom readings. Once I calibrated it, it gave readings that didn't agree 100% with my dexcom or my meter, but they were in the neighborhood. Feeling that the enlite sensor was working well, I took out my Dexcom sensor in the morning when the one week period of the Dexcom sensor was up Unfortunately, about 20 hours after insertion, my sensor abruptly failed. I got "weak signal" and they "sensor lost". I called the minimed 24 hour hotline and the woman on the phone said it might have been because I did tape it wrong, but they're replacing the sensor. Then I tried to put in another sensor... but the senserter did not deploy. I called the hotline again and they said my senserter had broken... and it broke that sensor with it. So they're replacing both.
I put on a Dexcom sensor.

The infusion site and tubing has been OK. It's been hard figuring out how to move in respect to it. I think I want longer tubing (I have 18"). I've been very relieved that I can't feel the insulin going in. In the evening my blood sugar rose a lot and I bolused four times to try to get it down, then gave up and did a shot with my jet injector, set an alarm, and woke up in the night ... and the blood sugar was fine! I woke up again around 6 am when my blood sugar was a little high (Dexcom read 161 and the pump read 148) and I tried to bolus 1 unit. The pump alarmed "No Delivery" but when I hit resume delivery I guess it did, because when I woke up again my blood sugar had dropped a lot. Breakfast spiked but came down and I had a mild hypo right after the sensor failed, so I guess I'm getting insulin.

Saturday, November 30, 2013

Your Doctor is Rated on Your Health

Through the Affordable Care Act, physicians will be publicly ranked for how well their patients are doing, especially if the patient is being paid for through medicare and the physician is in a relevant field.

For diabetics age 18-75, the physicians will be ranked on-

Whether or not your A1c is below 9%

Whether your blood pressure is high or not

Whether or not your LDL is low enough (<100 dl="" mg="" p="">
Whether or not you been examined for retinopathy in the past year.

Whether or not you have had a urine test for kidney disease in the past year.

Whether or not you've had an exam of your feet and ankles for neuropathy in the past year (it's unclear to me exactly how extensive they intend the exam to be).

Whether or not the doctor has looked at your shoes or other footwear to decide if it's safe or not.

Whether or not you have been tested for high blood pressure in the past 3 years (this one applies to adults up to 90 and not merely to 75).

Whether or not you meet ALL of the measures of a perfect diabetes patient: A1c below 8% PLUS blood pressure below 140/90 PLUS LDL below 100 mg/dl PLUS no smoking PLUS if you have heard disease, you're on aspirin

Whether or not you've had lipid testing in the past year.

If you have another doctor because of another disease, whether or not there is communication between the two doctors.

If you have a diagnosis of diabetic retinopathy and are any age, then whether or not the doctor has received proof of an eye exam in the past year.

If you have both a diagnosis of certain types of heart disease AND diabetes, then whether or not you have been prescribed a medication in the ARB or ACE inhibitor drug classes.

If you are 5-17 years old and have diabetes, whether or not your A1c has been tested in the past year.

And for all patients, the physician will be ranked based on things like what portion of patients have received a flu vaccine.

This ranking is theoretically intended to allow us to choose better doctors. But you can find even on the government website a list of reasons why this might not be a good idea- namely, your doctor will suddenly have an incentive to stop accepting or stop seeing patients that he has reason to believe will not look good for him. Also, this provides doctors with a reason to do unnecessary testing. For instance, there is really no good reason to test my lipids. Yet because my insurance company rates the doctor on these same things, my lipids have been tested annually for years. And guess what? Every year my TOTAL lipids is under 110 mg/dl and my LDL is under 50 mg/dl. That there is no point in testing the lipids of a person whose lipids were well below normal the previous year is not something that they take into consideration- as far as the writers of these standards are concerned, if I have diabetes, I must be at high risk of high cholesterol levels. Whereas I believe that previous lipid levels are a much better predictor, and that if the one test is not only not high, but below normal, there is no point in testing for high levels for another decade (or longer, probably).

Anyways. You can read or scan or search within the 1,369 page document here:

Thursday, November 28, 2013

Not My Idea- One Way Long-Term CGM Sensors Might Work

Dexcom years ago tried to develop a sensor that would be implanted for six months or a year- a long term sensor- which is why their sensors- the generation before the Dexcom 7, the Dexcom STS- was called Dexcom Short Term Sensors.

So obviously, making and marketing the Dexcom LTS hasn't gone much of anywhere so far.

But people are still working on developing long term sensors and playing with ideas that might make them work. One that I just read that seems sort of far out to me is the idea of using things to cause local growth of blood and lymph vessels near the placement of the sensor!

Wednesday, November 27, 2013

Intern/Volunteer Wanted by Pancreum for Artificial Pancreas Project

Not much of a fact, but if you know a mechanical engineer....

Some facts about glucagon...

#1 Most type 1 diabetics have abnormally high glucagon levels all the time. The local high levels of insulin in people whose pancreata make insulin tell the alpha cells to keep it down. In type 1 diabetics whose alpha cells are not too damaged, the alpha cells make a lot of glucagon.

#2 Injecting glucagon causes nausea in some people, even when used at low doses. This is one of the barriers to using an artificial pancreas; about 1 in 7 people who has gotten glucagon as a treatment for mild hypoglycemia by an artificial pancreas program has reported feeling nauseous. It's not clear why glucagon makes some people nauseous and not others.

#3  Although plans for glucagon pens have been around for quite a while, right now you can only get glucagon in the US as glucagon kits or glucagen kits. The glucagon kit from Eli Lilly is red. The GlucaGen kit is made by Novo Nordisk and it's orange. Both can be stored at room temperature.

#4 The instructions inside the kit are for extreme low blood sugar, where a person has passed out, and you might as well give a full dose intramuscularly. Some folks use a "mini dose" with a regular insulin syringe subcutateously, taking 0.15 mL (1mg/mL) for hard to treat but not quite as bad hypoglycemia, the full dose being 1 mL (- 1 mg) for adults and 0.5 mg for young children. My doctor recommends the mini dose but I never have tried it yet.

#5 Glucagon acts on the glycogen stores of the liver. In people with liver disease, or who are drunk, or whose glycogen stores are depleted due to fasting, illness, or previous use of glucagon, the glucagon will not raise blood sugar.

Tuesday, November 26, 2013

Islet Cell Composition

The islets of the pancreas contain five types of cells, named after the first five greek letters.
Alpha cells make glucagon, which stimulates the liver to release glycogen and produce sugars.
Beta cells (which make up 85% of the islets) make pro-insulin (which splits into insulin and cpeptide) and pro-amylin (which becomes amylin).
Gamma cells are often called PP cells; they make a  "pancreatic polypeptide". I find this a really silly name given that insulin is also a panceatic polypeptide.
Delta cells make somatostatin (we have other places in the body were somatostatin is made).
Epsilon cells make ghrelin, which is mostly made elsewhere in the body, and which is a hormone that has a lot to do with controlling hunger.

Monday, November 25, 2013

In embryonic development, the pancreas begins in two pieces.

The pancreas develops from two buds growing from the proto-duodenum, and they merge towards the end of the first trimester.
Human embryos begin making insulin around the beginning of the fifth month of gestation; during the second trimester.

Sunday, November 24, 2013

Reactive Hypoglycemia

The most common form of hypoglycemia in people who are not on medications causing hypoglycemia is reactive hypoglycemia.

In the average person's postprandial blood sugar curve, the blood sugar rises during and after the meal, then falls to slightly below the starting blood sugar, and then rises back to the starting blood sugar.

In many people whose pancreata are struggling to make enough insulin, or whose initial insulin response to meals is blunted (ie not making enough incretins or not responding to incretins), after the initial blood sugar rise, there is a drop in blood sugar that is much more dramatic than the normal drop that results in hypoglycemia. It is called reactive hypoglycemia because the hypoglycemia is caused by an over reaction to food.

It should be treated by a diet with no big meals, and snacks through the day. An extreme low carbohydrate diet is usually fairly good at preventing lows, but it isn't necessary; a diet in which a person eats a roughly consistent number of carbs every couple of hours during the day works as well (meals should get smaller in the evening). For milder cases, simply avoiding big meals- and especially avoiding big meals that are followed by exercise (a classical bad move would be a wedding banquet followed by energetic dancing)- is sufficient.

Saturday, November 23, 2013

Shoulder pain and reduced range of motion is more common in diabetics

I only found one decent study looking at age and length of diabetes in people with frozen shoulder. It claimed the risk did not go up for type 1 diabetics until they had been diagnosed 20 years.

None of the studies I found on type 1 or type 2 diabetes on frozen shoulder showed a strong statistical link of gHbA1c and frozen shoulder; some found a weak link.

When comparing the risk to nondiabetics, the studies I found had a range of risk of 33% increase compared to nondiabetics (a large large Chinese one year study where a large group of shoulder complaints was lumped together), up to 860% (a small Scottish study that looked specifically at confirmed frozen shoulder).

Friday, November 22, 2013

Mitochondrial Mutations Can Cause Diabetes

Most of the cells in our bodies have two sets of DNA. One set is our genes as we usually think of them- 23 pairs of chromosomes that determine the proteins that our bodies make. These genes come equally from our mothers and our fathers, although each of us has some new mutations- changes in the genes- of our own.

But the mitochondria within each cell have their own set of DNA. This set comes from our mothers, with no contribution at all from our fathers. But like our other DNA, this DNA sometimes has mutations, and mutations may be passed down maternally.

Some of the mutations in mitochondrial DNA cause diabetes. Diabetes caused by the most common mitochondrial mutation tends to be caused by gradual deterioration of beta cells and therefore a gradual inability to make insulin. It is not autoimmune, and it is not usually accompanied by insulin resistance. The diabetes is most often diagnosed in teens, young adults, or middle aged adults; less often in the elderly and rarely in young children. Of people with the mutation, half are diagnosed with diabetes by the age of 38, and 80% are diagnosed by the age of 60. 

Roughly 1% of all cases of diabetes are believed to be caused by mitochondrial mutations. Because hearing loss is often caused by the same mitochondrial mutations that cause diabetes, hearing loss that comes before a diabetes diagnosis, hearing loss and diabetes in a young adult is a good reason to suspect mitochondrial diabetes. Another good reason to suspect mitochondrial diabetes is a family in which there is a mitochondrial pattern of inheritance- all or most of the siblings born to the same mother developing diabetes and/or hearing loss, the mother and her siblings, the maternal grandmother and her siblings. Macular disease in the eye is usually mild enough that it doesn't cause problems, but if it is there at all, it should also raise a red flag.

Mitochondrial mutation related diabetes is associated with a lot of health issues that a person might want to know he was at risk for; it's also NOT associated with some of the health issues associated with type 2 diabetes (ie hypertension) or type 1 diabetes (autoimmune thyroid disease). 

Thursday, November 21, 2013

Some Things About Insulin Glargine

Lantus is the brand name of the insulin glargine sold by Sanofi worldwide. Profits are shared with Novo Nordisk because of a lawsuit Novo Nordisk filed some years ago saying that the patent belongs to them.
In India, insulin glargine is also available under the name Basalog, sold by Biocon.
In India, the price of insulin glargine has therefore dropped in the last few years, but in the US the price increased about 10% over the last year (the company's reasoning seems to be that they can get away with it).

Sanofi advertises Lantus as lasting 24 hours for a flat basal profile. This is not much backed up by their own data, which shows that:
-in the person that it lasted the least amount of time for, there was no glargine detectable in the person's blood 10 hours after injection
-they did not measure levels of glargine in the blood MORE than 24 hours after injection, only up to 24 hours. In most people, there was still glargine active at 24 hours. The half life of Lantus injected subcutaneously in lab animals is about 30 hours and there is some evidence that it is about that length in humans as well. In this study where glargine was made radioactive and injected into type 2 diabetics, the time it took for half the glargine to leave the body was 26.3 hours.

Lantus does not reduce A1cs in any studies by any statistically significant amount as compared to NPH or Levemir. Its strongest point is that it reduces the risk of hypoglycemia at night compared to NPH- and that it does pretty dramatically. It also lends itself to allowing a basal bolus system of injection when used with a fast acting insulin.

Lantus is currently the most prescribed insulin in the United States. Personally, I don't think it deserves the honor, given that most of the prescriptions are in people with type 2 diabetes who do not use any other insulin and would probably get as good or better control with once daily injections of NPH.

One large and well publicized study of Lantus users showed a higher risk of breast cancer in Lantus users as compared to users of other insulins (the risk was only in people using Lantus without another insulin). No other study has found the same results.

Glargine insulin is made by genetically modified escherichia coli bacteria for Sanofi, and by a genetically modified pichia pastoris yeast for Biocon.

Wednesday, November 20, 2013

Cancer Risk in Diabetics

My mother has a theory about why genes that pose higher type 1 diabetes risk are as common as they are. She thinks they lower the risk of cancer. This makes a little bit of theoretical sense, because a strong immune is more likely to kill off cancer cells. Does it hold up to the evidence?
I am not aware of a study of cancer rates in relatives of diabetics, and having diabetes doesn't tend to extend people's lives. The evidence does suggest, however, that type 1 diabetics are less likely to develop and die from cancers.

Type 2 diabetes (or maybe the insulin resistance) almost certainly is a cancer risk:
Japanese study on cancer death risk and impaired fasting glucose
Meta Analysis of studies with a total of over a million men concluding that diabetes is a risk factor for prostate cancer
Strangely enough, right after that I found this article about the reasons diabetes is protective against prostate cancer- go figure.
Australian study concludes higher breast cancer risk for people with type 2 diabetes

But type 1 diabetes, if anything, is protective.

Although type 1 is a risk factor for a handful of specific cancers.
Acute Lymphoblastic Leukemia is often diagnosed within a year of type 1 diabetes. Although I'll admit to wondering if the diabetes in these kids is misdiagnosed as being type 1.

Most studies on cancer rates and diabetes has shown that type 1 diabetics have about 95% of the cancer risk of non-diabetics. That is, we are 5% less likely to develop cancer than other people our age, which is not a huge difference, but which is considerably more interesting when you figure that type 2 diabetics have a cancer risk that is higher than that of age matched non-diabetics. And, this is also despite the fact that type 1 diabetics are at increased risk of a few specific cancers (stomach cancer because of celiac, pancreatic cancer because things that don't work right are more likely to become cancerous, ovarian cancer probably because of the increased PCOS, leukemia probably because its a white blood cell thing).

Tuesday, November 19, 2013


Fact: When most people eat, their stomachs (okay actually the ileum and jejunum) release incretins that cause their pancreata to make more insulin, even before blood sugar rises. That's part of how the body preboluses.
Some of the medications used to treat type 2 diabetes are incretins (Byetta and Victoza) or imitations of incretins, and substances that increase the levels of incretins in the body (the medications ending in -gliptin).

Some people with type 2 diabetes do not make normal levels  of incretins in response to eating; that is a fairly common contributor to the diabetes itself. 
Some people with type 2 diabetes as well as some forms of MODY are related to having beta cells that don't respond to incretins- if the beta cells don't respond to incretins, it's sort of like having a body that doesn't prebolus for meals, and after meal blood sugar levels are likely to be high, even though fasting levels are probably normal.

Partly because of incretins (and partly because of the speed of digestion), in non-diabetics eating sugar has a smaller effect on blood sugar levels than being injected with the same amount of sugar. In some type 2 diabetics (and in most type 1 diabetics), the effect on blood sugar level from eating sugar is the same as being injected with sugar.

However, it's pretty hard to tell whose diabetes is caused by less production of incretin, because diabetes also causes a lowered production of incretins- when people have other forms of diabetes, their bodies also make less incretins.

Monday, November 18, 2013

The University of Toronto Patented Insulin

The Canadian government issued a patent for insulin to the University of Toronto on September 18th, 1923. The inventors are listed as Frederick G. Banting and Charles Herbert Best. UT patented it in both the United States and Canada (it is Canadian patent # 234336, and licensed the right to make insulin (Eli Lilly was the first company they gave that legal right to).

Some excerpts from the Canadian patent (I recommend that if this makes you at all curious, you go read the whole thing, and I'd also like to point out that Canada had a general ban on the consumption of alcohol at this time). I have very slightly rephrased the wording to  in make it more readable, and I have omitted a fair number of passages that struck me as too technical or repetitive. It was interesting to me that the authors do not seem clear on whether or not insulin is a hormone, as they always call it a "secretion or hormone". Also, it is never once called insulin in the patent.
Previous investigators suggested that the ductless portions of such glands as the mammalian pancreas and the pancreas of cartilaginous fishes, known as the isles of Langerhans... contains a hormone capable of alleviating diabetic symptoms in patients and in laboratory animals; and others conducted experiments in which diabetic patients and diabetic laboratory animals were given extracts containing this secretion or hormone...
We concluded that the presence of toxic substances in the extract [used by past experimenters] caused local irritation followed by general reactions unrelated to physiological and therapeutic effects of the hormone.... We deemed it advisable to prepare the hormone in practically pure form.
 This was done by extracting the internal secretion or hormone from the fresh pancreas of mammalia, or, from the fresh pancreas of cartilaginous fishes, or from fresh related glands of boney[sic] fishes, with a solvent capable of preserving the activity of the internal secretion or hormone and then separating it practically free from injurious substances including inert associated gland tissue, proteins, proteolytic enzymes, salts, and lipids. The following are steps we employed
(1) Separation of the hormone from the fresh with solvents such as ethyl alcohol, methyl alcohol, methylated spirits, and acetone....
 (2) Removal of the major part of the proteins... heating to a suitable temperature [works]
  (3) Concentration by... evaporation
(4) Removal of the lipoids after concentration....
(5) Removal of the salts and a large portion of the remaining impurities by precipitation with alcohol.
(6) Precipitation of the internal secretion or hormone ... by a higher percentage of alcohol
 (7) Dissolving the precipitate in freshly distilled water , removing the admixed alcohol, concentrating by vacuum distillation, followed by sterilization of the resulting aqueous solution.
A buffer solution of 1/2 cc of 4% HaHCO3 was added for every five liters of filtrate before distillation was commenced to keep the hydrogen ion concentration within the p H range of 4 to

We obtained... an extract for administration to human patients which had the physiological and therapeutic characteristics of removing the cardinal objective symptoms of diabetes mellitus in patients and reducing the percentage of blood sugar in laboratory animals and which has a distinct value in the treatment of diabetes mellitus, and lowering blood sugar, decreasing the urinary sugar, checking acidosis, and raising the carbohydrate tolerance of a diabetic individual to whom it is suitably administered.  
What we claim as new is...
1. A substance prepared from fresh pancreatic or related glands containing in concentrated form the extractive from the ductless portion of the glands sufficiently free from injurious substances for repeated administration and having the physiological characteristics of causing a reduction of blood sugar useful for the treatment of diabetes mellitus. 
5. A method for obtaining a potent substance from the ductless portion of pancreatic or related glands in concentrated form and practically free from impurities... which consists of extracting said substance  from a fresh gland with a solvent capable of preserving the activity of the substance, precipitating said substance from the solution practically free from injurious substances, and making a sterile aqueous solution of said substance.

Sunday, November 17, 2013

Vegetarians are Less Likely to Develop Type 2 Diabetes

One of the things I find very reassuring about studies on vegans and vegetarians is that compared to meat eaters, they develop diabetes a whole lot less often. This was recently shown again by the Adventist study (which is a long term study following 7th day adventists on the theory that they are somewhat similar to each other on the non-measured lifestyle things) where the vegans had 40% the rate of adult onset diabetes (they actually didn't differentiate by type).
Given that it's extremely difficult to eat low carb and vegan at the same time (not actually impossible) this is an interesting observation. Meats don't have carbs. People who eat meat therefore probably eat low-er carb, right?

Of course, the vegans had considerably lower BMIs- they do in every study of veg*ns compared to meat eaters (and I have looked) and I figure that probably accounts for most of the difference. I suspect a lot of the difference, maybe all of it, would disappear if you controlled for BMI. On the other hand, if a high carb diet CAUSED diabetes, then I really really don't think that the lowered BMIs would be enough to reduce the rate of diabetes so dramatically in high carb eating people. So, simply put, I DO NOT think a high carb diet causes type 2 diabetes.
That is, of course, a separate issue from whether or not a high carb diet is healthy for people who already are diabetic, and there I don't think the evidence is nearly as clear. Because there is a lot of evidence that low carb diets do not have bad health effects and do lower A1cs at least in most people who are newly diagnosed with type 2 diabetes. There's less evidence later on in the course of diabetes or in type 1 diabetes, but frankly, it does seem to be one way of making blood sugar more manageable for some people.
That said, vegetarians with type 2 diabetes tend to have better blood sugar control than the average type 2 diabetic (maybe this has to do with already being good with dietary restriction? Don't know).

It is puzzling though, that both high carb and low carb diets seem to improve diabetes control as compared to a "standard" diet, and I think the answer is hinted at by the fact that nondiabetics who go low carb even for a few days start going higher on oral glucose tolerance tests. My hypothesis is that the body can adapt to a high carb consistent diet. It can do it very well. It can also adapt to a high fat, low carb diet. It does that very well. I suspect the problem with the standard diet is that the body doesn't do as well going back and forth from high fat low carb to low fat high carb.

Friday, November 15, 2013

Nursing Aides

Strange but true: One of the 21 skills that nurse's aides are required to have to get certified in the state of Illinois is cutting fingernails. However, it is illegal for a nurse's aide who is working as a nurse's aide at the time to cut the fingernails of a diabetic.

Thursday, November 14, 2013

Fifteen people with childhood onset type 1 diabetes who lived to be 80 or more

This is copied verbatim from an answer I gave on yahooanswers in response to the question, Can I live to be 80 if I have type 1 diabetes.

Insulin has only been available for injection since 1922 (that's 91 years). So far, about sixty people are known to have lived 75 or more years on insulin (I don't know what age you developed diabetes or were diagnosed, so I don't know how long you need to live with diabetes in order to live to age 80, but your odds are best if either you were diagnosed before puberty, or else the older the better).

The vast majority of people with type 1 diabetes, even since insulin has been around, have died young. However, it's not hard to find exceptions.

For instance, Barbara Wagler was diagnosed at the age of 12, in 1936, and she received a Joslin 75 year medal at the age of 88.

Robert Bates, who received the 75 year medal at the age of 80 (he was diagnosed at the age of 5) passed away this year at the age of 86.

Spencer Wallace was diagnosed at the age of 8 and is now alive at the age of 90.

Dorothy Jane Williams of Sarasota, Florida was diagnosed at the age of 11 and is now 87 years old.

Frank Newby of Johnson City, Tennessee was diagnosed at the age of 2, and is now 81 years old.

Patricia Youell was diagnosed at the age of 7 years old, and is 85 or 86.

Barbara Kimbrough was diagnosed at the age of 9 years old, and died earlier this year at the age of 90.

Jack Little of Australia was diagnosed at the age of 15 and he made it to at least 85:

Kathryn Ham was diagnosed at the age of 8 and is now 84.

Bob Dornhecker of Dallas was diagnosed at 11, and is now 86 years old.

Edward Danielson was diagnosed at the age of 10, and died last year at the age of 91.

Sheila Thorn, who was diagnosed as a baby, died at the age of 81:

John Hegan, who is 81 now, was diagnosed at the age of 3.

Gerald and Robert Cleveland, who are brothers, were diagnosed at the ages of 16 and 3, and were both alive at ages 92 and 88- not sure if they still are.

Wednesday, November 13, 2013

Diabetics Have Children, but not as many of them

Various large population studies in Europe have shown that people with type 1 diabetes have, at least historically, gone on to have fewer children than non-diabetics. As some adults with type 1 diabetes have told me, some of the reasons include their own poor health (feeling unable to deal with pregnancy and/or uncertain about being around and able to raise a child), doctor recommendations, fear of passing on diabetes (or causing birth defects to the child if the mother is diabetic), and infertility. 

Nonetheless, many people with type 1 diabetes have gone and had children.

The Finnish study of type 1 diabetics' says they have between 2/3 and 3/4ths as many children as non-diabetics, and that the difference is smaller for younger folks. 

This smallish German study showed less than half as many children being born to the type 1 diabetic parents. 

This study on pregnant women showed that the diabetic women (type 1 and type 2) had tried to get pregnant for a longer time period before getting pregnant as compared to the nondiabetic women:

This Swedish study found that type 1 diabetic women had 4/5ths as many children as non-diabetic women BUT that if they looked only at women diagnosed after 1985 who did not have complications (and how many such women were there, I want to know?), then there was no reduction in number of children.

Tuesday, November 12, 2013

Protein, Fat, and Carbohydrates ALL Raise Blood Sugar Levels

Over the last twenty years, the carb counting method of managing diabetes- where insulin is matched to carbohydrate intake- has gained in popularity. This has, I think, added to the perception that other sources of calories do not affect blood sugar, or affect them very little. But most studies have shown that proteins and fats raise blood sugar, at least in type 1 diabetics.

All of the study links I found that were prospective studies on this topic enrolled less than a hundred people. I am nonetheless willing to go with this as fact because most (all but one) of the studies did support this fact, the ADA agrees, and it fits my personal experience.

Monday, November 11, 2013

Type 2 Diabetes Is Less Significant in the Very Old

Type 2 diabetes is the most common diabetes diagnosis. It is sometimes diagnosed in childhood (although not in very young children) and the risk of being diagnosed goes up with age until, depending on the country we're talking about, somewhere between the ages of 70 and 90, although average fasting blood sugar does seem to increase even after the age of 70.

Type 2 diabetes is a major risk factor for dying younger when it's diagnosed in children, young adults, and the middle aged, but in the elderly its a much smaller risk factor, and numerous studies of very old people have found that it's not a risk factor at all until you start adjusting for various things, particularly BMI (diabetics tend to have higher BMIs, and in people above about 85, the higher the BMI, the lower the risk of death). In fact, in a number of studies on people above the age of 85, being diabetic was associated with a lower risk of death in a short period (2-5 years).

To me, the Georgia Centennarian Study (Georgia as in the southern state, not the country) is particularly interesting. Among the 244 people they studied, age 98 to 108, 21 of the 244 had been diagnosed with diabetes and another 8 people had high A1cs or random blood sugar and were diagnosed as part of the study. Although only 7 of them were diagnosed before the age of 80, 1 had been diagnosed with diabetes more than 50 years earlier.
After 1 year, 20% of the diabetics died and 27% of the non-diabetics died.  The diabetics did not have a higher risk of vision loss or a higher risk of diagnosed neuropathy.!po=53.4483

Saturday, November 09, 2013

Male Pattern Baldness Doubles Heart Disease Risk in Diabetics

Four times, the hair on my head has thinned and grown much more slowly than usual. The first time, shortly before I was diagnosed with diabetes, was probably malnutrition; after I was back on insulin it picked back up. The second time, shortly before I was diagnosed with gallstones and subacute thyroiditis, was probably malnutrition again. The third time was unexplained. All three times, all of the hair on my body grew more slowly.

About a month ago, after cutting my hair pretty short, the hair on the top of my head didn't seem to be growing in normally even though the hairs on the sides of my head were. This time, I'm thinking it looks like male pattern baldness (my hairline has seemed to be maybe receding for the last couple of years).

Just for the heck of it, I thought I'd google around for health associations with male pattern baldness. In men in general, it is associated, apparently, with a higher rate of death from heart disease- about a 30% increase in risk. Among men with diabetes though, that risk is even more dramatically increased.

In this study of men with all types of diabetes aged 30-95 in Taiwan, the ones with male pattern baldness matched for age and sex had more than double the risk per year of dying from heart disease. This might be partly because of things causing balding also affecting insulin sensitivity.. or not.
This article on the risk of balding and risk of heart disease in men in general has a lot of theories on pages 20-21 :

On Friday I left my Regular at some friends' place and didn't go back for it until today. I tried bolusing with intramuscular Lantus and it seemed to work moderately well... my blood sugar was worse than usual but not so much that you could pick it out as my worst day or anything, and I did have two large meals.

Thursday, November 07, 2013

The Complication Most Worth Testing For ...

is retinopathy (in diabetics of pubertal age and beyond).

There's little to no evidence that screening for kidney disease or neuropathy changes what happens with that kidney disease or neuropathy (although theoretically, once you know you have neuropathy, you should be taking more precautions). The things you can do for early, asymptomatic, kidney disease can be done without any knowledge that you have kidney disease (mostly, manage blood pressure), and there's nothing to do for asymptomatic neuropathy.

However, asymptomatic bleeding in the eyes can be treated.

The study that the ADA quotes in its 2013 statement about the worth of treating diabetic retinopathy with panretinal photocoagulation, saying that it reduces the risk of blindness in that eye from about 16% to about 6%, is an interesting study.
The Diabetic Retinopathy Study began in 1971, and it enrolled 1742 adults under the age of 70 with severe nonproliferative retinopathy and/or proliferative retinopathy, in both eyes. They randomly had one eye get treated, and one eye not get treated, which is a pretty cool control group IMHO.
The authors of this study believe that treating diabetic retinopathy should be best done neither early nor late; that it should be treated before vision loss occurs, but not in early stages of retinopathy, because of the risk of the treatments.
The value of knowing that a person has mild retinopathy is mostly only in being able to test for more severe retinopathy more often; retinopathy rarely develops very quickly.

There is some controversy/ discussion in medical journals about how often screening for retinopathy should be done, particularly in diabetics without retinopathy. Although the current clinic recommendation most often given is once per year, I think the evidence is pretty strong that screening every other year is sufficient... and I saw an eye doctor recently who thinks type 1 diabetics should be screened twice per year!

A metastudy:
Study listing with links:

Wednesday, November 06, 2013

Pets Responding to Blood Sugar

About 2/3rds of diabetic dog owners say that their dog has ever alerted them to a low blood sugar, and about 1/3 of diabetic dog owners say that their dog has alerted them at least a dozen times.

Last year I tried to look up data on how many blood sugar alert dogs are working in the United States (or anywhere) and I was unable to find that data. I was not even able to find data on exactly how many service dogs of any sort work in the United States (although I've seen a guesstimate of 20,000). I have met one person who had a service dog prior to developing diabetes whose dog alerts her to low blood sugar, and like most of us who read online, I have read of many stories of both pets (dogs and cats) and trained dogs alerting to high and low blood sugar. I have also read about dogs that were sold as having been trained, but which did not alert to high or low blood sugar.

It is entirely legal to train your own service dog, although not easy, and your dog does have to be well trained in order to be allowed to go everywhere with you (states differ in laws about service dogs in training).
Currently there are numerous programs that offer dogs to diabetics for blood sugar alerts; a google search for diabetes alert dogs brings up 12 different companies offering dogs to diabetics in the first thirty hits.


Tuesday, November 05, 2013


In the 2011 November facts, I posted about Wolfram's Syndrome under the heading of different ways in which diabetics lose their sight. The gene that causes Wolfram Syndrome was the first gene discovered that pretty much always causes diabetes (instead of just increasing the risk). But Wolfram Syndrome is really rare; its estimated that about 1 in 400,000 people have it. The form of diabetes that Wolfram Syndrome causes used to be classified as type 1b- it's a diabetes caused by inability to make insulin as the islet cells atrophy- but its not autoimmune. The diabetes is usually diagnosed before the age of 10 and is usually insulin dependent.

Since the gene was discovered in 1998, it has been discovered that other mutations on the same gene, mutations that don't cause Wolfram Syndrome, dramatically increase the risk that a person will be diagnosed with diabetes, usually type 2 diabetes. Although the study of this gene's contribution to diabetes development is still in early days, it may contribute to as many as about 15% of cases of diabetes diagnosed as type 2.

I say diabetes diagnosed as type 2, rather than type 2 diabetes, because type 2 diabetes is currently defined as diabetes caused by an insulin secretory defect on a background of insulin resistance... and it's very plausible that this gene would cause diabetes in the absence of insulin resistance.

Monday, November 04, 2013

Some risk factors for type 1 diabetes already present at birth:

- Family history of type 1 diabetes. Children with a parent or sibling with type 1 diabetes have a risk that's at least 4 times the risk of those without an immediate family member with diabetes, and depending on which family member and ethnic group, the risk may be more than 20 times as high.

- Genetic markers for type 1 diabetes - those with protective genes will not get type 1 diabetes. 

- Being born in a high risk country. The children of immigrants to high risk countries, who are born in those countries, have a higher type 1 diabetes risk than their foreign born siblings.

- Having a high birthweight. Babies that weigh more than 4500 grams (almost 10 pounds) at birth have about double the risk of developing childhood onset type 1 diabetes as compared to babies born weighing 2000 grams.

- Being firstborn. People without older siblings are about 10% more likely to develop type 1 diabetes (or at least, childhood onset type 1 diabetes).

Sunday, November 03, 2013

Prairie Dogs' Blood Sugar Levels

Apparently healthy prairie dogs of different types have normal blood sugar ranges that don't overlap. One study on black tailed prairie dogs found a blood sugar range of about 28 to 84 while another study on wild prairie dogs found a range of 138 to 510!
That might have to do with the fact that prairie dogs hibernate.
Other hibernating mammals (such as brown bears) show differences in blood sugar during different seasons, and even humans with diabetes tend to have higher blood sugars in the winter months.

Saturday, November 02, 2013

The Insulin Gene

Insulin is a protein and like pretty much all proteins, there's a gene for that.
It is usually called the INS gene or insulin gene, but has also been called the MODY 10 gene, the proinsulin gene, and a few other names, and it's located on the short arm of chromosome 11.
Mutations in the INS gene are a rare cause of diabetes; mutations in both copies of the gene lead to neonatal diabetes, and mutations in just one copy leads to MODY 10. Some mutations lead to an even rarer condition where the body makes too much insulin.

The gene actually does not code for insulin itself, but for proinsulin, which is insulin attached to c-peptide.

At 1,430 base pairs, the insulin gene is among the shortest human genes, which is probably why mutations in this gene are uncommon.

Friday, November 01, 2013

Diabetes Awareness Month 2013- Just Who is On The Guideline Committees?

As you all have probably noticed, my blog was offline for about ten months. I have removed a large number of posts due to privacy concerns but the rest of the blog is back. It may look a little disjointed if you read through the archives.

Today kicks off November, aka Diabetes Awareness Month. Many folks for this month get "facts" each day from diabetes organizations that are aimed at the general public about how big and scary and awful diabetes is without much real data or explanation. Sort of in reaction, I started posting a fact per day that is aimed at people who do have some familiarity with diabetes and that I do my very best to make sure really is backed up by some data.

I have taken down many of the posts that said much about me so this much you may want to know: I am a young adult diagnosed with type 1 diabetes at the age of 17. I have a small amount of medical training, but almost all of what I will post here about diabetes is self taught. I currently use a jet injector and inject Novolin R and an insulin pen for injecting Lantus. I also use a Dexcom G4. I live in the United States and I have private medical insurance.

So, welcome back if you're an old reader and also welcome if you're a new reader.

Today's fact is not directly about diabetes, but about guidelines for treatment. In the United States, you have probably seen A1c or other diabetes management recommendations from the American Diabetes Association and/or the American Association of Clinical Endocrinologists. 
You should know that 19 out of 21 authors of the AACE's 2012 diabetes guidelines had received payments from one or more company making diabetes drugs, and that 15 out of 16 of the authors of the ADA's 2012 diabetes guidelines had received money from one or more companies making diabetes drugs.

Of the 19 authors of the 2013 Guidelines, 10 report monies received or affiliations from the past year, but 15 have received money from a pharmaceutical company at some point, and that does not include two working for the ADA.

Nathaniel Clark received two thousand six hundred seventy five dollars for one speaking engagement for Eli Lilly in 2011.
Cyrus Desouza reports getting monies from grants from Novo Nordisk.
Martha Funnell has recieved monies for speaking and consulting for Eli Lilly and Merck in 2009, 2010, and 2011 to the tune of over $25,000 on only five days.
Allison Goldfine reports monies from public grant research funds as well as numerous pharmaceutical companies.
Richard Grant works for Kaiser Permanante and I'm not sure if he's the same Richard Grant who got $1,200 from J&J for a speaking engagement. He has also received monies from governmental grants.
James Lenhard has made more than $28,000 for speaking for AstraZeneca and Merck, and he also has received governmental monies for research.
Jennifer Marks reports monies for research from Eli Lilly as well as others.
Anthony McCall has gotten monies from Pfizer, Aventis, and Eli Lilly and the government.
Janis McWilliams was paid a little over two grand from Eli Lilly for "other".
Harsha Rao has received a little under a thousand dollars for "meals" from Pfizer.
Andrew Rhinehat reports only his research grants, but additionally has received almost 30 grand from Eli Lilly and Forest, none of it for research.
Henry Rodriguez received more than $50,000 from Eli Lilly in a two year period for non-research items, and he reports also money from the government, trusts, Novartis, Squibb, and Merck.
Debra Simmons reports research monies from Novo Nordisk.
Patricia Urbanski received three thousand dollars from Eli Lilly for consulting.
Carol Wysham, chair, received more than a quarter million dollars from Eli Lilly, although she reports only research money from charitable and governmental grants.