Adjusting Insulin Dosage

A lot of diabetics leave insulin dosage to their doctors, and don't see their doctors very often. Others adjust their dosage themselves, cutting out the middleman.

Most drugs on the market, including the insulin Karl uses, have "patient information" and "physician information." We always read both.

As an engineer, I've been amazed by the primitive nature of the dose-adjustment information given to doctors. I realize that doctors are not as well-trained in basic numerical methods as engineers, but still...

By the way, one of the larger barriers to doing a good job has been whittled away recently. The newer Bayer Contour blood glucose meters, the Contour Next series, lets you enter the number of carbs and the insulin dose on the meter, which now has the full set of information (time, blood glucose, carbs, insulin) to let you analyze the results without typing pages of information from a hand-written log.

Anyway, like most Type 1 diabetics, Karl is on a long-acting insulin (Lantus) and a mealtime insulin (Novolog).

The numerical analysis challenge goes like this:

• You want to give mealtime (bolus) insulin that's just right to handle the carbohydrates in the meal. This uses a carb factor, the number of grams of carbohydrate per unit of insulin. Get this right, and after the meal, blood sugar will be the same as before.
• In addition, if blood sugar levels weren't perfect just prior to the meal, you want to adjust the insulin up or down to correct for this. This is the correction factor, the number of points the blood glucose value changes per unit of insulin, in the absence of food.
• For long-acting (basal) insulin, you want a value that, in the absence of food, keeps the blood glucose levels steady all day.

Fair enough, but all three effects are going on all the time, so getting them all right is a bit of a trick.

I have not yet seen any publication where anyone even tries to do the number-crunching properly, or even to give useful rules of thumb, except for the long-acting insulin, where the rules are variants of, "Pay attention only to fasting (pre-breakfast) glucose levels, and adjust the dose of basal insulin, perhaps by one unit a day, until it yields numbers close to, but slightly less than 100, on average." 

Sure, they have rules of thumb for starting new patients on insulin, but that's about it.

Karl needed an increase in insulin dosages recently, as his numbers were creeping upwards, and what I did was as follows:

• Adjusted the long-acting Lantus dosage upwards to get his morning numbers below 100, knowing that some of the problem was really that he also needed more mealtime insulin, and that I'd likely end up adjusting them downwards in the end, at least a little.
• Looked at meals where Karl's pre-meal blood glucose levels were okay, so there was no correction, and thus the correction factor was not part of the picture. The difference between before-meal and after-meal blood glucose revealed the degree by which the carb factor needed adjustment. It turned out that he needed 9 carbs per unit instead of the previous 10.
• Looked at meals where Karl's pre-meal blood glucose was high, requiring a significant correction, and what the post-meal blood glucose was. This gives a hint at what the correction factor should be. This required a significant correction, from 85 points per unit to 45 points per unit. I also double-checked this with a rule-of-thumb table about what the corrective dose is likely to be, based on daily insulin consumption, though of course I value actual results with Karl specifically to anyone's table about patient populations in general. Note that if the carb factor and correction factor are both out of whack, it's hard to adjust them both at the same time with any precision, but you can do them sequentially.
• As predicted, the adjustments all called for an increase in insulin (not a surprise, since his blood glucose numbers had drifted upwards), and once the other two numbers were adjusted, the long-acting Lantus dosage proved a bit high, and we backed it off by one unit per day so far, and may back it off by one more.

This seemed to work pretty well. I probably ought to do regression analysis for more precision, and after we've switched over to the new Bayer Contour Next meters, I'm likely to take a stab at it.

The Bayer Glucofacts software that comes with the meter is pretty primitive, as I've said, but the better the control, the lower the standard deviation between readings, and the closer the average reading is to the target value. 

Test Results

Karl developed an aversion to blood work, but on the second try and quite a bit of help from the folks at Philomath Family Medicine, we got it done.

The trick was to get a prescription for Lidocaine ointment and apply this topical anesthetic an hour before the appointment, telling Karl that this would help a lot, and doing the blood draw in a different room from usual. This process was more to his liking, which is just as well, because it took the nurse quite some time to find the vein!

Blood Sugar Well-Controlled

Two years after being diagnosed with Type 1 diabetes, his average blood sugar level, as measured by the Hemoglobin A1c test, is at 5.3% -- within the normal range! Yay, Karl! This presumably means that he is no more likely to have diabetes-related health complications than a non-diabetic. Whew!

How do we do this? Karl's rigid dietary preferences help, because we know how many grams of carbs are present in everything he eats. And the sensible use of intensive insulin therapy, with long-lasting basal insulin and fast-acting mealtime insulin, is essential.

Adjusting insulin dosage. One thing we do that many people don't do is to adjust Karl's dosage based on experience. We don't wait for a doctor's okay -- or even ask, for that matter.  Some people wait until the next time they see their doctor to adjust anything, even if their next appointment isn't for a year and their numbers are way out of line! That's a bad idea.

For his basal insulin -- we use Lantus -- you can find the "titration" recommendation online. Basically, we look at the morning blood-sugar levels and adjust the dose until they're in the right range. This can be repeated if they wander out of the range again (looking at a few weeks as an average). Karl started out with about 7 units of Lantus a day and is getting 18 now. We split this into a morning injection and an evening injection.

Similarly, for mealtime (bolus) insulin, if the blood sugar numbers are consistently high or low after meals, we're clearly giving the wrong amounts of insulin. Over the past two years, the right amount has drifted from 20 grams of carb per unit of insulin to 10.

Being willing to adjust the dosage according to the data is probably a big reason why Karl's A1c numbers are so good.

Pancreas Hanging in There

Another important test, C-Peptide, measures how much insulin Karl's pancreas is still making. The "acceptable" range is 0.9-6.9 ng/ml, and Karl came in at 0.7. It was actually in the "acceptable" range last time we tested it, two years ago, which shows how blood tests don't tell the whole story, since being hospitalized with extreme levels of blood sugar isn't what I'd call acceptable!

People will tell you that Type 1 diabetics don't have any pancreatic function at all, and their C-Peptide tests will come in at zer0. That's nonsense. Pancreatic function in Type 1 diabetics is impaired, and typically declines, often to zero -- but writing the pancreas off is pessimistic and stupid.

To help preserve the pancreas, we do a few simple things:

• Maintain excellent blood-sugar control. This puts less of a load on the pancreas, which isn't literally killing itself by producing all the insulin it possibly can, 24/7, because the injected insulin is removing much of that load.
• Supplement with niacinamide and vitamin E. A few studies show benefit from this.
• Maintain a low-ish carb diet. The more carbs you eat, the poorer you blood-sugar control and the more your blood sugar spikes. Carbs reach the bloodstream faster than injected insulin, but the fewer carbs you eat, the smaller the spike And an error of, say, 20% in estimating the carbs in a meal makes only half as much difference if you're eating only half as many carbs. Karl eats only about half as many carbs as are recommended for a typical diet.

Two Years of Type 1 Diabetes

Next month will be the second anniversary of Karl's illness, hospitalization, and diagnosis of Type 1 diabetes, and it's so far, so good. 

His insulin usage has gradually increased, and is now at 17 units of Lantus per day for a basal dose, plus one unit of Novolog for every ten grams of carbohydrates in his meals. I rough figures, this is about twice his original dose, meaning that his pancreas has been gradually producing less and less insulin.

We were told to expect a "honeymoon period" of relatively low insulin requirements and stable insulin requirements that ended suddenly, but this didn't happen. 

Possibly this is because we deliberately decided to see what we could do to to stretch things out. Among these things were:

• Use of a relatively low-carb diet (averaging around 60 grams of carbs per meal, which requires less insulin and thus allows better control of blood sugar.
• Use of the Novopen Junior insulin pen, which officially allows insulin to be delivered with a half-unit resolution, but which probably allows precision to one-quarter unit or better. (For example, if we want to give 5.25 units, we dial in 5.5 units, then back off to halfway between the 5.5 mark and the 5.0 mark.)
• Karl likes the same meals over and over, so we get really good at comparing the before-and-after blood sugar levels to find the right level of insulin.
• Giving Lantus morning and night, rather than at bedtime only, because its response curve doesn't stay flat for a full 24 hours for everybody.
• Use supplements: Insulow and evening primrose oil at every meal and daily supplements of mutlivitamin, vitamin B3, and vitamin E.
• Refusing to freak out.

Karl's very fussy about maintaining his insulin kit just so, and that means we're almost never caught short when we go out to eat.

On the downside, Karl has refused to let the nurse take blood samples the last couple of times, so we haven't done detailed bloodwork recently. And his blood glucose regulation is getting less good as his insulin requirements increase, since the pancreas is far better at this kind of thing than any kind of insulin therapy, however spiffy and modern, so the less his pancreas contributes, the wider the blood-sugar swings, though we're still doing pretty well.

What's next? Perhaps reducing Karl's carbohydrate intake some more. The fewer carbs he eats, the less need for insulin, the lower the load on his pancreas, and the narrower his blood-sugar swings will be. We've done pretty well at keeping a typical meal at 60 grams of carbs, but Dr. Bernstein's regimen recommends no more than eight grams of carbs per meal, and that would be a huge change! 

So far, Karl has had no hypoglycemic episodes and no infections, and we'd like to keep it that way.

Illness and Diabetes

Karl's recovering from a cold, his first real cold since being hospitalized over a year ago. Man, oh man, did his blood-sugar levels go up! Bedtime readings were above 200 three days in a row.

Illness tends to raise blood-sugar levels in diabetics, and if there's dehydration involved, it gets much worse. Karl doesn't like our tap water, so we made sure the supply of bottled water and sugar-free drinks was unusually ample, using the lure of yumminess to overcome any reluctance brought on by lethargy.

Corrective doses at mealtimes were clearly not enough to bring his blood sugar back to normal, so we started a two-pronged strategy of increasing his long-acting Lantus dosage from 8 units a day to 10, and also added an extra 0.5 units per meal in addition to the calculated value. Both corrections together have gotten him into much better territory.

As things normalize, we'll first abandon the extra 0.5 units per meal and then back the Lantus off to suit. If Karl were to suddenly snap back to normal without warning, we might have to give a couple of small snacks to keep his blood sugar up, but it seems far likelier to me that he'll drift rather than snap back to normal.

I have not reviewed all my diabetes books for corroboration, since it's not rocket science to conclude that "if blood sugars are way too high, give more insulin."

We are also giving plenty of Insulow (alpha lipoic acid), which in some ways mimics the effects of insulin and also tends to reverse some of the damage of high blood sugars. While I suppose it's possible to overdo this, since alpha lipoic acid will lower blood sugars, and this could cause trouble if you're not paying attention, we've never seen any evidence of this. Karl normally gets 200 mg per meal, for 600 mg per day, but in his case giving twice this much has little effect on blood sugars. Apparently it's different for some people.

Karl didn't ask for cold medication. Dr. Bernstein's book paints a scary picture of aspirin, ibuprofen, and similar drugs, which can apparently  crater blood sugars unpredictably, and Tylenol, which is hard on the liver if you're dehydrated. Karl's pretty content with sugar-free cough drops as the only treatment.

Dairy Queen for Low-Carb Meals?

Normally, you'd expect a place with as heavy an emphasis on sugar as Dairy Queen to be a non-starter for people like Karl who need to watch their carbohydrates, but they have three different low-carb ice cream bars: my favorite, the vanilla orange bar, with 12 grams of carbs, the fudge bar, with 7 grams, and the no-sugar-added Dilly bar, with 19. So if you're on a low-carb diet, you can have dessert at DQ.

And, oddly enough, they have a very good (and large) grilled chicken salad that's very low in carbs, too.

Not to mention the usual fast-food expedient of bunless burgers, which have very low carbs indeed.

Karl's unwavering preference is for chicken strips, which are higher-carb than the above-mentioned alternatives, and he has a weakness for french fries. Sigh. What this means is that I'm eating a lower-carb diet than he is, though he's diabetic and I'm just losing weight. 

But for most meals he eats fewer than 60 grams of carbs, and in our experience it's looks like Dr. Bernstein was right, and eating fewer carbs not only means a Type 1 diabetic needs less insulin, but the blood-sugar levels are more stable. 

And Karl has gone from being painfully thin at the time of his illness to putting on weight steadily, so it looks like low-ish carb diets have real promise for weight stabilization. They're not just for weight loss.

See you at DQ!

Basal Insulin: What's the right amount?

The sad thing about doctors is that they aren't engineers, and this is doubly true for lawyers. So when you read about medications, everything's written by people who (a) don't think in terms of getting results out of feedback control systems, though that's what the body's all about, and (b) do think in terms of, "If I say something blatantly wrong, I'll get sued, but if I'm mealy-mouthed, I'm safe." This means that the dosage adjustment (titration) instructions for Karl's long-acting Lantus insulin are vague and simplistic, even by the standards of the intelligent layman: http://www.lantus.com/hcp/titration.aspx. Thanks, health-care industry!

So let's talk about feedback control systems, which is a fancy term for something that everyone already knows a lot about. If you set the cruise control on your car, the car will run at the speed you set, increasing the throttle when you're going uphill, decreasing it when you're going downhill, etc. Going up a really steep hill, you might not have enough horsepower to maintain your speed, so you slow down: "flat out" is all you've got. On a downslope, you might go faster than the speed you selected. So cruise control keeps everything constant under a certain range of conditions, but outside this range, you get what you get.

Karl's pancreas is still making some insulin. The pancreas has a very cool feedback control loop which, like cruise control, puts insulin into Karl's system at the right rate to keep his blood-sugar levels constant, within a certain range. The problem is that his pancreas can't make enough insulin to keep his blood sugar under control under most conditions. His pancreas' accelerator is pressed all the way to the floor and he's still not at cruising speed.

If we get his dosage right, injected insulin acts like a strong tailwind, allowing Karl to get up to speed without making his pancreas run flat-out. To continue the auto analogy, this reduces wear and allows it to cool off, so it will last longer.

Not only that, but by allowing the engine run most of the time at, say, 50% power instead of 100% power, if we get an unexpected hill, we can get over some hills without slowing down, because we have reserve capacity. Turning the analogy back to diabetes, if Karl eats a larger portion than we had calculated, so he receives less insulin than he should, or sneaks a snack without any insulin at all, his body has reserves that will cover it without spiking his blood sugar. Sweet!

Because Karl's pancreas is producing some insulin on its own, he can receive different doses of long-acting insulin and arrive at the same average blood-sugar levels. If he gets a low-ish does, his pancreas will provide more insulin, and if he gets a high-ish does, it will provide less. So within this range, the results will be equivalent -- except to his pancreas!

Since the goal is to keep his pancreas from being overloaded, we're aiming for a dose on the high side of this range. So far, we've determined that a dose of Lantus anywhere between 6 and 8 units per day gives about the same results, and we'll probably test 9 units soon. We're adjusting no more often than once a week, since day-to-day variations are wide enough to obscure the results.

Holding Steady

Eight months after being diagnosed with diabetes, Karl's holding steady, and, if anything, his blood glucose numbers are getting better.

We partly attribute this to our policy of adjusting his dose of long-acting Lantus insulin from time to time, with a rule of thumb, "If two doses give about the same results, give him the higher one, to take some of the load off his pancreas." The less hard his pancreas has to work, the longer it lasts (our goal is "forever"), and the more it can take up the slack if Karl has more cabs than expected. He was down as low as six units, we're at seven now, and we'll likely inch it back up to eight.

Because Karl's autism means that he likes his routine, he's settled into both a dietary and a daily medical routine very well. Because he doesn't pine for treats or change his mind about what he wants for dinner based on what's going on around him, it's very easy to stick with things that work. He's even very good-natured about going into Dairy Queen and getting one or two of their low-carb ice cream bars (the only things in the store that don't have inconceivable levels of sugar).

Everything would be golden if my employer, Citrix, hadn't shifted from an excellent Blue Cross health-insurance plan to a ghastly Cigna one. I'm thinking I'm going to have to buy supplemental insurance to maintain a decent level of care.

Here's Karl's blood sugar graph from the last month:

Karl at the Prom

Karl had a good time at the Special Ed Prom, though as usual he wore his headphones to cut down on the loud music. The local paper used a photo of Karl, so he's famous again.

On the diabetes front, he's relentlessly sticking to his routine and to his accustomed diet (peanut butter and honey sandwiches on reduced-calorie bread for every meal, and chicken nuggets for every meal but breakfast). We'd like him to have more variety! This keeps his carbs down, though.

He's starting to look forward to Atkins shakes as a snack, and since these have practically no carbs, they're something he can have without insulin. Most of the other no-carb foods (cheese, hot dogs, meat, etc.) he doesn't like until they're made into something like pizza, which sorta defeats the purpose!

Are You Positive It Isn't a False Positive?

We got a high blood-sugar reading the other night, a hundred points higher than we expected. "What happened? What did we do wrong? Did we forget to give him his insulin? Did he have a huge snack we didn't know about?" What we did wrong was forget that Karl was building a gingerbread house and hadn't washed his hands before doing his blood-sugar testing. D'oh!

He gets so little in the way of sugary foods that we don't always remember that "sugar on his fingers will mess up his blood-sugar tests."

Precision: Matching Carbs to Insulin

How can we get more precise control over blood sugar? We've been experimenting with a way that seems obvious but I haven't seen mentioned anywhere: Match the carbs precisely to the insulin dose.

For example, suppose Karl is going to sit down to a lunch of 51 carbs, for which, at 15 carbs per unit of insulin, he's supposed to get 3.4 units of insulin. Well, we can't give him 3.4 units. We can give him 3.0 or 3.5, but not 3.4. 

Because giving too much insulin is worse than giving too little (since giving too much leads to low blood sugar and the troubles that it causes, from hunger to fuzzy-mindedness to loss of consciousness), we round down, and Karl would get 3.0 units, which is four-tenths of a unit short of what he needs. We'd expect his blood sugar to end up about 40 points too high because of this.

On the other hand, if he had 52.5 grams of carbs, that would work out to exactly 3.5 units, so we could be right on the money. All we need to do is find 1.5 grams of carbs in a Karl-friendly package. What we use are Ritz Bitz, which come out to about 1.5 grams each. Problem solved!

So now he's getting anywhere between zero and four Ritz Bitz with every meal. It seems to make a real difference when his carbs are just short of the next insulin increment.

If you're using an insulin pen but aren't using one that allows half-unit dosage, you should be! The NovoPen Junior has been working very well for us. If we had to give insulin in one-unit increments, the number of Ritz Bitz we'd give Karl per meal would go as high as nine! He'd like that, but we wouldn't.

Adjusting the Carb Ratio

Looking at Karl's blood-glucose logs yesterday, we discovered that Karl hadn't had low blood sugar once in two weeks, but he'd had high blood-sugar many times. Since insulin is what brings his blood sugar down to normal, being never low but often high means, "too little insulin."

We use what's called "advanced carb counting," meaning that we figure out the grams of carbohydrates in every meal and divide this number by the "carb ratio" to get the number of units of insulin. 

I'm told that Type 1 diabetics whose bodies no longer produce any insulin have a carb ratio of around 8, meaning they get one unit of insulin for every 8 grams of carbohydrate. For Karl, we've been using a carb ratio of 16, meaning that he gets only half a much insulin as the diabetic with no insulin production of his own. This implies that Karl's pancreas is working at half efficiency.

Frequent high readings means that 16 is the wrong number, so we started using 15. The two numbers don't sound very different, but on a typical meal this happens to give Karl an extra half-unit of insulin. We've only been doing this for 24 hours, so the results aren't definitive, but it did exactly what we wanted: his post-meal ("post-prandial") blood sugar was in the 120 range, and his pre-mail blood sugar was in the 80-100 range -- both very good numbers!

It may also turn out that Karl's poor blood-sugar control with higher-carb meals was a simple "not enough insulin" problem.

One thing I've found frustrating in the literature is the weak, hand-waving descriptions of how to adjust insulin dosage. Come on, guys, this is the Age of Science! Is it possible that the problem is that, if you're clear, you might be sued, but if you're incoherent, you'll be left alone? Because basic information is so often mumbled when it needs to be shouted from the rooftops, and this means I haven't found any good "How to adjust your insulin dose" links for this post.

What we do is look at the trends for the last two weeks or more, especially comparing the frequency of lows compared to highs. Lows are scarier than highs, which is reflected in the normal guidelines, where it's considered okay to have several times more highs than lows. If that's not what we're seeing, then our dosage or our technique are not right. 

If your glucose meter comes with adequate software, as the Bayer Contour USB meter does, then you want the averages to be where they should be, the standard deviations to be as small as possible, and the highs and lows to be under control, biased towards modest highs rather than modest lows.

The standard deviation tells you how tight your control is but not whether it's centered around the right value. Getting tighter control, for us, happens when we:

• Reduce the number of carbs eaten
• Calculate the carbohydrates in each meal accurately
• Use the Novopen Junior for its half-unit dosing (insulin pens that only give insulin in one-unit increments are too imprecise)
• Don't cheat in any way.

Having mealtimes at the same time every day doesn't seem to matter for Karl, and if the portion sizes vary, this doesn't matter either if we've figured out the right carb ratio, so long as he doesn't go crazy on the carbs.

Random diabetes tip: The Swiss Miss sugar-free cocoa is sorta thin and unsatisfying, but adding a tablespoon or two of cream into the mix straightens that right up with only 1-2 grams of added carbs. (Avoid their "No Sugar Added" cocoa, since it's full of sugar.)

Leftover Insulin Drops on Injection

Last night at dinner, I noticed a drop of insulin still on the needle after his Novolog injection, and dampness around the injection site. We keep the needle in until a count of eight after the injection, which is supposed to prevent such things.

This time, though, it clearly wasn't effective, and Karl's blood sugar shot up to over 200, meaning that a whole unit had been lost, at least.

We're using the Novopen Junior insulin pen and BD 4mm nano needles and haven't seen this problem before. The instructions for the nano needles say, "Don't pinch the injection site," but Karl wants us to (it's how we were taught to give injections), so we still do. Arguing with an autistic kid is hard work, so we do things Karl's way when it seems okay to do so.

I wonder if the pinch is actually counter-productive with short needles? I wonder if there's a protocol for dealing with a seemingly messed-up injection like this? Estimating the loss seems like a difficult task.

A one-unit corrective dose got him headed in the right direction, though of course these never have the instant effect you'd like.

Fun? With Advanced Carb Counting

The bad thing about diabetes management is that it isn't perfectly simple. The good thing is that, with glucose test strips, you can measure blood sugar any time and find out if things are working, so you never have to be in the dark if you don't want to be.

If you eat the same meals over and over, which is what Karl prefers to do anyway, you can nudge you insulin doses up and down until you hit your targets. The same meal at the same time of day should require the same insulin time after time. This trial-and-error approach means that the theory behind dose calculation doesn't have to be very precise, which is just as well, because it isn't.

We're using what's called "advanced carb counting," which we use like this:

• Add up all the "net carbs" in the meal
• Divide by Karl's "carb factor" (20 for breakfast and lunch, 16 for dinner)
• Adjust if his blood sugar is high or low
• Inject the indicated amount of insulin

What are Net Carbs?


Net carbs, or net carbohydrates, are "the carbs that matter to you." What these are depend on context. What kind of carbs don't matter?

• Fiber doesn't matter, because it's not digestible and doesn't effect blood sugar.
• Sugar alcohols, artificial sweeteners like sorbitol and xylitol, don't matter for people on the Atkins diet (at least, they don't matter for me), but they matter at least some to diabetics.
• "Slow carbs" are in the same boat as sugar alcohols.

What We Do About Fiber

Fiber is technically a carbohydrate, but it's indigestible and has no effect on blood sugar, so if you give insulin for it, you're giving more insulin than you should, which can result in low blood sugars.

We were told to ignore fiber values under 5 grams, but if there were 5 grams or more, to subtract the entire amount of fiber from the total carbohydrates, and use this smaller number.

Why use two rules? Apparently, most people (a) hate subtraction, (b) have hardly any fiber in their diets anyway, and (c) don't much care about an error of up to 4 grams in their calculations. So the "Ignore fiber amounts under 5 grams" rule allows them to skip the subtraction almost all the time, at the expense of accuracy.

I find the five-gram rule more annoying and confusing than the subtraction, and I don't like the inaccuracy it imposes, so I always subtract the fiber.

What We Do About Sugar Alcohols

Most artificially sweetened candies, ice cream, and so on use sugar alcohols as sweeteners. If you're non-diabetic, sugar alcohols are "slow carbs" that metabolize so much more slowly than other carbs that they're almost like proteins or fats as far as low-carb diets are concerned. If you're diabetic, on the other hand, this doesn't seem to work. Being metabolized slowly is good, but it's not enough.

Apparently, different people react differently to sugar alcohols, and some people can count a gram of these as being equivalent to half a gram of ordinary carbs, but for Karl we have to count them at 100%.  This works fine (and it's what our diabetes team told us would happen), though it makes us wonder, "What's the point of using this stuff at all, if diabetic customers have to use as much insulin as with regular sugar?"

Fortunately, some low-sugar foods, like the no-sugar-added Umpqua ice cream that Karl likes, have zero-calorie artificial sweeteners doing most of the heavy lifting, though they also have sugar alcohols, so the total carbs really are a lot lower than with regular ice cream, with a half-cup serving of vanilla weighing in at only 11 net carbs.

What We Do About Slow Carbs

We like Dreamfields Pasta for low-carb meals. Karl isn't interested in pasta, so we haven't seen what effect it has on him. The word on the Web is that the "unavailable" carbs in this stuff aren't as unavailable as all that, and they get digested slowly over a period of hours, which is great for non-diabetic low-carb dieters, but for insulin-dependent diabetics, their mealtime insulin tends to run out of steam before the digestion is done, causing delayed high blood sugar.

Routine Doctor's Visit

Since there isn't a doctor who specializes in diabetes in our area (an endocrinologist), we're seeing our GP, Shawn Foley, for Karl's checkups. We got a thumbs-up, as we knew we would, and Karl had gained a couple of pounds since last time, which is good, since the bout of diabetic ketoacidosis that resulted in his being hospitalized and diagnosed left him thin. And we got a prescription for more test strips per month, since 200 test strips is only 6.7 per day, and we often go above that. (Especially because the school is very good about testing his blood-sugar levels before and after P.E. to ensure that he doesn't exercise his way to low blood sugar and its attendant problems.)

Is Karl's "honeymoon period" lasting? That's the big question. At the hospital, the doctors claimed that his pancreas' ability to produce insulin was a write-off, and it would vanish soon, in a matter of weeks or months (it's been four months now). This would reveal itself through degraded blood-glucose control and increased insulin requirements. Before this week's doctor's visit, I printed out the trend graphs from Karl's Bayer Contour USB blood-glucose meter, and noticed that his blood-sugar was under slightly better control during the most recent 30 days than it was during the previous 30 days, and his insulin requirements are about the same. That's good!

He might have been doing a little better before that, but it's a little unclear because we didn't start taking after-meal blood-sugar readings in the early weeks, so this revealing piece of the puzzle is largely missing. On the whole, it looks like he's holding steady.

This is good, because, for example, if he comes home from school with a blood sugar of 140, it will be down to a more normal 80-90 by dinnertime without corrective insulin. While we want to keep the load on his pancreas down to a minimum, we're glad it's there to fill in the gaps!

We have been adjusting Karl's insulin dosage on our own, which seems to be standard practice these days. The first several times we did it, we bounced it off someone on our diabetes support team, but since they always agreed with us, we kinda stopped. It's always a matter of tweaking doses up or down by half a unit and keeping a watchful eye on happens, and Karl eats the same meals over and over, so this is not rocket science! What works and what doesn't is revealed quickly.

Does Eating Extra Protein Result in High Blood Sugar?

Even though we're using "advanced carb counting" for Karl's meals, the calculations are incomplete. For example, our diabetes educator and many sources of diabetes information point out that protein will digest partly into glucose, and a "big protein meal, like eating a large steak, will cause a rise in blood sugars," but what to do about this? Most sources are strangely silent.

I've read many times that when diabetics deviate from their normal meal plan -- pizza is often mentioned -- they get high blood sugars even when they do everything "right." So it seems to me that we need a method that's better than "advanced carb counting."

Diabetics on very low-carb diets know that their blood sugars rise more after a meal than the small levels of carbs would indicate, and Dr. Bernstein talks about giving insulin for protein in his book, Dr. Bernstein's Diabetes Solution: The Complete Guide to Achieving Normal Blood Sugars (which has been released in a new edition just this month). He doesn't have as much detail as I'd like, because his approach involves eating the same amounts of protein and carbs with every meal, so his description is a little simplified.


There is whole book about this very topic, or at least I think there is: T.A.G.: A Diabetic Food System, by Mary Joan Oexmann, which covers the effect of protein and fat on blood sugar (in addition to carbs), and what to do about it. Like most interesting books, this one is out of print, but I've ordered a used copy and look forward to reading it.




There's a discussion forum for folks using the TAG system, and I'm starting to absorb the info there: http://www.tudiabetes.org/group/tagers.

As I understand it. the T.A.G. system tells you to consider the carbs, protein and fat in a meal. A gram of protein is considered to be equivalent of 0.6 grams of carbs, and a gram of fat is considered to be equivalent of 0.1 grams of carbs.

In a meal with significant amounts of protein or fat, this gives a number much larger than the grams of carbs alone. For example, Karl's current low-carb peanut-butter sandwiches have only 19 grams of net carbs, but the TAG value is 35. If one unit of insulin is the right amount for this sandwich, then, using carb counting, Karl's carb ratio is around 19, but his TAG ratio is around 35. Or something. I need to read the book!

I'm also looking for research on the subject. No doubt the bibliographies in the two books will be helpful.

Do We Count Grams of Protein as Well as Carbs?

Some people count grams of protein as well as carbs, since digestion converts some of the protein into blood sugar. This is normally ignored in insulin calculations, but the standard diabetes diets are low-protein anyway.

Keeping Karl's carbs down means keeping his protein up, so maybe it's worth doing. Dr. Bernstein, who promotes a very low-carb diet, has his patients track their protein consumption.

Our motivation to look into this is that Karl always has chicken nuggets for dinner (sigh), and this is his highest-protein meal of the day, and it's typically followed by high blood sugar, meaning we didn't give him enough insulin. Maybe if we didn't ignore the protein, the math and the results would align better?

I'm looking into this...

Love-Hate Relationship: the Bayer Contour USB Glucose Meter

How does Bayer manage to prevent their Contour USB blood-glucose meter from getting a five-star rating from me? It wasn't easy! It's easy to use and has a lot of good features, including keeping a record of every blood-sugar reading we've ever taken (over 700 so far). It's easy to use, reliable, and rugged. It has a good display, with color and everything -- and it uses scary colors to get your attention for high and low readings! The user interface with the three on-board buttons works just fine. The battery lasts for weeks and recharges quickly from any USB port.

The USB port is a good feature. The Contour USB looks basically like an oversized USB flash drive, so you can plug it directly into a USB port without a cable. And it comes with a cute little wall charger if you don't habitually charge it on your PC, like I do.

But their software has a clumsy and unfinished feel. The built-in clock doesn't know about daylight savings time. The "Glucofacts" software is supposed to run automatically when you plug the unit into your computer's USB port, and it doesn't. The user interface of the software is hard to figure out for such a simple program. Sigh.

But these are minor annoyances. The reports are pretty cool. Here's a trend from the last month (click the chart to enlarge):

Note that the chart says at the bottom left, "Created on Saturday, November 5, 2011," but the chart has data through November 7. The program hasn't noticed that I launched it two days ago and it isn't November 5 anymore! This is the sort of thing that bothers me about this product. The main features work, but the details are sloppy. Come on, Bayer! Get it right!

The green area is within target (70-140 mg/dl), while orange means low and yellow means high. This graph shows, to my eyes at least, that Karl has a few lows and more highs, but the graph is hard to interpret otherwise.

But check this out, a "Standard Day" graph that overlays all the readings onto a single 24-hour chart (click the chart to enlarge):

Wow, look at how Karl's blood sugar rises in the afternoon and evening! Not all the time, but there are essentially no high readings during the morning, but by the time he gets home from school, between 3:30 and 4:00 PM, we start seeing more and more highs. Not only that, but we see fewer lows later in the day.

To us, this implied that either (a) his bedtime Lantus wasn't lasting 24 hours, or (b) he needs a smaller carb ratio for lunch and dinner (that is, more insulin) than he was getting. So we're trying both twice-a-day Lantus and more insulin at lunch and dinner, and we'll see what happens.

So where do I stand with the Bayer Contour USB product? I've tried three meters, and this is my favorite, hands-down. It's worth it for the display and the easy USB data download so I can put the charts together. I've also used the Bayer Contour, which has a lot fewer features (while taking the same Bayer blood glucose test strips and using the same Glucofacts software). I've also tried the LifeScan OneTouch Ultra Mini, which is okay as a meter, but is bigger than the Contour USB and doesn't have the color display or the built-in USB plug, so I was unimpressed.

My verdict? I love the meter and I'm annoyed at the software, but my advice is, "Buy the Contour USB right now!"

And, Bayer? Do us a favor and hire some software engineers who know what "user experience" means.

We got our Contour USB for free at the hospital (suppliers like Bayer really want your lifetime test-strip business, and a free meter is their best chance to get it). If you don't know anyone with a drawer full of these, they're available cheap almost anywhere. Check out the Amazon link below, for instance.

Check the Obvious When Blood-Sugar is High

Karl's been coming home from school with blood-sugar levels higher than we'd like, and we were mystified. What could explain it?

It turned out that, during the period when we were considered changing Karl's carb ratio from 20 to 25, because of occasional morning lows, we'd actually told the school nurse that this was a definite change! So he was getting 3 units of insulin at lunch, when the right dose was 3.5, and that explains his after-school numbers.

Or so we think. A few more days will tell, but we think that's it because his weekend numbers (where we still use 20 carbs per unit) have been better.

So, when in doubt, go back over the basics with everyone involved.

Keeping Carbs Down and the Ultimate PBJ

If it were me with the diabetes, I'd go straight to Dr. Bernstein's super-low-carb diet, since I've been on a low-carb diet since 2008 anyway. The fewer carbs you eat, the less your blood sugar spikes the less insulin you need, and the fewer health issues and complications you have.

Karl isn't me, though, and he has strong food preferences that don't align very well with a super-low-carb diet, so what do we do?

Karl loves peanut-butter sandwiches. Here's how we game the system to get his carbs down without shorting him on calories.

Step 1: More Peanut Butter!


Peanut butter is a low-carb food that's also rich in fats and proteins, meaning that, while it's low-carb, it's high-calorie. Since the game is to reduce Karl's carbs while keeping his calorie intake high (he was painfully thin after his trip to the hospital), peanut butter is exactly the kind of food we want!

In theory, a "serving" of peanut butter is two tablespoons, but you can get three onto a sandwich without having it drip around the edges. We use tablespoon measures to get this right.

Calories: 270
Carbs: 12 g
Fiber carbs: 3 g
Net carbs: 9 g

Step 2: Dinky Little Diet Bread Slices

Using less bread is the obvious way to cut the carbs way down. Our method is to use bread aimed at dieters, which uses small loaves and thin slices to cut the calories and carbs in half.  We've used Orowheat's 40 calorie per slice bread and Sara Lee's 45 calorie per slice bread. Here's the breakdown for two slices of Sara Lee "45 Calories & Delightful 100% Whole Wheat and Honey" bread:

Calories: 90
Carbs: 14 g
Fiber carbs: 5 g
Net carbs: 9 g

We prefer the Orowheat bread. The Sara Lee is okay, except that it often has big holes in it, which I find annoying at bread that costs over $4.00 what (let's face it) amounts to half a loaf!

Step 3. A Dash of Honey

Karl prefers honey on his sandwiches. We started out with a teaspoon (6 grams of carbs), but when we reduced it to half a teaspoon (3 g), he didn't mind.

Calories: 20
Carbs: 3 g

Step 4: Add It All Up

Calories: 380
Carbs: 29 g
Fiber Carbs: 8 g
Net Carbs: 21 g

We base insulin dosage on net carbs (total carbs minus fiber), which is considered to be "advanced carb counting" for some reason. See http://clinical.diabetesjournals.org/content/23/3/120.full for a discussion on the basic flavors of carb counting.

Variations

Karl prefers to have a peanut-butter sandwich as part of every single meal (sigh), unless we're eating out. He takes one to school with him each day to round out the food he buys in the cafeteria. The school has little in the way of low-carb options, and the sandwich is a key to keeping him well-fed at 75 carbs or less per meal.

Because Karl loves peanut-butter sandwiches so much, it pays for us to spend some energy on getting this optimized for him, and, under the circumstances, we think that a 380-calorie entree with only 21 g of net carbs is pretty good!

A peanut-butter sandwich is the gold standard of kids' meals anyway, so there are worse things.

I've tried substituting calorie-free pancake syrup for the honey, which Karl doesn't mind, but since the honey has only 3 g of carbs and he specifically asks for honey, I don't do this as a regular thing. Using jam instead of honey would be fine for anyone but Karl, who doesn't want jam. Just use an amount that matches your goals.

Tiny BD Nano Needles

Karl uses the tiny little 4mm BD Nano Pen Needles (for insulin pens), which are really short and have worked great for us -- no problems at all, and their reputation is that they're safe and painless -- it's hard to get into trouble with a needle so short! We use them for Karl's disposable Lantus pens and for Novolog in his refillable Novopen Junior insulin pen.

For some reason, our doctors started us on the mini (5mm) needles instead, so we had to get a new prescription for the nano needles. I'm told that some people are using even longer pens -- the original pen needles are three times as long as the nano!

The BD Nano page has a link to a rebate form that will let you try the needles for free.

Let's hear it for pain-free injection!