Continuous Glucose Monitoring Part 2

... This is what happened.

Just before the date I got lazy and then sick. The perfect storm that for me translated into a period of massive insulin resistance change. My insulin requirement more than doubled in three days. My blood glucose was not well controlled during this time. Three days into this perfect storm, I was well enough to start exercising again. How this affected my blood glucose control even surprised me. I almost immediately move into a period of great control.

The experience that I had when using the CGM was completely different when I was well controlled from when I was poorly controlled.

When poorly controlled.

Within the poorly controlled period of three days I experienced every possible reading error you can think of. I.e. The CGM told me that my blood glucose was high when it wasn't. It told me that I was normal when I was high. It told me that I was low when I was normal. It told me that I was normal when I was low.

When well controlled.

Within the well controlled period the CGM reading was extremely accurate. After 12 hours between calibrations (the maximum time between calibrations), the tested reading was often only 0.1 mmol/l out.

It is hard to understand why there is such a discrepancy. I suspect that it is largely due to the calibrations. When you re-calibrate the sensor there is an assumption made by the device that at that time your subcutaneous glucose reading is the same as your blood glucose reading. We know that the readings are not the same because there is a 10 - 20 min lag in the subcutaneous glucose reading. This is the reason why it is recommended that you calibrate the device when your blood glucose is stable and not changing quickly, limiting this error. When poorly controlled it is hard to find a time when your blood glucose is not changing quickly. The sensor has certain calibration requirements that may mean you are forced to calibrate when you would have preferred not to. This would result in inaccurate calibrations which in-turn results in inaccurate readings.

Calibration could not have been the only reason for the inaccurate readings. My blood glucose wasn't changing fast enough at calibration time to account for more than a 1mmol/l error (huge error), but I sometime had errors greater than 5 mmol/l. It could not be the hardware, because the well controlled period followed the poorly controlled period using exactly the same sensor (sensors get worse with time not better).

Whatever the reason, CMG is a much more useful if you are well controlled.

CMG may not be the silver bullet I was hoping for, but it is still an incredibly useful tool. It helps to get the timing of your basal changes correct. You can see exactly when the change needs to be made. It also checks you blood glucose when you normally wouldn't. There are likely to be one or two surprises. There was for me.

Maths, your health depends on it.

The truth is that a certain amount of maths is required to manage type 1 diabetes well. Another truth is that you don't need to be good at maths to succeed in managing diabetes well. It's a question of balance.

I am on my mountain bike at the bike park. It's a blue route. There are two small technical up and down hills followed by a much larger even more technical steep uphill, down and uphill again. Getting up is a bit of a balancing act similar to how I find managing diabetes. You cannot lean too far forward or the back wheel will slip, nor can you lean to far back or the front wheel lifts. You cannot go too slowly because you need momentum to get up the last part, nor can you go too fast or you won't have enough power for when you do slow. You also need the correct line and balance. Basically everything needs to be perfect to succeed. I have done the route before but it still takes more than one attempt to get up.

Then it's the trail around the edge of the mountain. The ground drops away on my left and I feel that sense of exposure. I love it. This is my fortnightly weekend morning getaway. It's great exercise and great for taking my mind off things. It is not working today though. I am in a deep conversation with my brother in-law who is right behind me. I am panting now because the dip with a hair-pin turn to the left is followed by an up-hill I say, "... but It's not that hard. Why would anyone let percentages and basic addition get in the way of good health?" We are at the top now. It's a fast series of tight massively cambered turns. Mind the rock. "You don't get it Paul. Some people are just terrified of maths!" he says. I am lucky to have been blessed with an aptitude in Maths and this is taking some time to sink in. It's a steep downhill and I pickup speed. I need to concentrate now because of the sharp rocks. They are a bit tricky at speed and you don't want to fall here. Regrouping on the other side I say, "... but it's their health! What can me more important?"

My brother-in-law and I are discussing a topic that has been bothering me ever since I was diagnosed with type 1 diabetes. It started with comments like "Don't take advice from that patient. They didn't have the aptitude to work it out so they are on a different treatment." or "You are so lucky you can work these things out. Most people don't and just suffer." I am using a bit of poetic licence here, but you get the message. Another saying "As jy dom is moet jy hard kak" comes to mind.

Initially I just accepted what I was told. Relieved that I 'got it' I moved on. Now that I have had diabetes for more than 2 years and know what is required to know pretty well, these early comments now haunt me. There is a lot that you need to learn, but the maths that you use daily is very basic. Managing diabetes well is more about understanding and balance than maths. I have never heard someone say "You are just not clever enough to ride a mountain bike". Balance is instinctively learned once the problem is understood. I think the fault lies with either the teacher's ability to teach, or the student's willingness to learn, or access to a good teacher, or just the sheer amount needed to be learned. So either find a good teacher, or change your attitude, or get a good book.

If you are not yet sold by my argument, and are 'terrified of maths' or struggle with your daily maths requirements, I have some other suggestions:

1. Get the gadget. You can buy a scale that does carb-counting for you. You put the food on the scale, enter the food type and ... it tells you the weight of the carbs.
   
2. Phone a friend. If you really struggle with the numbers or just want some confirmation, then ask for help. This is serious stuff and most friends are more than willing to help out.
   
3. Do the easy maths before the hard. There is always more than one way to get to an answer. Get someone to help you find the easy way. Another example is to learn to carb-count using exchanges first. This is a lot easier than the gram based carb-counting.
   

What about mental maths?

Now you should obviously not try this until you are confident doing the calculations on you calculator. However if you have got the hang of it there are many useful tips and tricks you can use that make mental maths easier. I want to share one of these tips that I use quite often when carb counting. I.e. how to calculate percentages in your head? Here is an example.

I am quite hungry and in a rush. I need a decent meal quickly. I grab the loaf of low GI bread. (This loaf is not really low GI, but rather 'lower GI' than normal bread. It is quite safe for me to eat.) I grab four slices and pop it on the scale. The scale reads 166g. Recently I have found that the bread thickness varies too much so I prefer to weigh the food. The nutrition label on the loaf says that there are 38g of carbs in each 100g of bread. I read that as 38%. This means that I need to calculate 38% of 166g to get the total carbs in grams of the four slices that I am about to eat.

The trick to mental maths is to change the problem slightly to make it easy. There is nothing easier than working with tens or a multiple of tens. That's what I look for.

38% = 40% - 2% = 4 * (10%) - 2 * (1%)

Now getting 10% from 166g is easy because you just need to move the decimal place by one. 
I.e. 10% of 166 = 16.6g   
and similarly 1% of 166 = 1.66g
Let's start the calculation: 40% of 166g = 4 * (16.6)
We don't need an exact result so we can safely round to the nearest half gram.
≈ 4 * 16.5 = 66g

Now just subtract 2 % of 166g = 2 * 1.66 ≈ 2 *1.5 = 3g
Total carbs = 66g – 3g = 63g.
There are 63g of carbs in the meal. (As it turns out in this case the exchanges method would have got us close enough and is a lot easier to calculate.)

Mental maths requires a bit of practice before you get good at it. So try it more than once, and soon you will find this quicker than reaching for a calculator.

Loading on the carbs

Eating a large carb meal can be tricky for someone who is living with type 1 diabetes. The larger the carb meal the harder it is to avoid hyper or hypoglycaemia and the more likely it is to have a severe hypo. This is because any error in insulin dose gets amplified by the size of the meal. I will explore in more detail the things that can go wrong when eating a large carb meal and how to avoid them.
Since I have started writing my blog, I have become more involved in the diabetic online community. One thing that I have noticed is that many people with type 1 diabetes don't eat large carb meals well. One guy even described the hyper that followed a high carb meal as if it was a punishment that he needed to endure for breaking his diet. I believe this is wrong. A high carb meal doesn't have to be a bad experience.
I love eating carbs and I eat a lot of them. I am practically carbo-loading all the time. This meal choice has a lot to do with the amount of exercise that I do, which is a lot. Because of this I have become quite good at eating large carb meals. I still get the occasional meal wrong, but I get most meals right. This is what I have learned.
Getting the insulin dosage correct is not an exact science so there is always an error. The answer to eating large carb meals well is to keep the errors small.

1. Carb-counting needs to be more precise. Weigh the carb separately if you can. Start counting the carbs in the parts of the food you normally don't count. E.g. the 2-5grams in the sauce etc...
2. Any mismatch between the food's GI and your insulin absorption will be amplified. If you are using a rapid acting insulin analogue like Humalog or Nova rapid, then you need to particularly watch out for fatty or low GI meals. Increasing these meals can result in a hypo after the meal.
3. I think it is not wise to more than double the amount of carbs you normally eat for a meal. The insulin-carb ratio you are using is probably not accurate enough for such a large change.

The above points are not always that obvious to identify. I would like to share two examples where it wasn't. In both examples I detail what happened to me when I had a second helping of a meal that I normally eat well with no ill effect.

Single portion pasta meal

I have a favourite pasta dish that I frequently eat. This pasta dish is a favourite not only because it tastes good, but because I always have a good blood glucose response and I feel good afterwards. One day I felt particularly hungry and had a second helping.

Double portion pasta meal

What I did not realise was that the nutrition label was out by 6% (normally a 3 grams error). I also never bothered counting the carbs in the pasta sauce (normally a 2g error). This is a total undercount of 5g of carbs in my normal portion size. I can tolerate a 5g undercount without any problem so I didn't notice the error (check the graph above). When I doubled the meal size, the 5g undercount became a 10g undercount. The second graph shows my not so great blood glucose response. I had a hyper of more than 10mmol/l after the meal and my blood glucose stayed high until the next meal.
The first time that this happened to me I found it very confusing and it took me a while to work out that there was an error in my carb-counting in both meals.

Single portion slightly fatty meal

Here is another example. This time I ate a double portion of a slightly fatty meal. The first graph shows my blood glucose response when eating the single portion. My blood glucose response took a dip at the 1 ½ hour mark but stayed within the normal range and I didn't notice the insulin – food GI mismatch. When I doubled the meal size then I had a hypo an hour later. The graph below shows what happened.

I now recognise a potentially too low GI meal and prevent any ill effects by using a duel – wave bolus (50%, 50% with a 30 min square wave) on my pump. (If you don't have a pump delay giving the insulin can achieve a similar result.)

Double portion slightly fatty meal

When eating a large carb meal, I am particularly careful when counting carbs. I avoid low GI or fatty large carb meals. I also often use my blood glucose response from a large carb meal to fine tune my insulin-carb ratios. I am continuously working at improving my carb-counting checking my food GI and fine tuning my insulin carb rations.


The opposite is true too.
If increasing the carbs in a meal amplifies any errors, then why not decrease the carbs you eat in times of uncertainty. For me that time is when I am on holiday.
One of the reasons why I go on a holiday is to unwind a bit, but this de-stressing completely throws out my insulin – carb ratios. This can make my holiday unpleasant and stressful in a whole different way. What I have started doing is to reduce the amount of carbs that I eat when I am on holiday. My insulin-carb ratios still change and I need to adjust for these changes, but I suffer from less hypo's and hyper's. My holidays are much more pleasant for everyone.