Monthly Archives: January 2013

CGM Calibration

A common complaint about CGM systems is the lack of accuracy. I frequently see posts on Facebook or Diabetes forums from people who are frustrated that they can’t get reliable results from their CGM. I believe that the problem is often down to calibration; it is very easy to get inaccurate numbers if you calibrate too frequently, or at the wrong time. Since starting on the CGM, I have managed to develop a calibration routine which gives me pretty accurate BG readings, at least with my Medtronic device.

The ‘rules’ are as follows:

  1. Do not calibrate too often. You need to calibrate at least two times a day, although three times a day may be more appropriate, as it gives more flexibility with timings and won’t involve testing in the middle of the night.
  2. Choose the time you calibrate carefully. CGM readings tend to lag further behind meter readings when your BG is falling or rising quickly. Ideally do not calibrate within 2 hours of a bolus and calibrate at least 15 minutes before food. I find that lying down can have an impact on my CGM reading, so I try to avoid calibrating in the 15 minutes before I go to bed. The 15 minutes is because that is how long it takes the CGM to calibrate.
  3. Do not use “Auto Calibrate”. This means any BG you take with a link meter will be used to calibrate the CGM, whether your BG is stable or not.
  4. Try not to calibrate when your BG is too far outside of your normal range. If I calibrate when my BG is much higher than usual (e.g. 15) the CGM tends to be more inaccurate when it drops back to my normal range.

I find the most convenient times to calibrate are when I wake up (around 9am), before dinner (around 6pm) and before bed (about midnight). As long as you don’t spend more than 12 hours asleep, or have more than 12 hours between breakfast and dinner, this sort of scheme should work as well. I know that my blood sugar is usually fairly flat at these times so they are the best times for me.

I have found that trying to follow these rules gives me a Mean Average Deviation of 3%-6%  each day according to the Carelink software. This means that my CGM readings are within about 6% of my meter readings. The exception to this is when I have a very high BG, as the CGM tends to under-read in these situations (e.g. the CGM might read 17 when my BG is 22). In practice, I find that this isn’t too much of a problem, because if the CGM claims my BG is 17 I’ll be paying quite close attention and testing frequently anyway.

I hope that someone finds this useful and manages to get better CGM readings as a result.


Share Button

The Importance of Changing Pump Sites

The packaging for Medtronic’s pump sites states that you should change them every 48-72 hours. I’ve never been particularly good at sticking to this, sometimes making it as far as 6 days before changing it. I didn’t seem to have any issues with infection or irritation of my pump sites, so I tended not to worry about it too much, as it didn’t seem to have a noticeable impact on my blood sugars.

Since my control has improved, I have noticed a much bigger difference when a pump site starts to get old. Over the weekend I let a pump site get to about 4.5 days before changing it. During the last 24 hours or so before it changed the site my BG levels were noticeably worse, running at 8-10 when I would usually be 6-8. I suspect that it has always had this sort of effect on my blood sugar, but because my control was poor I didn’t really notice the difference between 15 and 17. From now on I’ll have to be more careful about how often I put in new cannulas.

Share Button

The impossible CGM graph

This morning I was very surprised to see my CGM graph was perfectly flat for the last few hours. I’ve seen almost flat lines before, or flat lines which completely disagreed with what my meter was telling me, but when I tested it was within 0.2 mmol/l of the CGM reading!

This never happens

This never happens

My porridge experiment has been continuing, and I am very impressed with my BG results. I think I have had the same result consistently enough to say that it isn’t just a fluke. I am regularly spiking no higher than 10-12 about 60-90 minutes after eating, under 10 at 2 hours and back where I started (or near enough) by 3 hours. I just wish I’d discovered this earlier!

My Insulin doses have increased a little further – I am now on 1u:10g carbs for breakfast, 1u:15g for lunch and 1u:20g for dinner. Breakfast and lunch are both almost correct but dinner still needs some work. At the moment I’m usually 4-7 pre-dinner, and I can’t quite bring myself to do the full dose of insulin when I’m < 6 due to anxiety. It is something that I am working on, and hopefully I’ll be able to get rid of my large evening spikes in the next week or two.


Share Button

The Porridge Experiment part 2

Wow. I’m really liking porridge for breakfast!

Today I went for 40g oats, 135ml milk, 80ml water for a total of 30g carbs,  I managed to locate a large pyrex jug, capable of holding the best part of a litre. Apparently this still isn’t big enough for a bowl of porridge! Thankfully I managed to stop the microwave before any of it escaped. I decided to add a tiny pinch of salt and it really does make a big difference to the taste – I’m not sure I’ll need to add anything else to it.

As for the effect on my blood sugar level; I was 7.5 before eating, and I bolused just after finishing the porridge. According to my CGM, the highest BG I reached was 10.4, I was down to 9.5 at two hours post-food, and down to 6.9 before lunch! That is the best post-breakfast result that I can remember.


BG post breakfast

BG post breakfast

The small black marker on the lower left of the pump display shows when my breakfast bolus was. I was already rising slightly from waking up (which might be the next thing I need to look at). It was quite a gentle climb compared to what I normally see with cereal, so I guess the next thing to try is bolusing before I eat.

Share Button

The Porridge Experiment part 1

For the last few years, I have stuck to a very similar weekday breakfast. For the past couple of years, it has predominantly consisted of 30g Weetos and 125 ml milk. I found that consistency helped with control but eventually it starts to get a bit boring and I was having problems dealing with the large fast spike that breakfast cereal brings.

Often I would spike from 6 or 7 when I woke up to 16 or 17 before my blood sugar started to drop again. I originally attributed this to bolusing after I finished eating and tried bolusing before I ate to give the insulin a chance to get in to my system before the cereal did. My blood sugar still spiked to at least 13 or 14, which is better but still not ideal.

After doing a bit of reading, I decided to try porridge. I’d heard mixed reviews of porridge as a diabetic breakfast. Some people are left feeling full all morning with a completely manageable blood sugar while others are hungry after 10 minutes with blood that approximates syrup.

As porridge has a much lower G.I. than breakfast cereal, I decided to leave bolusing until after I ate just in case it was far far slower than my insulin. This turned out to be a good thing. I dutifully calculated how much milk/water/oats I would need for 30g carbs, weighed it all out and put it in a bowl before putting it in the microwave to cook. When the instructions say ‘Place in a large bowl’, they really do mean it. The bowl I used was only half full when I put it in the microwave, but soon it was overflowing with porridge! Before I could open the door to stop the cooking, a rather large proportion of my oats were all over the bottom of the microwave. After clearing them up and letting it finish cooking, I tried to weigh my remaining porridge to have some idea of how many carbs were left in the bowl and estimated around 20g and bolused for it. I’m not entirely sure how close I was.

The porridge itself was nice and definitely filling. By the time I got to lunch I was only just about hungry again. The important part, however, was my blood sugar. From a pre-breakfast reading of 8.0, I spiked up to around  13.3, before dropping back to 9.7 pre lunch. While the numbers themselves weren’t ideal, a rise of around 5 is pretty manageable, especially considering I bolused after eating and wasn’t entirely sure how many carbs I had. I’m looking forward to trying them again tomorrow to see if I can do any better!

Porridge Experiment, part 1. Status: Success.

Share Button

Hey! I Need That To Live…

Diabetics tend to build up a collection of tools they need to help manage their diabetes. Around the house I have several BG meters (some of which have probably only been used once) as well as various methods of delivering insulin. What follows is a description of the tools I currently use, and would love to hear some suggestions about better products which others use.

I am lucky enough to have an NHS funded insulin pump. Relatively rare in the UK, insulin pumps are the closest thing we have to actually simulating what a normal pancreas does. Currently I use a Medtronic Paradigm Veo.

Medtronic Paradim Veo

Medtronic Paradim Veo

Yes, I do have a purple pump. My original one was black, but I dropped it a few too many times and eventually the case started to crack and the battery connection died ( I’m surprised it didn’t give up sooner).

As someone who is often worried about BG levels, I am never very far from my BG meter. When I was diagnosed I was given a Bayer Asencia Contour and I’ve pretty much stuck to this brand since – I upgraded to a Bayer Contour Link as it communicates directly with my pump, saving time with BG entry.

Bayer Contour Link

Bayer Contour Link

I’m definitely looking forward to being upgrade again when the new version (Bayer Next Link) is released in the UK. For now I’ll just have to look longingly at the US product page.

Due to my fear of lows, I test my BG rather a lot. Over the past 4 years, I’ve tried just about every different finger-pricker-thingy that I can get my hands on in an attempt to find the one that hurt least when making up to 16 new holes in my fingers each day. Until recently, the Bayer Microlet was the winner – until I got a free Freestyle Insulinx meter through the post. I was starting to struggle tog get blood out of my fingers using with Bayer lancet, but the Freestyle one seems to get more blood out with almost no noticeable needle impact at all – in fact I’ve had a couple of finger pricks which didn’t hurt at all which I struggled to stop bleeding. My fingers definitely appreciate the upgrade.


Freestyle (left) and Bayer microlet (right)

The last bit of technology I have is a Constant Glucose Monitor (CGM) from Medtronic. Sadly, these are virtually impossible to obtain on the NHS, so I’ve had to take the path of self-funding to be able to obtain one. It consists of a small sensor that sits just under the skin and transmits a blood sugar reading to my pump every 5 minutes. While not entirely accurate, it is a valuable tool to pick up trends in BGs, as well as catching low and high blood sugars before they occur. This currently costs around £200 for the transmitter, and £55 per sensor, which lasts for 6 days.

Hopefully I’ll manage to get NHS funding for the CGM eventually, until then I’m going to be paying for it out of pocket because I find it so useful.

Which devices do others use? Why did you choose them?

Share Button

Tasty Snack

Finding nice snacks which don’t consist of either plain vegetables, or a lump of cheese can be difficult for a diabetic. Lots of snack foods have quite a large carbohydrate content, which isn’t exactly ideal for maintaining steady blood sugars. I was very pleased to discover a recipe the other day over on Diabetes Daily for “Cheese and Sausage Muffins”. These are low carb (5-6g each) and quite filling. Having said that, a batch of 5-6 can still disappear rather quickly.


1 pound sausages
3 large eggs
1 cup almond meal / almond flour
½ teaspoon salt
¼ cup grated Cheddar cheese
¼ cup Parmesan cheese


  1. Preheat oven to 375°F (190°C)
  2. Remove sausage from casings
  3. In a mixing bowl, combine the sausage meat, egg, almond flour, cheese and salt.
  4. Mix together well.
  5. Divide evenly between 5 or 6 greased muffin tins.
  6. Bake for 25-30 mins.

Make sure they are thoroughly cooked all of the way through!

So far, I’ve tried this recipe with plain Beef, plain Pork and Pork with Chorizo sausages, and they all taste great. You could add BBQ sauce to give them a bit more of a kick, or use any type of sausage  you prefer – I’d love to hear about any new sausage combinations which work particularly well.

Share Button

So It Begins

Diabetes has been a big part of my life for the last 4.5 years. Diagnosed just days after my 22nd birthday, I found myself having to think about food, exercise and my body. Since the day I was packed off to the hospital with a blood glucose (BG) of 22.4 and ++ ketones, a huge amount of time and effort has been spent trying to get my BG under control. Yet, despite this, I have never really managed to achieve the control I desire.

Most of my problems can be traced back to one single event – a hypo while shopping. About 6 months after diagnosis, I found myself standing in the local Sainsbury’s with a BG reading of 2.7. A rather large amount of sugar eventually fixed the problem, but left me rather shaken. From then on, whenever I felt my BG dropping, or my reading was slightly lower than expected, I panicked. Only, I didn’t realise it was panic. The adrenaline rush of fear feels incredibly similar to the adrenaline rush of a hypo. What should be considered normal blood sugars soon felt too low, not being comfortable with BG readings below the mid teens. I also stopped going out much due to the fear of having a hypo.

After about a year of becoming more and more reclusive, I ran in to a major problems – I started to have panic attacks when leaving the house for anything, even work. At this point it became obvious that it had gone way too far. The last 3 years have been spent trying to overcome my anxiety about hypoglycaemia. I’ve tried multiple treatments for the anxiety, including therapy, beta blockers and SSRIs. All of these helped, but none of them cured the problem.

During December I had an a1c test, which came back as 9.2%. I was getting increasingly frustrated with my inability to do things I enjoy without the constant worry of going low and more and more worried about the complications I was lining up for the future. As a result, my new year’s resolution is to fix my blood sugar levels. Since the start of January my average daily insulin dose has increased by 30% (18u -> 24u) my average daily BG has fallen by 33% (15 -> 10) and after spending time at some more sensible blood sugar levels, my anxiety has reduced. The purpose of this blog will be to keep a record of my progress and to hopefully share any useful information uncovered along the way.

Share Button