Before this all goes completely off-topic (the diabetes topic, that is), I thought I’d share my bicycle cycling experience on the two brevets I did this year. Complete with insulin types used and stupid mistakes made and everything. Might come in handy for someone else looking into it. Possibly. At least I’ll be a bad example.
Insulins: Lanthus basal, Novorapid bolus. ICT with one Lanthus injection in the evening.
The trip started on a fine Saturday morning. I reduced the basal injection the night before (around 9pm) to 75% of normal . The morning after I had to get up at stupid-o’clock to travel to the starting point. A large breakfast with loads of tasty carbohydrates was consumed – with slightly more Novorapid to cover for the lack of basal insulin. Morning blood glucose wasn’t too bad, actually, which surprised me with the reduced basal rate. After that: cycle to the train station, hop onto a train and check the blood glucose levels, approximately 1.5 hours after breakfast – a duigusting 220mg/dl (12.2mmol/l) was the result .
Commence heroic correction (yes, that was stupid)! Meet up with all the other strange people, who enjoy cycling until it hurts, for a common breakfast at the starting line. Another blood glucose test found 80mg/dl (4.4mmol/l). Which I considered a bit low, just before sports. The correction for the “low” consisted of three pieces of cake – without a bolus injection, which turned out to lead to a high blood glucose during the morning – which isn’t that surprising, in hindsight).
Start of the actual event was at 9.15h in a large group of cyclist in front of a backdrop of black clouds, sunshine and a strong tailwind. Up to the first control (places to stop and get a stamp at – to prove you went the right way) I then had to keep correcting my way too high blood glucose (I used Novorapid at 1 I.E. per 50mg/dl(2.8mmol/l) reduction – which is identical to what I use outside of sports ). I attempted to test my blood glucose on the bike with the Accucheck Mobile taped to my top tube – which didn’t work. At all. The lancet device requires two hands. We did, however, stop occasionally allowing me to test. For the record: The highest blood glucose tested during the beginning of the event was 500mg/dl (27.8mmol/l) .
After roughly 75km (46 miles) my blood glucose had dropped to 120mg/dl (6.7mmol/l) and no further corrections were necessary. The speed at which we were travelling was set to about 24km/h (14 miles/h) – brevets are not a race, more of a friendly cycletour – so I felt quite happy at this blood sugar levels.
At that time, I fell back to my normal feeding plan for cycling. Which is not ideal for weight loss, but I’m all for the “can’t have everything” motto. Approximately every 45 minutes I consumed 40g of carbohydrates in varied shape and form (all of them relatively quick: bananas, dates, coke, sport-gels) and at every control (~ every 40 miles) I added something savoury. Typically a cheese or ham sandwich, so I wasn’t forced to eat sweets all day.
The good news: blood glucose stayed stable for the rest of the day. Between 105mg/dl and 130mg/dl (5.8-7.2mmol/l) – which apparently means I did everything right!  An interesting (and not diabetes related) problem in the afternoon: We stopped at a supermarket to get a cup of espresso and I grabbed a bottle of Strawberry/Banana-Smoothie instead of the juice I had had in mind. This is drinkable from a bicycle bottle (barely), but it’s not ideal. At all.
There were problems, of course. A crash in our group forced two to quit due to damage to the bikes (too many broken spokes on one, a broken brake lever on the second bike). A few hailstones fell on our jerseys in the early afternoon, when the thunderstorms caught up. My elbows started to hurt (at around the 150km mark (95 miles)) and my bottom complained (starting around the 175km mark (110 miles)). Any climb over 18% (luckily there were only two) was not much fun – especially in my weight class.
Useful side effect: I noticed that my Lanthus dose worked for almost exactly 24 hours (this is apparently sometimes debatable). After 24 hours 15 minutes my blood glucose started going up, which I covered with more food and some extra bolus to prevent complete lack of insulin.
The recovery period lasted for about 8 hours afterwards – with a still reduced insulin amount (again: 75% Lanthus) and more food (which I was quite happy for at that time, I’ll freely admit).
Total time for the 232km (144 miles) was about 13 hours. With plenty of breaks and a lovely group. The plan for the next ride was to go about everything a good bit more relaxed and to not correct too much (in both directions – cake and starting sugar).
 A Lanthus problem. As the period of effect is 24 hours, I need to reduce the basal rate long before the sport actually started. This obviously leads to high blood glucose levels in the morning. This got better on the next attempt, using Levemir, and should not be a problem any more with the pump. Yay for technology in this case!
 Yes, I should have known this was only the post-breakfast high and that the breakfast bolus was still working. I probably was just too tired (or stupid, but I shall heroically ignore that possibility).
 And this is probably too high. Not recommended for imitation.
 General consensus is: this is too high to do sports. I could not detect any ketones and actually felt good, so I continued. I would definitely recommend against this, though. Blame it on my stupidity.
 Lest this sounds too impressive. I got lucky. My group also had a work collegue (who cycles a lot more than me and who’s got plenty of long-distance experience) and his wife (who cycles a tiny bit slower than me). They set a pace that was ideal for me with only minimal exertion. This was probably perfect for my “diet plan”. Next time I’ll be on my own and will have to eat more, as I’ll probably not be able to pace myself as well.
 Oh “Along the river”? A horrible LIE! The part along the river (which is by definition relatively flat) was limited to about 32km (20 miles) – the rest was a bloody up and down the surrounding hills.