My interest in understanding fuelling for diabetic endurance runners is something that has been troubling me since I finished the Bamburgh Coastal Trail Ultra Marathon last October. I seriously underestimated the fuelling levels required on that run, and though I finished in good time and high spirits, I soon found myself having a diabetic episode. I cannot tell you for sure if my Blood Glucose Level (BGL) was too low because, as a Type 2 Diabetic, I had ceased to be prescribed test strips for my BGL meter in 2008. I was told there was no need now that my circumstances had changed. Briefly, the background is this:
I’ve been diagnosed T2 diabetic since 2001 and a runner of sorts since then completing my first marathon in 2006 in 3:59:01. For about six years my BGLs were completely controlled by exercise. I monitored by BGL religiously and eschewed any attempt to be put on meds. In 2007 all that changed when I lost my Anterior Cruciate Ligament (ACL), landing me in hospital and unable to run. I then had a ‘clean up’ operation in 2008 leading to another protracted period of reduced exercise. During this period I was put on a regime of 3x 500mg Metformin per day (1x per meal). I was informed there was no longer any need for me to monitor BGL levels and was no longer eligable for test strips for my meter. However, as my running built back up, I engaged in longer and longer endurance runs, determined to better my pre-op marathon time (easily beating my previous PB time in 2010). I also moved up to Ultra distances and noticed that all was not well with my fuelling strategies on the run (I discussed this in a previous blog – Running with Asthma and Diabetes).
I raised concerns about my symptoms during and after long runs (and I expressed my desire to continue with the longer distances), with Dr Ellie Dow at Ninewells Hospital who was both interested in what I was doing and insisted I have a BGL/Ketone meter again. This has proved invaluable as I noticed some curious anomalies such as shorter, sharper runs appearing to lower BGL dramatically while longer runs seeming to leave my BGL considerably higher than when I started, or at best, hardly changed. Obviously this has a lot to do with the fact that for a 6 mile run I would not take fuel on board, while on anything over 10 miles I might perhaps fuel at the start (BGL depending) and also again on the run if I felt weak/nauseous. This strategy has seen me through over a dozen marathons and ultra marathon distance runs previously, but in truth, apart from my first marathon, I’d never had the chance to actually test what was going on with my BGL during or after a run.
Once I had the meter I did some random checks to see how I was at the end of either a run or a cycle and two things struck me: Firstly the often repeated notion that a Type 2 diabetic cannot have a ‘Hypo’ (Hypoglycemic Attack) – or a BGL lower than 4 – was blown out of the water on my first test. That came after a six mile run in Sweden where I was shocked to find my BGL was sitting at 2.1. Having lived for a year with a T1 diabetic I knew this was serious, but took on some carbs and put down the result as an anomaly. But over the next month I found I was regularly down at a BGL of between 3-4 after exercise (once had a reading of 1.8 – but that just has to be wrong). The solution was to ensure I was better fuelled at the start of a run or bike ride. I also started to suspect that when I started with a higher BGL, that a run would reduce the glucose level by a couple of points. However a swift ride on the mountain bike through the forest seemed to strip the sugars out of the system more consistently than running for the same period of time. This appeared confusing to me, not least as cycling feels like less effort and I feel less tired at the end of it. So on 20 July I decided to do a blood test before going out on the bike – it came out at 5.4. I wondered what would happen if I took a 51g Zipvit carbo gel followed by a thrash round the forest. About 50 minutes later I was surprised to see the BGL was pretty much the same despite the huge carb intake. So I decided to keep a log for a couple of weeks, recording BGL results at the start and end of exercise (and if high at the end, about 30-40 minutes afterwards). My desire to log these results was given new impetus when I read Gavin Griffiths’ (T1 diabetic) very interesting blog on Professional Advice for the diabetic endurance athlete. Gavin’s blog notes why BGL can be higher at the end of exercise rather than lower, and is well worth a read.
The routes I cycle/run are flat forest tracks which themselves are a mix of soft sand, earth/mud and grass tracks. Comparing a 50min/one hour workout in this environment, it is appears that while a run will usually drop the BGL a couple of points on the UK scale (say from 8 to 6), a bike ride on the trails can reduce the level by as much as 8 points (see the 10 mile bike ride, one hour after a main meal on 6 August: 14.2 down to 5.8 in 49 minutes). Also worthy of comparison are the results from the bike ride on 20 July to those of 26 July where I took the same 51g Zipvit carbogels at the start of a 15 mile run. After two and a quarter hours on the hoof, I found the BGL higher than when I started, confirming the advice given to Gavin last month.
These results put me into a bit of a quandary as I started running to help keep my BGL down, yet I cannot sustain a distance run without taking on carbs – so a simple conclusion one could draw is stick to the short fast runs or stay on the bike. BUT – while I may have started running for that reason, that is not the only reason I do it now. As an asthmatic I find running distances is highly beneficial for my breathing and I simply love long distance running. So as the experiment continued I reviewed the way I had fuelled on those races I’d done earlier in the year to see if I could fuel sufficiently for distance while keeping the BGL at a safe level.
On the two 30 mile ultra runs I’d done in May for RelayGB, I had used 30g Lucozade carbo gels every 40 minutes and a half a banana every other hour. At the end of both runs (undertaken on consecutive days) my BGL had been 5.9 and 3.9 (actually first test read 2.4, but was 3.9 within munutes) respectively. On the two longer ultra runs I’d done in June (32 and 50 miles respectively) I had taken on far more carbs than required and finished at 10.5 and 7.6. Having now charted these results I opted to go for a long run yesterday and experiment with the meter on the run. I was surprised to find that over 20 miles I had taken on 36g at the start and 50g after six miles and needed nothing until the end (finishing with a BGL of 5.8). I felt hungry and under fuelled, but clearly the Zipvit gel was good for a full 14 miles!
On reflection then, I think I’ll relegate the Zipvits for use on my cycling or long ultras and return to the trusted Lucozades for my daily running (if required and with bananas where applicable), taking 2x 30g gels evenly spaced rather than having one massive hit from the 51g Zipvit and then running ’empty’ [and the Lucozades are lighter and taste better]. I’ll keep the log running throughout the build-up to the Speyside Ultra (36.5 miles) on 25 August and maybe thereafter. I really do want to understand and get this fuelling dilemma sorted.
But there is also an issue raised by fellow diabetic runner (T1) Dave Sowerby relating to this experiment. On his blog post What’s Better? High or Low? Dave raises an interesting, if controversial point about striving for the best HbA1C average by keeping BGL low (the HbA1C roughly corresponds to the BGL, and mine has been at steady 6.9 for the last two years). Now that I have taken cognisance of the fuelling I seldom get as low as previously, but this must necessarily push my HbA1C average up in the longrun – the very thing I must avoid to remain within the ‘safe’ limits for a T2 diabetic). Unlike T1 diabetics who can regulate (to a degree) with insulin, I have no antidote to excessively high BGL other than to take more exercise, and this is simply not always convenient despite my best endeavours. There is a part of me that agrees with Dave Sowerby’s conclusions; that the lower results I was getting in April may be better for me in the long run, but I will continue to monitor my fuelling strategies in consultation with my diabetic support team at least until my last race of the year in September (Tamarindo Marathon, Costa Rica). And as for that apparently obvious choice of prioritising cycling over running – I don’t think so, at least not yet.
|Activity & Date||BGL Before + Carbs||BGL End||BGL (30-50 mins after)|
Previous exercise induced lows before checking BGL before exercise
|6 mile run (6/4/12)||2.1 – HYPO|
|8.5 mile run (29/4/12)||2.7 – HYPO|
|12 mile bike (1/5/12)||3.8 – HYPO
|11 mile bike (8/5/12)||3.4 – HYPO
|6 mile run (30/5/12)||3.3 – HYPO|
|6.5 mile run (11/7/12)||5.8||3.1 – HYPO|
BGL post Marathon/Ultra distance runs (where known)
|31.6 mile run, RELAY GB (15/5/12)||5.9||6|
|29.4 mile run, RELAYGB (16/5/12)||2.4 (checked within minutes 3.9) – HYPO|
|32 mile run (12/6/12)||5.4 (30g)/23 mile 8.2 (30g)/
26 mile (50g)
|50 mile run(23/6/12)||5.6 (50g)/18 mile 7.2 (50g)/
42 mile 7.9 (50g)
Daily results (taking cognisance of BGL pre-exercise & introducing carbs accordingly)
|10 mile bike (20/7/12)||5.4 (50g)||5.6||4.8|
|6.5 mile run (21/7/12)||7.4||6.1|
|Mountaineering (22/7/12)||No Info|
|6.5 mile run(23/7/12)||7||7.3|
|6.5 mile run (25/7/12)||8.2||7.5|
|15 mile run (26/7/12)||7 (50g)||8.1||8.7|
|10 mile bike (27/7/12)||10 – Post meal HIGH||3.8 – HYPO|
|8 mile run (28/7/12)||6.8||7.7|
|11 mile bike (29/7/12)||6.9||3.2 – HYPO||4.1|
|12 mile run (30/7/12)||6.1 (30g)||8.4||7.2|
|Mountaineering (31/7/12)||No Info|
|6.5 mile run (1/8/12)||7.9||4.9 – LOW||5.1|
|11 mile run (2/8/12)||6.6||5.8||6|
|13 mile bike (3/8/12)||7 (50g)||4.7 – LOW||5.4|
|15 mile run (4/8/12)||7.2 (30g)||9.2||7.6|
|5 mile run (5/8/12)||5.9 (30g)||8.7||7.2|
|10 mile bike (6/8/12)||14.2 – Post meal HIGH||5.8||7.2|
|20 mile run (7/8/12)||7.6 (36g)/
6mile 6.3 (51g)/