Dapagliflozin, alcohol, and ketones

[HairySmurf]

Hi folks,

I had an experience last night that I’d like to share and I’d like to ask if anyone else has had a similar experience.

I am taking Xigduo twice daily, which contains 1000mg of Metformin and 5mg of Dapagliflozin (AKA Farxiga).

On New Year’s Eve I drank a lot of alcohol, which is not at all advised when on either Metformin or Dapagliflozin. I went to sleep at around 5am and didn’t wake up again until nearly 1pm. I was thirsty with a dry mouth, definitely dehydrated, but not so much that I had a pounding headache, and felt very tired and shaky, and slight nauseous. I put it all down to a hangover, had something to eat and decided not to take my meds.

I had been advised by my diabetes doctor when Xigduo was prescribed that I should not take it when sick. I was told that it continues working, dumping excess glucose via the urine, even when a person is not eating well and is dehydrated due to illness. This, I was told, can lead to further dehydration and other complications. I decided that my hangover counted as ‘sick’, so I decided to skip my meds at that time.

I drank a pint of water and went back to bed for a while. I woke up at around 9:30pm on New Year’s Day and again felt a little dehydrated, tired and shaky, and generally ‘felt weird’. This was unusual for me given how long it had been since I had my last drink the night before. I decided to test my BG and Ketone levels.

At around 22:30 my BG was 7.8 mmol/L and my ketones were at 2.4 mmol/L. I had been told by my diabetes nurse that my ketones should be 0.6 mmol/L or less. I had tested my ketones daily in the past while I was put on another medication (Gliclazide) for a week. I was dieting at the time, losing weight fast, and as a result my ketone readings were generally at the high end of normal, between 0.4 and 0.6 mmol/L. One time it was 0.8. I had tested my ketones only once before while on Xigduo and got a reading of 0.8. A reading of 2.4 mmol/L seemed alarmingly high. A quick Google led me to the NHS website page for Diabetic ketoacidosis (DKA) which states:

“under 0.6mmol/L is normal. 0.6 to 1.5mmol/L is slightly high – test again in 2 hours. 1.6 to 3mmol/L means you're at risk of DKA and should speak to your diabetes care team for advice. over 3mmol/L is high and means you may have DKA and should call 999 or go to A&E.”

Quite worried at that point I called the out-of-hours service associated with my GP practice and spoke to a nurse. She advised that I probably did not had DKA as I did not have enough of the associated symptoms, and that my ketone levels were likely due to the alcohol. She advised that I drink plenty of water, test my ketone levels before going to sleep to ensure they were not rising, and test them again in the morning. She also advised that if I suddenly felt worse, in particular if I started vomiting, I should go to the hospital.

I was reassured though I had serious doubts that a ketone reading of 2.4 could be due solely to a hangover. I ate a meal, decided to skip my meds again, just in case, and otherwise I did as I was told. I tested my ketones again at 1am and the reading was again 2.4, which seems to confirm that it’s a ‘real’ reading and not a glitch in the testing device.

Late this morning, January 2nd, I tested my ketones again and the reading was 1.3 mmol/L, still over double my normal readings from a few weeks ago but very much better. I decided I definitely did not have DKA and took my meds after breakfast. I tested again a few minutes ago at 18:20 and got a ketone reading of 1.8. I re-tested, in case this was due to a bad meter reading, and got a result of 1.4. This suggests that my ketones are steady or possibly rising slowly since this morning.

Does anyone else have experience of testing their blood ketone levels while on an SGLT2 inhibitor like Dapagliflozin? Has anyone ever tested their ketones after excessive alcohol intake? Has anyone seen a reading as high as 2.4 mmol/L either following excessive alcohol intake or in the normal run of things while on this medication? How worried should I be?

I had already planned to give up drinking for the month of January, though now I’m thinking it would be best to obey the advice leaflet and stick rigidly to 2 units of alcohol or less on any given day while on this medication. I intend to call my diabetes nurse tomorrow if my ketones don’t come down further and ask for advice, and I’ll bring it up with my doctor at my next appointment regardless, though some patient stories on this topic would be very much appreciated.

Thanks for reading.

 

[Antje77]

What have you been eating?
If you haven't had many carbs they could just be normal ketones from low carbing.

 

[HairySmurf]

What have you been eating?
If you haven't had many carbs they could just be normal ketones from low carbing.

Hi and many thanks for the reply.

Generally I've been on diet which has me in a calorie deficit of around 1200 calories per day with diet and exercise combined, calculated based on my observed weight loss of just over 1Kg per week, measured weekly. It's not a low carb diet per se but I have eliminated all meals that are very high in carbs. It's not any kind of formal diet, I just measured the calories in everything I was eating for 6 weeks, eliminated all the worst meals, and replaced them with better ones. This, combined with lots more exercise has been enough to do the trick, so far. It was on this diet that I saw readings of around 0.4 to 0.6 mmol/L ketones, measured every day for a week.

On New Years Eve however I ate more carbs, a few biscuits, and I ate around 75 grams of roasted cashew nuts late that night in addition to the three meals I usually have. I wouldn't put it down to food alone certainly. Alcohol, as I understand things, causes blood sugar to rise for a while but can then cause low blood sugar for a period lasting up to 24 hours. This then causes ketones to rise, though I have no idea if it can cause ketones to rise to four times my observed 'normal' levels.

 

[Guilty]

Does your doctor know about your diet at the moment? With the Dapigliflozin it may put you at higher risk of DKA.

Happened to me (with Empagliflozin).

Couldn't tell you what my Ketones were leading up to my DKA, as I didn't recognise it as DKA at the time. So didn't test. And they were off the scale to begin with at the hospital (just showed as HI).

Sounds like you are doing the right thing though, monitoring. And being aware of the symptoms.

Don't take any chances with it if you develop worse symptoms. But your nurse has already told you that.

And, if it makes you feel any better. On previous occasions I've had ketones of 4+ in the past from low carb / calorie restriction combined with Empagliflozin and was fine, no DKA.

 

[HairySmurf]

Does your doctor know about your diet at the moment? With the Dapigliflozin it may put you at higher risk of DKA.

Very many thanks for the reply.

Yes I told both the doctor and diabetes nurse on the day I was prescribed it. Neither raised any concerns, and when I mentioned to the nurse that my diet was likely to be raising my ketone levels she said 'well we certainly don't want you to stop your diet and exercise'.

To put that in context, I was given the blood monitoring device and a prescription for glucose and ketone test strips along with a one week prescription for Gliclazide. I was told that the Gliclazide was to confirm that I had type 2 diabetes. This makes sense as that medication only has its intended effect if the pancreas is capable of producing insulin. I was asked to test my BG twice per day, before morning and evening meals, and ketones once per day, then report the numbers to the nurse over the phone a week later. Once she was happy with the results I was given the ok to start Xigduo. If not for the need to do that testing I'm not entirely sure that I would have been given a blood monitoring device at all.

 

[Antje77]

I was told that the Gliclazide was to confirm that I had type 2 diabetes. This makes sense as that medication only has its intended effect if the pancreas is capable of producing insulin.

This doesn't make much sense to me. In adults, the pancreas keeps producing insulin (just not enough) for a fair while with T1 as well. Gliclazide is not used as a diagnostic tool for insulin production as far as I know.
With a flozin into the mix, it will tell you even less.

If your doctor or nurse wants to know about your insulin production, the usual way is a C-peptide test, which shows how much insulin you produce.

 

[HairySmurf]

This doesn't make much sense to me. In adults, the pancreas keeps producing insulin (just not enough) for a fair while with T1 as well. Gliclazide is not used as a diagnostic tool for insulin production as far as I know.
With a flozin into the mix, it will tell you even less.

If your doctor or nurse wants to know about your insulin production, the usual way is a C-peptide test, which shows how much insulin you produce.

I don't know what to tell you. It's possible I misunderstood the reason for the seven day prescription with accompanying self testing. There was a whole lot of info thrown at me that day, but I tend to have a good memory.

I was initially prescribed Metformin 500mg x 3 per day by my GP. The diabetes specialist put me on Gliclazide for seven days and at the same time gave me a six month prescription for Xigduo, to begin after I reported blood numbers to the nurse. I was not advised to keep up the daily testing after day 7 (I still don't know whether they want me to test every day). There was no overlap between the Gliclazide and Xigduo during the test period.

I do remember the doctor saying something like 'If you have type 1, this will give you a tummy ache' or something to that effect. If not for the purpose of verifying the type of diabetes I have, is there any logical reason for the 7 day prescription? It seems bizarre to me to do that and ask for blood self test results unless it was a test of some kind.

 

[Antje77]

I do remember the doctor saying something like 'If you have type 1, this will give you a tummy ache' or something to that effect. If not for the purpose of verifying the type of diabetes I have, is there any logical reason for the 7 day prescription? It seems bizarre to me to do that and ask for blood self test results unless it was a test of some kind.

I have absolutely no idea!

Or maybe, but this is pure speculation, their thinking was along the lines of ruling out you don't produce insulin at all. Gliclazide forces the pancreas to produce more insulin, which lowers blood glucose. If you can't produce insulin at all, gliclazide doesn't work. So maybe, just maybe, they used it to see if you are at least producing some insulin. If you wouldn't produce any insulin, in a way you would develop a 'tummy ache', which is a giant understatement to describe DKA, but yes, a tummy ache is one of the symptoms.
Which would be pretty creative thinking, but who knows?

It's a useless way to tell T1 from T2, with the exception of having a very quickly developing T1. Which isn't common in adults, we tend to slowly lose our insulin production.

 

[HairySmurf]

Update: My blood ketone readings did fall back close to what I consider my current normal - 1.5 mmol/L yesterday (the 3rd) and 0.8 today.

My diet and levels of exercise haven't changed much, so the ketone spike seems very likely to be due to alcohol intake alone or some interaction between alcohol intake and/or dehydration and the Dapagliflozin.

Interestingly I believe my rate of weight loss has increased substantially, which may explain why it took 3 or 4 days for my ketone levels to drop. This might have had a bearing on the spike, and is presumably due to the glucose dumping effect of this medication.

Keeping calm and carrying on.

 

[HairySmurf]

This doesn't make much sense to me. In adults, the pancreas keeps producing insulin (just not enough) for a fair while with T1 as well. Gliclazide is not used as a diagnostic tool for insulin production as far as I know.
With a flozin into the mix, it will tell you even less.

If your doctor or nurse wants to know about your insulin production, the usual way is a C-peptide test, which shows how much insulin you produce.

Just updating this thread in case my post about gliclazide as a test confuses anyone who comes upon it.

I had an appointment with my endocrinologist today and I asked him what the purpose of putting me me on gliclazide for a week was. It wasn't any kind of rough test for T1 - there was a medical purpose. I gathered that pushing down my blood glucose with gliclazide reduces levels of glucose toxicity that pancreatic beta cells are exposed to. Basically it seems that lower BG levels, as close to normal as possible, helps a dysfunctional pancreas recover a little, increasing the chances that the Metformin and Dapagliflozin combination (Xigduo) would be sufficient to keep my BG levels down after that point. I assume that the purpose of testing my BG twice per day and reporting those numbers to the nurse before starting the Xigduo was to verify that the job had been done, that the dose of gliclazide was sufficient to produce a week of near-normal BG levels, and possibly to decide whether Xigduo would be enough medication to go on with at that time.

It's probable that the only reason he explained that to me in any detail was I had just asked him about glucose toxicity in the context of T2 remission. My last HbA1c in January was 39 mmol/mol, while on the Xigduo, down from 89 mmol/mol last October. There's no way that particular medication could do that by itself and I'm eating 150 to 160g of carbs per day so it looks like my weight loss is having a big effect; that my dysfunctional beta cells are recovering. He offered me Trulicity or Ozempic, if I wanted it, despite my low HbA1c. I asked if taking Ozempic along with the Xigduo would help dysfunctional beta cells recover faster by keeping BG control as tight as possible and thereby reducing levels of glucose toxicity. He thought about it for a second and said there are no guidelines on this. I was hoping for a little more than that - some guidance on the best way to transition from meds to remission after losing a load of weight. He said that as I'm experiencing no side effects I should keep taking the Xigduo, that the decision about Trulicity or Ozempic was up to me, and that the only reason he'd recommend taking it on top of the Xigduo would be to help with weight loss and help in maintaining a lower weight. I chose to stick with the Xigduo, at least until my next appointment (likely in six months) and refused the Ozempic. No technical remission for me any time soon.