Type 1 Keto

Covid, Keto and Insulin

Covid, Keto, and Insulin Through the Roof

My personal Covid experience.

This is about how Covid provides opportunities for learning about Insulin in Type 1.

‘Have you had Covid?’ people would ask?

‘Of course not, I have a keto cloak of invincibility!’ would come the reply.

Doctors above all should know that such puffed-up confidence is never likely to end well.

I expected that it would be bad if and when I did get Covid and likely finish me off. Bravado is never a proxy for certainty. There are so many health benefits of keto lifestyle in general. Especially the  improved glucose control that reduces complications of Type 1. In theory Keto should help me against Covid. Ketogenic lifestyles, even in Type 1, reduce insulin needs, and so down-regulate the cellular receptors that Covid latches on to. Keto reduces inflammation and, should confer some protection against Covid. It’s only my theory, but it’s logical. But there might be other mechanisms that exert a counter effect and are  driving  bad outcomes. Who knows? But I felt well prepared for this fight partly because of the keto lifestyle. I am also triple vaccinated against Covid.

But I don’t get ill these days (there I go again!) I recover quickly from injury, and overall, this keto lifestyle suits me. It does help my T1 diabetes control. And, let’s face it, glucose control is so important in diabetes of all types. Glucose control is an essential determinant of day-to-day well-being and also of long-term complications. Keto makes Type 1 management easier and safer.

In very stark detail, a sobering paper presented at the EASD in 2020 showed how important good glucose control is. It turns out that for every year that HbA1c is above 7.5% or 58mmol/mol, 100 days of life are lost. Over a lifetime from diagnosis, that adds up to 8 years on average.


So, if you start counting from the 1st of January, you will lose every single day from then until the 10th of April. Or the 9th of April if it’s a leap year. Personally, I could miss January, the first two weeks of February and November, but I don’t think this paper says we get a choice!

Now, at a Public Health level, this might just about be acceptable if a few people with complex needs were poorly controlled to this level. We could focus on specialist teams to work with this group. But, no. According to the Diabetes audit for England and Wales 2017-2018, a whopping 70% of people with Type 1 diabetes are in this situation.

Seven out of every ten people are losing a third of every year they live. The only saving grace is that we don’t know how long we have left.

And this figure hasn’t changed for years. Is anyone bothered? Yes, I am. Because research on keto lifestyles has shown that HbA1c can be normalised such that the vast majority of

people can get their 100 days back. In 1752 it is rumoured that people rioted to get just 11 days back when the Calendar was adjusted as a one-off. Our loss of 100 days is every year!

Raised glucose as the first sign of Covid.

One morning, my blood glucose gradually rose and plateaued at around 11 mmol/l for no explicable reason. I felt well on waking, but I had symptoms of headache, mild sore throat, and light dry cough over the next  few hours. A Covid lateral flow test was positive as was the follow up PCR. The rest of that day was spent feeling more tired with some myalgia. My glucose levels had been a Covid early warning sign.

I was already on a keto diet. My appetite for food had gone, but my sense of smell and taste was usual. So, despite around 50% fewer calories and increased bolus injections of Novorapid, I could not drive the glucose down. This was the situation for the next three days. It seems intuitive to want to correct glucose with Insulin.

My usual total daily Insulin is eight units of Tresiba twice a day and two units of Novorapid in the mornings. This goes in fasted to counter the dawn effect where blood glucose rises due to stress hormones. Typically four units for evening meal, on a 16 to 18 hour fast and two meals a day strategy. Therefore, I use around 22Units of Insulin, give or take, every day. And this level of Insulin keeps a good glucose profile to keep my HbA1c within the non-diabetes range.

On day one of Covid, I injected  12 units of Tresiba in the morning and 14 units at night, with 4,6,6,6 units of Novorapid throughout the day, sometimes without food. 48 units in total.

My day 2 was 14 units of Tresiba twice a day, six units of Novoapid fasted, then another eight units 30 minutes later. Overall, on day 2, I injected 54 units of total Insulin, close to that volume the next day, 38 units on day four and today on day five; it looks like around 28 units. Today the glucose will be back to normal. In fact, on reflection, I wonder what would have happened if I had injected my usual volumes.

My glucose eventually settled into the normal range, but my Covid test remained positive for another 48 hours. The CGM graphs nicely show the run in to Covid and the recovery.  You can immediately spot the crazy lines when the glucose sensor was malfunctioning; it was then a manual reset. I tested several times a day with a finger-prick device because of sensor issues

Continuous Glucose Meters are indispensable but not without flaws.

Apart from cold symptoms and tiredness on day 1, I was slightly better on day 2; I was 70% better on day three, 90% recovered by day four, and symptom-free on day 5. My glucose told a different story by 24 hours; it normalised on day 5.

I felt it necessary to dust off the ketone meter. I rarely use it these days. I was in and out of ketosis, but mostly out. On one occasion, I had a glucose of 16.6 mmol/l  and ketones of 0.8. Hospital admission levels. I could detect the glucose level physically.  After a few hours the glucose drifted down to its plateau of around 12mmol/l,  so I elected not to retest ketones.  But if both had kept rising after a few hours, the story might have been different. I might have been entering Diabetic Ketoacidosis (DKA). DKA is a risk for all Type 1. Especially during infections like this. But you will not go into DKA as a result of a keto diet.

Day 1, run in to Covid. Red dots are calibration measurements using capillary device.

Symptoms: Headache, mild sore throat.

Insulin total 48U

Ketones 0.3mmol/l

Day 2. Despite high volumes of rapid insulin and reduced calories on a keto diet, insulin resistance is clearly demonstrated.

Symptoms: Headache less, tired, dry cough, sore throat gone.

Insulin total 54U

Ketones 0.3 mmol/l

Day 3. Sensor failure! Plateau glucose continues, worse if anything.

Symptoms: Tiredness easing, dry cough. Headache resolved.

Insulin total 54U

Ketones 0.8 mmol/l

Day 4. Frustratingly, the new sensor is slow to stabilise. Calibration red dots show reducing glucose.  Signs of hope by the evening.  At this point great care needed with rapid insulin dosing. Insulin resistance has largely resolved.

Symptoms:  Minimal Covid Symptoms.

Insulin total 38U

Ketones 0.2 mmol/l

Day 5. Return to normal . Sensor still playing up overnight and morning reading shows there was unlikely to be a hypo in the night. Certainly no hypo symptoms.

No Covid symptoms

Insulin total  28U

Ketones 0.2 mmol/l

Day 6. Normal day.

Insulin total 26U

Ketones 0.5mmol/l

So, what on earth is going on here? An almost tripling of Insulin with little effect on control. I mean, for me, one unit of Insulin reduces glucose by 2 mmol/l, yet here I was shooting up 14 units with no food cover and not a flicker on the glucose needle. This is acute insulin resistance caused by inflammation caused by Covid infection.  If you have Type 1 diabetes, you will have noticed the same effects of insulin resistance with constant snacking, tiredness and mental stress. The glucose goes up to 11-12 mmol/l and refuses to budge. It seems that whatever metabolic processes are happening, Insulin is not working. Or if it is working, then it has reset the glucose threshold and parked it there for some reason. There are no wild swings of glucose is a stubbornly flat trace when one considers the volumes of insulin injected.  You might notice that this level is just about on the level at which the kidneys fail to recover all of that filtered glucose and at which glucose then spills into the urine. Is that a protective mechanism against rising glucose? A type of diabetes drug called an SGLT2 inhibitor exploits this mechanism after all.   I had no means of checking this as I had no urine glucose strips. However, in the absence of carbs, the glucose rise can only be through glucagon. Glucagon is a counter hormone to Insulin and raises blood sugar. Glucagon also responds to stress hormones  released during infection. Start adding carbs to this situation where Insulin is not that effective, and you can see how all of this is going to get out of control. The system’s capacity to self-regulate will be overwhelmed by glucose from carbs and that from stress hormones. It would likely keep on rising.

Of course, Insulin doesn’t just have a role in glucose control, though that is one of its significant functions. Insulin interacts with at least 23 different hormones at my last count! Glucagon is the obvious counter hormone to Insulin. Insulin lowers glucose, and glucagon raises it. But thyroxine, oestrogen,  testosterone, vitamin D, growth hormone, cortisol, adrenaline, and many digestive hormones interact with Insulin and affect how it is expressed and vice versa. It’s a right old chemical soup but as tightly regulated as a Swiss watch. You can see that Insulin has many interactions. Pouring in huge glugs of the stuff without respect for these subtle interactions isn’t going to help. Insulin also leads to the production of pro-inflammatory signalling molecules called cytokines. Cytokines perpetuate inflammation, and stress hormones continue to rise. Glucose rises, Insulin doesn’t work well. What this is doing to all the other hormones is anyone’s guess! It is all code red until the immune system kicks in or the inflammation stops. Lobbing in a few carbs at this point is surely provocation? 

Not Covid, but six insulin molecules around a zinc atom. It’s normal storage state. It’s very pretty and you need more in infection.

Type 1 diabetes helps us explore how this insulin requirement changes during infection. It was surprising that I used almost three times the usual dose of Insulin to maintain, let’s face it, rubbish control. I expect this type of metabolism is happening in everyone, diabetes or not. Lack of Insulin and the need to inject just highlights it. Those who do not need Insulin because they don’t have type 1 diabetes will never know. Type 1 might be of use after all! We are the lab rats with no insulin secretion.

Finally, at some point, I knew that I would recover from the infection. And I might become insulin sensitive very quickly. At this point, and it can happen in hours, any excess insulin will drive down the glucose from its plateau. And it can be dramatic from experience. I was injecting large volumes of Insulin and getting nowhere and had to bear in mind that a massive hypo might be imminent. As soon as I felt better, I knew it was time to start tapering the insulin injection. I would have done this also if my glucose trace had normalised. Being ever vigilant but with patience was the key here. A CGM is so helpful. I was safely timing my rapid insulin dosing to ensure the previous dose had worn off before reinjecting. Three hours to five hours is quoted, everyone will have their own figures. I think that if I had day three again ( and I’m sure that I will get the chance someday!) I would not be quite as bold with insulin dosing. Insulin wasn’t working because the trace was flat whether I chose to inject or not. You might question the need to increase the Insulin at all? I guess it depends on whether the glucose plateaus or rises. I was reacting out of habit, plus I wasn’t in the best frame of mind with a headache and a temperature. Excuses, excuses! But i do wonder what would have happened if I maintained my usual dose. That’s for then next time! I never used more than 2 unit corrections at night—big doses when asleep would have been reckless. When I stacked fourteen units I knew I would need around 140 g of carbs to cover that if I wasn’t ill.  But, I knew that I was Insulin resistant on day three, so I had more safety. The CGM graph was confirmation. Had I not had an infection and glucose was raised, I would have taken a different approach: vigorous exercise and hoover the glucose up by that route. This would by-pass the long process of waiting for the gut and liver to get their act together. I’m not sure it works in inflammation. And also,  when ill, it is something you just don’t want to do.

The crisis is over, and it is now back to the usual round of fasting, keto diet, physical activity, daily monitoring, and daily injections. Given the complexities of Type 1 diabetes management, one would think that some sort of competency would be required before people were trusted with ownership of this disease. For one, even as a medically qualified person, I would agree with that statement. I would have also failed the test. It is not in my nature to be so thoroughly organised or disciplined to be trusted with Type 1 diabetes daily. But like most of you readers, I have got it and have to get on with it. I get through each day mostly unscathed. I’m still here. I can still refer to the Frank Sinatra song ‘I did it my way’ in the present tense. We all do it our way. So please take the following blog as a personal account. I hope you pass his on to your clinical teams. Some individuals still have learning needs around keto management of diabetes. It will help them to understand our lives. And rest assured, many of those would fail the Type 1 test, too. I don’t know how others manage. Please share your story. Every one of the 400,000 people in the UK with Type 1 diabetes will have a different story to tell of how they manage infection and the insulin demands on the body that goes with it. Those with kidney disease will have a vastly different experience. Those whose diabetes has reduced glucagon sensitivity will have another story to tell. As will those on day 1 compared to those on day 10,000. I hope my account will encourage you to tell yours.

12 replies on “Covid, Keto and Insulin”

Hi Ian

It’s dean Johnson here we had a consultation early in the year about running and diabetes management. Just read your blog and it is nearly a carbon copy of what happened to me I was unwell and injecting large numbers of insulin with no movement on levels. This recovered after about 2 days of this. In other news I am now on levimar long acting insulin and starting to up the miles I am doing 10k today which will be my normal pattern from now on. I am aiming for marathon distance by April or May time and am fully keto.

Keep up the good work

Hi Dean, Thanks for your story. I assume your half marathon will be with your son on your shoulders! It gives you an incentive to get it done sooner rather than later. Please let me know how you get on.
Best wishes,

Thank you for sharing your experience with Covid as a T1. I very am sorry you had to experience this, but it gives me good insight and knowledge for what to expect and perhaps even helps me know how I can prepare if it happens to me too.

Hi Gina, I was disappointed my keto cloak slipped. The symptoms generally with this variant seem mild and it’s good to know that I fared no worse than non T1. Keto or not I don’t know, but I’m sure it didn’t make things worse. And of course with Keto one gets the benefit of good control the rest of the time. Hope you are well.

Thank you so much for this, Ian. I do hope medical teams are equally prepared should T1s need hospitalisation. From experience of illness requiring hospitalisation during pregnancy, aged 27, I remember the hospital team using a sliding glucose:insulin scale over the weekend and watching, mortified, as my carefully controlled bg levels rose hour on hour, with my precious cargo on board. Sadly, it took a lot of raising of voice and demanding my own injections back to alert the DNS who came in, out of hours, and, not without some angry urgency, spoke sternly to the nurses. Thanks to her swift intervention, we didn’t die and my baby is now a healthy, fit 27 year old himself. It could have been so different. Thanks for sharing the heads up, and so glad you are well.

Thank you for this very interesting account of dealing with Type 1 and Cofid. I agree with you in feeling that people need some sort of qualification to be entrusted just with Type 1 diabetes! How crazy is that!

Thanks for the update. Very interesting. Is a LADA/T1 diabetic in honeymoon I dont take any insulin. Its been 7 years since diagnosis and started low carb then. Intermittent fasting for the past few years.

When Covid started I was worried that I might suddenly need insulin and so requested some. (I was given it and instructions 7 years ago). Doctors were worried that I wouldnt know what I was doing which is fair. However, I thought that I didnt want to be trained in a panic when my mind was fussy with a headage and illness. Eventually, they gave me it and some refresher training over the phone. Thankfully I havent needed it yet as, to my knowledge, I havent had Covid (yet).

Brilliant news, Rob. I hope you are well. Is this a honeymoon or are you now in a stable state? It’s a long time. Only your biomarkers will tell you. Lets hope its the latter.

Leave a Reply

Your email address will not be published. Required fields are marked *