Type 1 & the Big "M" - Night Sweats and Hot Flushes (Part 1)
If you ask anyone to think about words that they associate with the menopause, then I am guessing that the top spots will usually be occupied by “night sweats” and “hot flushes”. For the purposes of writing about them, I will describe my experiences of both. Many people describe the physical reaction to both as being the same and the name purely reflects the time of day when they occur. I have found however that my physical response to both is slightly different.
My night sweats have occurred in the very early hours of the morning. They remind me of the water pistol fights I would have with the kids on very hot summer days in the garden. They would get me so drenched (their aim was always better than mine!), that I would have to change out of clothes that were so wet they stuck to my skin. Night sweats would leave me equally drenched. The sweating has been sufficiently extreme that nightwear and bed sheets both look and feel obviously damp. It is severe enough that I can’t just roll over and go back to sleep. I have to get out of bed, change nightwear and roll back the duvet to allow the sheets to dry sufficiently before I can contemplate trying to resume sleep.
Hot flushes make me think of the dial operated fan heater that provided the heating in my first flat. It can feel like someone is playing with the dial, turning it up and then back down at random, for varying lengths of time. I glisten whilst the strength of feeling of internal heat can rise quite quickly and I need to shed layers of clothing or roll back the bedclothes, to cool down. In the earlier stages of peri-menopause, they tended to happen most commonly at some point between the hours of 9pm and 3am. They could last for an hour or continue on and off for a period of several hours.
Night sweats have been relatively rare. I have probably had only six or so episodes. Hot flushes however have been a very regular companion. They were more spread out in the early stages of the peri-menopause. I could go for weeks or a couple of months without any symptoms and then have regular symptoms on successive nights with maybe a few days respite before flushes returned. Again, this could last for several weeks before returning to several weeks or months flush free. Hot flushes are what cause the greatest problems. I cannot predict when they will happen or how long they will last and they cause blood glucose mayhem along with disturbed nights. Although not yet fully menopausal, as I approach this next stage, hot flushes are happening more frequently, can affect me now during the day as well as at night and can last significantly longer.
So, what have been some of the frustrations in managing the above?
The seeming reluctance from some HCPs (and it is only some), to recognise that even with type 1, we will be equally prone to night sweats and hot flushes and they will not mean we are suffering the symptoms of a low blood sugar,
Both night sweats and hot flushes have had a pronounced impact on blood glucose behaviour,
The methods I have found that work for each are slightly different and will generally go against HCP recommended behaviour in living with type 1, but that doesn’t make them wrong. It just means they are untested in any sufficiently large population of peri-menopausal women to give confidence to them being recognised as an acceptable and safe technique to use…. and you need the support of technology to be able to use them safely.
Hypo v hormones - the cause of sweats
Having self-funded CGM for almost five years, I am aware of blood glucose trends when these symptoms occur.
I have suffered a sweating response to hypos in the past and tend to find that for this to be my reaction, my blood glucose will be below 3 mmol/l and continuing to fall. When stable above this level, I don’t tend to suffer a sweating response. And the combination of other symptoms when I am low – feeling hollow and tired, the ability to only focus on myself and the internal crankiness I feel if demands are being made on me by either people, their conversations or other events, due to the lack of energy to be able to do anything other than focus on treating the low – make it very clear that I am low.
A sweating response with a night sweat is very different – I feel mentally and physically coherent (albeit tired), no brain fog or exhaustion, just hot and sweaty!
And even for any HCPs that still believe the symptoms might be hypo related, the trends and alarm function on my CGM make it very clear that this is not the case.
Blood glucose response
My blood glucose levels rise with both night sweats and hot flushes. It has been suggested that this rise might be as a result of the fat or protein content of evening meals, continuing to impact blood glucose levels through the night. I appreciate that it is difficult to ever isolate one particular factor causing blood glucose behaviour but again, I am inclined to rule this out. I do not eat a high fat evening meal and I aim to finish dinner three to four hours before going to bed, which gives a good idea of the food and insulin response post dinner. We tend to stick to a similar selection of evening meals on a weekly basis and there is no obvious pattern with the food eaten to the blood glucose behaviour during particular nights. Lastly, the symptoms of night sweats and hot flushes make it very clear that this is what is happening when you are going through them.
I set alarms on my CGM to alert me when my blood glucose levels reach 8 mmol/l. I often find that if I have gone to bed with an in range blood glucose, if I am woken with symptoms of night sweats or hot flushes, my blood glucose will be in the 8-9mmol/l range anyway.
Is this too low an alarm setting? If I am having issues on two, three or more consecutive nights, probably. When I get to that stage and exhaustion is setting in, I will invariably increase my alarm to 10 mmol/l or switch off altogether in an attempt to regain sleep and some sanity. However, I generally find it better from a management point of view to be alerted at this level rather than at a higher level of over 10 mmol/l, as the longer I wait and the higher the blood glucose is at the point of starting to address, the longer it will take to reverse. So although the lower alarm setting may lead to greater potential for interruption to sleep and alarm fatigue, when sleep is disturbed anyway with the hormonal symptoms, and if I would be awake for longer trying to reverse a higher out of range reading, I figure that this is the least costly behaviour in terms of coping with the repercussions.
So, both night sweats and hot flushes lead to an increase in blood glucose levels but for me this is where their similarity ends.
Blood glucose management of night sweats
By the time I wake up drenched, with my CGM alarming to state that blood glucose levels are over 8 mmol/l, they often appear relatively stable at this level, with the CGM trend showing a horizontal arrow as opposed to a diagonally or vertically upwards arrow.
Although I calibrate my CGM before bed, due to the rapid increase that hormonal fluctuations have on blood glucose levels, I always choose to do a self monitoring blood glucose (SMBG) for comparison. I will then plug my blood glucose reading into my pump and use this with pre-set insulin sensitivity factors (ISF) to administer a correction dose. I silence the alarm function on the CGM for a set time that I feel comfortable with and will attempt to go back to sleep.
Depending upon my blood glucose level at the point of giving the correction dose, I may find that my CGM re-alarms after the time for which I have chosen to silence it. This is likely to be due to playing catch up, with the time it takes for the insulin dose to peak and begin to work on a blood glucose already out of range. So, as long as my blood glucose is not significantly higher than it was at the first alarm, I will re-silence with no further intervention and return to sleep. This is usually sufficient to deal with any blood glucose excursion and more often than not, my blood glucose is closer to a more normal range by the morning.
Night sweats have been one-off events in a night. I have not experienced multiple episodes of waking with a sweat in the same night.
Blood glucose management of hot flushes
There are a number of reasons why hot flushes are much harder to manage. They have lasted for varying amounts of time, anywhere from one to six hours, so the length of time the counter-regulatory hormonal response will last cannot be predicted when starting to correct. In my experience, correction doses based on ISF have been completely inadequate in returning blood glucose levels to a normal range. Not only must consideration be given to the need for more insulin to correct an increasing blood glucose value rather than one that is stable or declining but when playing catch-up, it will take longer for insulin to start working to stabilise higher blood glucose levels and return them to a normal range. So balancing this understanding with the knowledge that a more aggressive correction dose is likely to be required to manage the on-going hormonal response adds to the difficulty in determining an appropriate correction dose.
As with night sweats, I will perform a SMBG to compare CGM readings to blood glucose levels. Alarming for values over 8 mmol/l, the CGM is far more likely to be showing a diagonally upwards arrow. When I first started experimenting with correction doses I would limit the amount given to a dose based on my ISF. I would silence alarms for a pre-set time and go back to sleep. It became obvious over the many episodes and my experimentation on myself, that this was never enough. My CGM would alarm again at regular intervals throughout the night, depending upon the length of flush. However, due to the risks associated with correcting too aggressively overnight, I remained cautious in trialling different ways of managing this.
The concerns centre around insulin stacking. “The simplest definition of ‘stacking’ is to take a rapid acting insulin dose while another previous insulin dose is still working thus causing an overlap in the insulin effects.” (Ponder and McMahon, 2014, p200). Given that insulin can peak anywhere from 45 to 90 minutes and last between 2 and 8 hours, the concern is that doses given in relatively quick succession will eventually lead to hypoglycaemia. However they also point out that smaller insulin doses have both a shorter peak and duration of action than larger doses. (Ponder and McMahon, 2014, p196). I want to make it clear that Ponder and McMahon do NOT recommend taking overlapping insulin doses at night recognising that people tend to be more sensitive to insulin in the overnight hours. It has become obvious however in considering how my blood glucose levels respond to hot flushes and following their education around dynamic diabetes management, many of the tools they discuss in relation to matching insulin doses to a late rise in blood glucose as a result of particular foods eaten seem to have equal relevance for me in managing any prolonged hormonal impact on blood glucose levels.
With the use of CGM, alarms and a willingness to jump in to treat any potential over correction, I have experimented and accepted the greater textbook risk with administering a several overlapping micro-bolus doses when managing hot flushes. So is this insulin stacking or therapeutic accumulation?
I have shown an example from a CGM graph.
I woke up with hot flush symptoms and my CGM was reading 7.2mmol/l with a horizontal trend arrow. From my experience, to be woken before my CGM alarmed was an indication that this was likely to be a more prolonged flush but I still felt the need to remain cautious in choosing a correction dose.
I gave myself a 1u correction dose at 3.27am.
At 3.41am my CGM alarmed to say I was 8.1mmol/l with a diagonally upwards trend arrow. After only 10 minutes I would not usually contemplate giving myself a second correction dose so close to the first but I was desperate for sleep. This followed on from several nights of disturbed sleep and I banked on my quick risk assessment that this night was going to be more of the same. I recognised that the first dose would not have had any time to start working but based on my experiences on the last few nights and over previous years, my gut reaction was that my initial 1u was likely to be sufficient and so I gave myself an additional 1u dose and silenced the alarm for a pre-set time.
I woke up 5 hours and 20 minutes later, to a fasting blood glucose of 5.2mmol/l at 9.05am.
So is this due to a poorly set basal rate? By definition, I suppose the only answer is yes? If my basal rate does not match my body’s requirements on any night, it is not a good pre-set dose. However, when the same pre-set doses can provide the overnight graph below for a night without hot flushes, how do you manage this?
So how many micro-bolus doses would I give?
Like all type 1 management, it is not formulaic, and what is right for me will not be right for you. I have tended to give a maximum of two micro bolus doses that in total are the equivalent of double the correction dose recommended by my ISF. Apart from the above two corrections given only 10 minutes apart which are in effect the same as giving one correction dose of double what my ISF would prescribe, I have not previously wanted to administer all in one go. Even with a CGM, the constant awareness of the risks of nocturnal hypoglycaemia is ever present.
Even if I re-alarm after these two micro-doses, I will rarely give any further doses within two to three hours, as if blood glucose levels appear to be stubbornly refusing to reduce, it may just be a matter of waiting. In the worst cases, I will then find my fasting blood glucose levels remain at a level of 8 -10 mmol/l. However, in writing this what has become obvious to me is the impossibility of coming up with a set of guidelines that always works. (Much like with anything to do with the management of type 1!). This will become apparent if you follow through to part 2 (Blog 6) where I deconstruct a hot flush, taking you through my CGM graph for a bad night and the decisions taken at each stage.
Combining temporary basal rates with micro-bolus corrections
I have been asked why I don’t combine an increased temporary basal rate with a correction bolus at the start of a flush. My maximum basal setting is currently programmed at 0.55u per hour so to get the same effect as say a 1u micro correction, I would need a significant percentage increase. I find a temporary basal rate initially takes too long to begin to work against blood glucose rises and by the time it starts working, chances are that the hot flush has passed and so I will end up with blood glucose levels falling dramatically from the percentage increase that is finally beginning to work and now significantly higher than required. This has caused me to spend more time managing low blood glucose levels for a significant time after.
So, I have thought long and hard about whether I should be specific about my management of hot flushes and the use of overlapping insulin doses. I do not want others to adopt this as a blueprint. It would be wholly inappropriate to consider copying all that I do, as my physiology and lifestyle are different to yours. You will need to find your own workable techniques. However, if going through this stage of life yourself, I hope you have greater opportunities to discuss suitable techniques (particularly if they go against traditional textbook teaching) with HCPs. And PLEASE, if you choose to experiment, take sufficient time and care when trialling procedures that may work for you.
Ponder, S.W. and McMahon, K.L. (2014), “Sugar Surfing: How to Manage Type 1 Diabetes in a Modern World”,
I am writing a series of blogs about my experiences of managing type 1 as I approach the menopause. I hope this will help others to understand the range of changes that may affect them. However, I am very aware that my experiences may well be very different to yours…. So please use this for ideas, discussion with HCPs and not as gospel. Please talk to your team before experimenting with techniques that have worked for me. Please experiment slowly and safely, to see how you react when trying to find techniques that may work for you. It has taken me nearly five years of self-funded CGM use and trial and error, to come up with my tool kit. Even then, it never works all of the time!