recent insulin study

[Sid Bonkers]

Hi Sid,

Thanks for your thoughts. Unfortunately (for us insulin users!) I don't think that is the most likely interpretation of the study results, for the following reasons: 1). people in all comparison groups were 'incident cases' (that is, the authors used a criterion that to be included people had to be recently diagnosed and diagnosed for the first time), so in theory at least they won't have been people who had failed to comply with any therapy for years and gone through an escalation of medication...............

Hi, for the sake of space I have only posted a small quote, I am not trying to quote you out of context or anything

I dont understand your answer really as if all the teat subjects were newly diagnosed and as you say not those who have disregarded advice for years how can any conclusions be drawn from the data? I thought that intensive insulin treatment for T2's was a relatively new idea so I dont see how any long term data could have been collected.

I was placed on multiple daily injections (mdi) plus 4 x Metformin after diagnosis I started out using 10 units of Humalin I at night to stop my insulin rising over night and 10 units of humulin s before meals, this quickly escalated to 12 units across the board to get my bg to what I considered to be a safe level, then as I started to lose weight I slowly reduced my insulin accordingly until roughly 12 months later I stopped using insulin altogether and just relied on the Metformin to hold my bg in the 5%'s (HbA1c)

After another 4 years and after losing another stone or so I have reduced my Metformin to just two a day now, so I conclude that early intensive insulin treatment helped me a great deal.

I wonder if I am included in the data you quote? First 12 months of intensive MDI treatment and then the following 4 years without problems, I can only say what I see and for me insulin has almost certainly had the opposite effect for me at least, but I was very pro active, I lost the weight that I needed to loose and have maintained that weight loss, I did not increase the fat in my diet so my trigs have stayed low .5 at the last check as I remember and all my other markers have been in range for the last 5 years.

Heres a pic of the first page of my insulin log book from 5 years ago when I started insulin treatment in hospital after diagnosis, notice the first recorded bg level of 14mmol/L. The hospital had reduced it down from 29 mmol/L with an insulin/glucose IV drip.

My initial target (top right) was to get my bg level to under 10 mmol/L which I managed in 5 days.

Been off insulin now for 4 years so when used as an initial intensive treatment I found it to be very effective indeed.

 

[noblehead]

Hi Noblehead. I've been a type 1 diabetic for 20 years and moved to a low carb paleo template diet a year and a half ago. You are absolutely spot on about insulin requirements. Initially when I moved over to low carb, my insulin requirements went as low as 15 units tdd from about 45 units on my former diet. However, after a period where my body adjusted to a fatty acid metabolic pathway and the consumption of nutrient rich food, this low total promptly changed and sits at about 30 units daily- 12 units of that is my basal (levemir). Included within this tdd, however, is the need to inject for green tea and coffee as caffeine raises my blood sugar one hour after consuming it. So again, for me, you are spot on with the amount of injections I take - typically around 7 a day on average. I have contemplated giving up caffeine but...nah I've got to have some vices.

Anyway, with respect to my insulin requirements, it took a while to understand why, but, after a lot of reading, the increase in my insulin needs is largely to do with the insulinogenic nature of the foods I now eat and the bodies adaptation to this style of eating. For example, on a low carb diet, there is a high glucagon response from the body whenever I eat a meal that has protein in it. The body requires insulin to allow the amino acids from protein consumed to enter the cells otherwise they can't be used. Part of this process involves pumping out glucagon to prevent the person eating this diet from suffering hypoglycaemia from the insulin response to the amino acids. Of course as a type 1, I don't produce insulin and so need to inject accordingly, otherwise my blood sugars go high because of this glucagon response. This glucagon response does not happen if I were to revert back to eating a high carbohydrate meal (as glucose is consumed with the insulin there is no risk of low blood sugar in a non-diabetic and no need for the body to pump out glucagon). So, bottom line, for me, is that I calculate for 40% of the protein that I eat and bolus accordingly. Interestingly, I don't seem to get that increase in blood sugars from protein consumption, after a couple of hours, that affects other type 1's on a low carb diet, and so don't need to double bolus for it. And I try and inject twenty minutes before eating as the glucagon response happens fairly quickly. This way I prevent my levels from spiking. And if there is a rise in blood sugar, it tends to happen very slowly. Same with lows.

I'm definitely in the camp where, for me, insulin is hugely beneficial (for obvious reasons), and I tend to dismiss a lot of studies because they are observational and do not take into account lifestyle factors, nutrition, stress, and any underlying inflammation within the body, which I believe is the main reason for diabetic complications (type 1 or 2 although the underlying conditions and risk factors are hugely different). Bottom line for me - inflammation, insulin resistance, and high blood sugars together = big, big problems. I'm less convinced that high(ish) blood sugars by themselves, in the absence of inflammation and insulin resistance is as problematic.

Many thanks for the feedback Craig and all makes perfect sense. Interesting what you say about green tea and coffee, although I do drink decaffeinated coffee now I can't recall it raising my bg, I do however enjoy green tea and drink 2 cups most days and not noticed any effect bg wise, it just goes to show how different we all are!

 

[smidge]

First off read the article and you'll see why those who VLCHF need quite a lot of insulin for the small amount of carbs they eat. Also, having been on this forum for several years and talking from experience from having low(ish) carbed before it's quite common for low-carb type 1's to have to inject 6-10 times a day compared to those who don't and inject 4-5 times, this is due to the slow rise in bg many hours after eating caused by the slow releaese of protein and the fat content of the meal.

Mmm...this is a difficult one to respond to really. I agree that relatively speaking, a low-carb diet will require a higher insulin:carb ratio than a high carb diet - but then carb ratios are not actually relevant and applicable to a very low or no carb meal.

I tend to go by a rough insulin:carb ratio, but adjust the dose depending on how much carb is on the plate in relation to other food types - I'm sorry, but i'm not explaining that very well!

As an example, 2 rashers bacon and 2 fried eggs has about 1g carb but usually needs 1.5 units Apidra - so that looks like a pretty bad ratio. A chicken breast and a plate of mixed veg with about 8g carb would need 2 units Apidra for me - probably 1 to cover the veg and 1 to cover the chicken I guess (although I don't think of it in that way), and a 100g plain yoghurt and 60g raspberries (about 9g carb) would need 1.5 units - maybe because there's no protein involved?. So, I think in a roundabout way, I'm saying that it seems to depend more on the size of the meal than the content when you low-carb - at least for me. I sort of use carb ratios as a starting point and then add on a small amount depending on the size and type of the meal. However, if i want potatoe with my meal, I would use the carb ratio and my actual dose of insulin would be very much higher. So, my chicken and veg example (above) is about 8g carb and needs 2 units - add 3 new potatoes (maybe 25g carb), and i'd have to add another 3.5 units Apidra to the meal, so 5.5 units overall.

I sort of see it as needing a minimum amount of Apidra just because I'm eating and then extra depending on what and how much I'm eating. I wish i could explain that better, but it works for me! It certainly isn't determined by a carb ratio.

Smidge

 

[mo1905]

Makes sense Smidge and you've explained it well. When I saw a dietician I was told have eggs and bacon or an ommelette or something and you don't need to inject but my levels still rise if I don't !


Sent from the Diabetes Forum App

 

[academicdiabetic]

Hi Sid,

Thanks for your reply and for going to the trouble of scanning in your hospital card. Regarding the study follow-up period, just to clarify, the follow-up was for three years, it was just that in order to capture all 'episodes' they selected people who had been newly diagnosed at the start point (to avoid the issue that you raised about the possibility of prior treatment/treatment 'failure' or other experiences having an impact on outcomes).

Regarding your history with insulin, I haven't said at any point that insulin doesn't lower blood glucose, obviously it does and lowering blood gluocse is the only outcome that any research looked at until very recently, now researchers are beginning to query what insulin 'achieves' in terms of subsequent morbidity and mortality. This trend, ironically given your experience, comes from the fact that the ostensibly 'good' outcomes from the DCCT trial for intensive early insulin therapy turned out on later analysis to be 'bad' (intensive therapy is now not recommended), so people have started questioning what insulin does other than lower blood glucose and the answer from the study I have been citing is that by the end of three years on insulin therapy (normal, not intensive), seems to be that the overall mortality of those on insulin (rather than metformin) is increased by around 2.5%, their risk of renal failure is increased 4x, risk of retinopathy and so on also increased, I can't recall by how much. None of the outcomes favour the insulin group (although it may well be that their end-point blood glucose was lower, I can't recall from the study now).

Obviously I don't know why you were in hospital, nor do I know why you were on insulin for a year with the blood glucose figures in the record card above (unless your figures were about double that or more without insulin) as I wouldn't have thought that would have been the correct approach. You are assuming (I think?) that the early intervention with insulin was what resulted in teh good outcomes on-going and up to four years later, but as you also put a lot of effort in yourself and lost (a lot?) of weight, surely it could equally have been that?

Insulin is one of the body's major hormones, as we all know, and hence it would be optimistic to assume that injecting amounts of it without perfect knowledge (which we can't have currently given the complexity..) would have only good effects. The new analogue insulins are significantly different to 'natural' insulin so ditto... I think the (surprisingly novel) focus on mortality and morbidity in the research literature (as opposed to solely a focus on levels of blood glucose without reference to any other desirable/undesirable outcomes) is going to be extermely valuable, but given outcomes to date I don't think it is likely to make those of us taking insulin very happy. Maybe it will and these early studies are just an aberration, who knows? Either way, I prefer having information to not having information so that I can act accordingly,

As you say, we are all different, so I wouldn't ever suggest a route for someone else to follow, 'diabetes' almost certainly isn't one disease (in fact currently its just a definition made, it seems, on rather shaky grounds) and even if it were, it would present differently with different people and the routes into and out of it would differ also.

Ho hum, I guess we all just have to get on the best way we can! At least having the opportunity to discuss our expereinces is positive!

 

[Dillinger]

To a point reducing carbs works and insulin usage does decrease (I know this to be true), however as we all know from personal experience and reading others on the forum that when you start to replace the carbs with too much fat and protein it can create more problems than it solves:

http://care.diabetesjournals.org/content/36/4/810.full

Whilst I do enjoy arguing with Noblehead that is actually a very interesting link; and one that I see sometimes with my diet. I low carb but wouldn't say I ordinarily go high fat but if I eat a 'low carb pizza' essentially cream cheese, linseeds, eggs plus pizza toppings then I will get a large delayed bump in blood sugars even when the pizza itself works out at about 5 carbs. Other than that slightly crazed fat based meal I don't see this though.

I don't consciously bolus for protein but as most of my meals are protein based with some carbs, so I bolus based on varying insulin/carb ratios through the day and kind of automatically count for protein I guess.

The thing is I think my control would be worse if I were to eat carbs as the highs and lows are just too difficult to control.

Back to the original post here's an article from the slightly crazy 'wanna buy a supplement?' Dr Mercola but it's interesting and on topic at least

http://articles.mercola.com/sites/articles/archive/2001/07/14/insulin-part-one.aspx

Best

Dillinger

 

[noblehead]

I sort of see it as needing a minimum amount of Apidra just because I'm eating and then extra depending on what and how much I'm eating. I wish i could explain that better, but it works for me!

No that makes perfect sense Smidge, I'm sure your method of bolusing for meals depending on portion size is similar to how SamJB boluses for his meals, if I'm not mistaken isn't this based somehow on the Bernstein method?

It's some time since I last had just bacon and eggs for breakfast but I needed to split my insulin and inject 2 units before and 2 units 3 hours postprandial, It's amazing how so few carbs still needs 4 units of QA as where I just need to inject 5 units for a breakfast of porridge oats, milk, seeds, yogurt and blueberries (45g of carbs in total). As you say it's all about what works for you that's important!

Whilst I do enjoy arguing with Noblehead that is actually a very interesting link; and one that I see sometimes with my diet. I low carb but wouldn't say I ordinarily go high fat but if I eat a 'low carb pizza' essentially cream cheese, linseeds, eggs plus pizza toppings then I will get a large delayed bump in blood sugars even when the pizza itself works out at about 5 carbs. Other than that slightly crazed fat based meal I don't see this though.

Dillinger, I would like to think we debate rather than argue as arguing never achieves anything. I appreciate what you say about being best able to control your bg better the way things are and it's all about finding what works for the individual, there's never a wrong or a right way and I'm sure we will have these sort of discussions for many years to come, thanks for the reply btw!

edit, apologies for the small print, haven't a clue why it does that some time

 

[SamJB]

Stats, numbers, surveys, percentages can all be "tweaked" or biased. That is also fact.

Apologies to pull you up on this Mo, but the item being discussed here is the results of the National Diabetes Audit. I am on the steering committee for the National Diabetes Audit and there are some of the finest diabetologists and medical statisticians that I have worked with on this project (my day job is in pharma/medical statistics for a global pharma company) - two steering committee members have CBEs. The NDA is, in fact, the largest clinical audit in the world. An enormous amount of care and intelligence has gone into producing the reports and I can assure you that the results have not been tweaked and bias has not been introduced.

You may have had a poor experience where you felt a survey's results were not to be trusted, but you can 100% trust the results of the NDA.

 

[mo1905]

Apologies to pull you up on this Mo, but the item being discussed here is the results of the National Diabetes Audit. I am on the steering committee for the National Diabetes Audit and there are some of the finest diabetologists and medical statisticians that I have worked with on this project (my day job is in pharma/medical statistics for a global pharma company) - two steering committee members have CBEs. The NDA is, in fact, the largest clinical audit in the world. An enormous amount of care and intelligence has gone into producing the reports and I can assure you that the results have not been tweaked and bias has not been introduced.

You may have had a poor experience where you felt a survey's results were not to be trusted, but you can 100% trust the results of the NDA.

Happy to be corrected Sam ! I guess I meant surveys in general but I'm sure there are some that are legit, as you have pointed out. 93% of diabetics not meeting targets is pretty scary then.

 

[SamJB]

No that makes perfect sense Smidge, I'm sure your method of bolusing for meals depending on portion size is similar to how SamJB boluses for his meals, if I'm not mistaken isn't this based somehow on the Bernstein method?

Yep, well remembered Noblehead! I'm about to repeat myself, but for the benefit of those who have not read this before...

When you eat your food will make its way to your large intestine. When it enters, it stretches the walls of the large intestine. This triggers glucagon production in anticipation of the meal, so that you can benefit from it straight away.

I bolus according to the portion size, i.e. to how much my large intestine will be stretched.

Some low carbers may have more fat/protein, there's a few ways of low carbing. Me? I just replace the carbs with extra veg.

 

[SamJB]

Happy to be corrected Sam ! I guess I meant surveys in general but I'm sure there are some that are legit, as you have pointed out. 93% of diabetics not meeting targets is pretty scary then.

Sorry to pull you up on it, Mo! You are right though, some surveys do need to be taken with a pinch of salt.

 

[noblehead]

Yep, well remembered Noblehead! I'm about to repeat myself, but for the benefit of those who have not read this before....

Thanks for the confirmation Sam.

 

[smidge]

No that makes perfect sense Smidge, I'm sure your method of bolusing for meals depending on portion size is similar to how SamJB boluses for his meals, if I'm not mistaken isn't this based somehow on the Bernstein method?

Yes, it is a sort of adaptation of Bernstein. To be fair, Bernstein applies a scientific method to the calculation of insulin based on the individual's size, amount of carb and amount of protein - his calculations of the insulin dose are pretty exact, but the amount and type of food you have to eat each day is very unforgiving and I know I could not stick to that. I accept his method is the best for complete BG control, but I have to have a regime I can stick to, so I take his principles and keep my carbs low. I try not to vary the amount of carb too much from day to day (because keep changing the amount of carb causes the basal dose to need adjusting), and I have worked out through trial and error and a lot of testing, recording and reviewing what various foods and quantities do to my BG - and then I have a sort of 'black art ' that goes 1:8 insulin:carb for the carbs in the meal+1 extra unit for the protein (or 1.5 if it's steak - no idea why, but maybe it's just bigger than a chicken portion LOL), half a unit extra just because I'm eating - and if it's a high fat meal (which is rare for me) reduce the Apidra by half a unit and check in 2 hours to see if I need another unit. Not as scientific as Bernstein, but pretty effective. It's quite difficult to explain to my consultant though! It's a lot of effort to start off with, but once you've sorted out the amount of bolus for particular sizes of meals it becomes pretty straightforward - sounds much harder than it is!

Smidge

 

[smidge]

To a point


To a point reducing carbs works and insulin usage does decrease (I know this to be true), however as we all know from personal experience and reading others on the forum that when you start to replace the carbs with too much fat and protein it can create more problems than it solves:

http://care.diabetesjournals.org/content/36/4/810.full

That's an interesting study. I'm not sure really what it objectively tells us about fat content of a low-carb diet though, as the meals given to the subjects were all high carb - and I don't know any diabetic who would recommend a high fat high carb diet as was fed to these subjects. I also felt the study failed to assess how the foods they were feeding the subjects normally affected them - for example, the high fat meal involved bread (a cheese sandwich) which spikes me uncontrollably no matter what I eat it with whereas the low fat meal contained rice which, in small portions, I am able to control with injected Apidra. So, I think the results would need further investigation to really tell us the effect of a LCHF diet, but it is certainly interesting in terms of the effect on BG and insulin requirements of fat in a HCHF diet and also in emphasising what most of us know but the medical profession seems unwilling to accept - that all three nutrients play a significant role in BG levels and insulin action, timings and doses need to take this into account instead of focussing only on carb counting.

Smidge

 

[fatbird]

To academicdiabetic

Is this the paper you have referred to?

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3612791/

FB

 

[Garr]

Apologies to pull you up on this Mo, but the item being discussed here is the results of the National Diabetes Audit. I am on the steering committee for the National Diabetes Audit and there are some of the finest diabetologists and medical statisticians that I have worked with on this project (my day job is in pharma/medical statistics for a global pharma company) - two steering committee members have CBEs. The NDA is, in fact, the largest clinical audit in the world. An enormous amount of care and intelligence has gone into producing the reports and I can assure you that the results have not been tweaked and bias has not been introduced.

You may have had a poor experience where you felt a survey's results were not to be trusted, but you can 100% trust the results of the NDA.

Is the audit just the collecting of all the results from our regular trips to the specialists? I'm sure it must be more complicated than that though, but If so, why don't they find out what the 7% who are meeting the target are doing to achieve it?

 

[SamJB]

Is the audit just the collecting of all the results from our regular trips to the specialists? I'm sure it must be more complicated than that though, but If so, why don't they find out what the 7% who are meeting the target are doing to achieve it?

Yes, that's right. The NDA collects hospital and GP records of your check-ups. The data from your HbA1c and other health checks are collected and analysed.

Interesting point about finding out what the 7% are doing correct. I've been mulling over this for a while as it is data that has a great deal of value attached to it.

The quantifiable factors that go into maintaining good BGs are not collected in a controlled environment. Factors such as carb doses, insulin doses, doses of other drugs, whether they've been ill or stressed and how frequent they have exercised are down to the patient to record. These are not recorded at check-ups. If a questionnaire were to be sent out to the 7%, how can we tell if a patient has recorded results correctly, how do we know if the values they have given have remained static throughout the HbA1c period? If we were to collect results in this fashion, it would be uncontrolled and people would quite correctly question the reliability of the results. (See Mo's point, above)

In addition, there will be some of the 7% that are in that category because of frequent hypos. Hypo frequency is not recorded at check-up time either and any data on this would be uncontrolled too. In other words, I suspect that the percentage of well controlled diabetics, i.e. have low hypo frequency and an HbA1c < 6.5% is less that the 7% that achieve this HbA1c target.

The only way we could find out what the well controlled group are doing would be to conduct a clinical trial, which will record the above factors in a controlled environment. Can we provide the details of the 7% to a research group? I don't know, but I expect data protection would prohibit this. It would also be measuring several variables, which is possible, but it can leave the results looking less robust.

In addition, most clinical trials are not all they are cracked up to be. A lot of papers referenced on here and by Diabetes UK only have a few dozen or a few hundred patients over a period of a few months, are conducted by a couple of clinicians and perhaps have a single statistician/analyst who is shared throughout the hospital/CCG/research faculty. In clinical trials conducted by pharma companies you have thousands of patients across multiple countries over a period of years and perhaps a couple of hundred analysts, statisticians, informaticians and clinicians who have decades of experience. This is why you get a lot of disagreeing papers with the former, whereas the latter setup is a lot more robust and objective. The problem is, is that the latter costs around $1bn.

So it's a great idea and I for one would love for this information to be shared, but it is not without its challenges.

 

[Garr]

Aye, agreed, just seems a waste that they can't use the evidence that's straight in front of them. I get a " everything's good, carry on doing what you're doing and I'll see you in 6 months. Any questions?" They don't ask what I'm doing. A simple questionnaire on diet, carb intake, snacks, exercise etc. might reveal patterns among people who are currently keeping tight control and those who don't. But nothing's ever simple though. Cheers for brilliant reply.

 

[phoenix]

The thread has morphed ,it really hasn't a lot to do with the original post which was about a T2 study not a T1 one. The mortality risks for people with T1 have decreased somewhat since pre insulin days! Data from Scotland shows that in recent years mortality rates have improved considerably. (in spite of fewer than 13% of Scottish T1s achieving less than 7%HbA1c ).
.http://www.eurekalert.org/pub_releases/2013-09/d-uss092413.php
I don't think that the original paper mentioned has actually been linked to, if not it's here:
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3612791/#!po=44.2308

FB quotes the results of the DCCT/EDIC study (seemingly favourably) and concludes

Of course -it is how good control is achieved that makes all the difference

Intensive control is what many of us try to do now compared with the conventional therapy of the time. I think it's a great shame that many are still not shown any methods of doing this. .

Conventional therapy was simply one or two injections a day with no targets other than avoiding the symptoms of hypo and hyperglycemia.

Intensive therapy in the DCCT included:

• 3 or more injections a day or the use of a pump.
• self monitoring at least 4 times a day with targets of 3.9mmol/l to 6.7mmol/l pre meals and less than 10mmol/l after meals.
• HbA1c target: less than 6.05% (but see results)
• Insulin self- adjusted according to diet and exercise.
• individualised advice on diet (and exercise) : Individual strategies to compile diets and adjust insulin (Healthy Food Choices, exchange systems, carbohydrate counting, and total available glucose.(TAG there are threads on this on the forum )
• diet composition goals : initially 10–25% protein, 30–35% fat, 45–55% carbohydrate, a polyunsaturated:saturated fat ratio of 0.8:1.0, and <600 mg cholesterol . The study protocol was amended in 1988 in response to the National Cholesterol Education Program (≤30% fat, ≤10% saturated fat, and <300 mg cholesterol) with further advice/medication if LDL cholesterol targets weren't achieved.
• Consistency of carb amounts and meal times was stressed as important. (This seems to differ from DAFNE , my own doctor/dietitan still stress this, and personally I find better/.easier control when I do this .However being able to be able to be more flexible also has benefits for lifestyle)

Some results :
1) Average HbA1c in the intensive group was 7.2% (It was 9.1% in the conventional group).
(so they didn't achieve the HbA1c target)
2)Those who had greater residual insulin found it easier to keep to achieve lower levels as did those who developed T1 at older ages(tended to have more own insulin) and those on pumps
3) Complication risk increased as HbA1cs above 7% and increased sharply at over 8%. Going below 7% didn't reduce the risks by much
Levels below 6.5 % increased risk of serious hypos (almost mirror image of complication graph)
(this was using older insulins than most of us use now ie NPH or lente rather than lantus/levimir which have flatter profiles and regular insulin which needs pre-meal dosing rather than rapid insulin which acts quicker and also the newer doesn't last as long in the body as regular )
The well known risk/hypo graph

http://www.medscape.org/viewarticle/560226
4) Analysis of available data suggests that it was the HBA1c or average that mattered rather than the everyday ups and downs of glucose levels.
Consistency in HbA1c was also important (ie not good to have periods of low HbA1cs followed by higher ones)
(see Kilpatrick analyses)
6)Those that reported following the meal plans tended to have lower HbA1cs.
http://care.diabetesjournals.org/content/16/11/1453.full.pdf+html
In the intensive arm, diets higher in fat and saturated fat and lower in carbohydrate are associated with worse glycemic control, independent of exercise and BMI
http://ajcn.nutrition.org/content/89/2/518.full#T2

NB This trial was almost stopped early on because of the early worsening of retinopathy with fast reductions in HbA1c. (It's why people should now be recommended to lower levels gradually)

After the trial (EDIC)
the HbA1cs of those in the IT arm rose , but the better control seen during the trial seemed to have a residual beneficial effect.
Importantly, 30 years after the trial there have been persistently reduced complications in the intensive arm.86.7% of that risk reduction was explained by between-group differences in HbA1c levels
There has for example been a 39% reduced risk of microalbuminuria and a 61% reduced risk of macroalbuminuria in the intensive arm. (P<0.0001 for both).
http://www.medpagetoday.com/MeetingCoverage/ADA/40047
I think that is a really important finding. (and may be significant for mortality rates). A very recent Danish analysis of mortality in T1 showed that life expectancy was much improved but that kidney disease could be an important factor in those that had reduced longevity. Those that avoided it now tended to have a near normal life span. http://www.ncbi.nlm.nih.gov/pubmed/23949580

 

[noblehead]

Yes, it is a sort of adaptation of Bernstein. To be fair, Bernstein applies a scientific method to the calculation of insulin based on the individual's size, amount of carb and amount of protein - his calculations of the insulin dose are pretty exact, but the amount and type of food you have to eat each day is very unforgiving and I know I could not stick to that. I accept his method is the best for complete BG control, but I have to have a regime I can stick to, so I take his principles and keep my carbs low. I try not to vary the amount of carb too much from day to day (because keep changing the amount of carb causes the basal dose to need adjusting), and I have worked out through trial and error and a lot of testing, recording and reviewing what various foods and quantities do to my BG - and then I have a sort of 'black art ' that goes 1:8 insulin:carb for the carbs in the meal+1 extra unit for the protein (or 1.5 if it's steak - no idea why, but maybe it's just bigger than a chicken portion LOL), half a unit extra just because I'm eating - and if it's a high fat meal (which is rare for me) reduce the Apidra by half a unit and check in 2 hours to see if I need another unit. Not as scientific as Bernstein, but pretty effective. It's quite difficult to explain to my consultant though! It's a lot of effort to start off with, but once you've sorted out the amount of bolus for particular sizes of meals it becomes pretty straightforward - sounds much harder than it is!
Smidge

Admittedly I don't know a great deal about the Bernstein diet only that what I've read on this forum and some US ones, it does seem rather too restrictive for most people and it makes sense to adapt it to your own choosing. It's interesting just the same how you dose adjust for your meals Smidge and at first glance it does seem rather complicated, however you seem to have mastered it and have good diabetes control, at the end of the day that's all that matters.

That's an interesting study. I'm not sure really what it objectively tells us about fat content of a low-carb diet though, as the meals given to the subjects were all high carb - and I don't know any diabetic who would recommend a high fat high carb diet as was fed to these subjects. I also felt the study failed to assess how the foods they were feeding the subjects normally affected them - for example, the high fat meal involved bread (a cheese sandwich) which spikes me uncontrollably no matter what I eat it with whereas the low fat meal contained rice which, in small portions, I am able to control with injected Apidra. So, I think the results would need further investigation to really tell us the effect of a LCHF diet, but it is certainly interesting in terms of the effect on BG and insulin requirements of fat in a HCHF diet and also in emphasising what most of us know but the medical profession seems unwilling to accept - that all three nutrients play a significant role in BG levels and insulin action, timings and doses need to take this into account instead of focussing only on carb counting.

Smidge

I'm not too sure about whether it's just relevant to HCHF Smidge. I've long suspected the delayed rise in bg maay hours after eating (even with a low-carb high-fat meal) couldn't possibly be down just to the slow digestion of the protein, and this is why I suspect that some type 1's that do go down the low-carb high-fat route need to inject twice for one meal a few hours apart, as I said earlier I need to inject twice for bacon eggs without any visible carbs otherwise my pre-prandial bg levels would be high the next time I eat.

Both Dillinger and I were discussing this subject last year when this study came out about high-fat diets causing insulin resistance in type 1's, Dillinger will tell you that his consultant believes that his resistance to the insulin he injects is down to the fat in his diet, and if you read this recent thread from a long-term low-carber you can see that eliminating some fat in his diet resulted in a big reduction in insulin requirements:

http://www.diabetes.co.uk/forum/threads/saturated-fats.48599/#p438010

It's all interesting stuff and the more we understand how food effects our bg the better we can dose adjust for those meals, that is why I'm looking forward to the Joslin finding and their advice on bolusing for fat in the diet.

Phoenix, as ever a great posts and so much to read, I'll take a look at the links later when I have more time.