May 28, 2008

Insulin Right After Diagnosis Dramatically Improves Type 2 Outcome

Two studies just published in the journal Lancet show you just how mistaken is the current practice of starting Type 2s on oral drugs and withholding insulin until their A1c with a full load of oral drugs is 10% or higher.

These are the studies:

Effect of intensive insulin therapy on β-cell function and glycaemic control in patients with newly diagnosed type 2 diabetes: a multicentre randomised parallel-group trial. Jianping Weng et al. The Lancet 2008; 371:1753-1760


Intensive insulin therapy in newly diagnosed type 2 diabetes. Ravi Retnakaran and Daniel J Drucker. Lancet 2008; 371:1725-1726. (Subscription required)

In the first study, "The patients, with fasting plasma glucose of 7·0–16·7 mmol/L [126 - 300] , were randomly assigned to therapy with insulin (CSII [pump] or MDI [basal/bolus shots]) or oral hypoglycaemic agents [oral drugs] for initial rapid correction of hyperglycaemia. Treatment was stopped after normoglycaemia [normal blood sugar] was maintained for 2 weeks. Patients were then followed-up on diet and exercise alone."

Here's what happened:

"A year after stopping therapy, the remission rate was 42% among those who reached normal blood glucose levels during the treatment period, the researchers said.

But the rates were 51.1% among those who were treated with insulin infusion, 44.9% among those given insulin injections, and only 26.7% in the oral hypoglycemic agents group."

What this means is that almost twice as many newly diagnosed people with Type 2 diabetes who received intensive insulin treatment right after diagnosis were able to achieve normal blood sugars using only diet and exercise than did the people treated only with oral drugs. Even though the patients given insulin were taken off insulin after experiencing only two weeks of normal blood sugars!

This is a monumental finding and one that should make you insist that your doctor give you a basal/bolus insulin regimen as soon as you are unable to maintain normal blood sugars with diet and exercise alone. If you can't get truly normal blood sugars by cutting the carbs and increasing your physical activity, skip the expensive and ineffective oral drugs and go to the drug that always lowers blood sugar: insulin.

Why does insulin work so much better than other drugs?

The answer is probably because it is the only drug that reliably drops blood sugars below the level that cause secondary insulin resistance. Many doctors do not seem to understand that if your blood sugar is high the high blood sugar itself causes insulin resistance no matter what your underlying physiology might be. And this additional blood-sugar related insulin resistance starts at relatively low levels--much lower than doctors understand. I personally see a huge difference in my insulin resistance after meals--measured by how much insulin I need to cover a given number of carbs--when my fasting blood sugar is 108 mg/dl and when it is 85 mg/dl.

But when you take an oral drug that does a feeble job of lowering your blood sugar, you have to contend not only with the damage caused by the too-high blood sugar, but also with the additional insulin resistance caused by your too-high blood sugars. This IR packs on additional pounds and hastens the burnout of your insulin producing beta cells because they must make much more insulin to cover the meals you eat.

A telling fact that came out at the recent AACE conference that got no play in the media at all is that in the last decades the average A1c of people with diabetes in America has risen dramatically.

As reported in the Endocrinology Today newsletter: "Between 1988 and 1994, NHANES data reported 44.5% of patients reaching a target HbA1c of 7.0% or less. Between 1999 and 2000, that percentage dropped to 35.8%."

The reason fort this? The endocrinologists scratch their head but admit that with the greater choice of oral drugs, fewer patients are using diet to control blood sugar. And though the article doesn't spell this out, it is also likely that because there are so many new, expensive, highly promoted oral drugs, doctors are delaying the move to insulin for much longer than they did in the late 80s when they had few oral drugs to try and moved to insulin faster. The article does report, "Insulin use in the United States remains low."

Doctors like oral drugs because they don't have to follow up with patients, educate them, or worry about hypos. Drug company reps make it sound like their drugs can provide healthy blood sugars, even though the prescribing information (that doctors rarely read) shows that most lower A1c by no more than 1% and many by only .5%--in patients whose blood sugar starts at levels of 8% or higher.

Until now we had vague information suggesting that using insulin immediately after diagnosis could preserve the beta cell function of people with LADA. Now, with this new data, we see that using insulin right after diagnosis benefits Type 2s, too.

So don't let your doctor tell you that it's better to try all of the many oral drugs before you start insulin. It isn't true, and waiting three or four years while taking drugs that can't normalize your blood sugar may mean that by the time you start insulin you have few beta cells left to save.


Anonymous said...

Your writing is so informative and interesting. I remembered an earlier post you wrote about insulin being more suitable than oral drugs. My glucose readings are normal because of the low carb diet, but once in while I'll eat too much fruit or other carb and my blood sugar shoots up. I have no recourse, even excerise doesn't help me, to lower the blood sugar. But if I had access to insulin and was taught how to use it judiciously, then I would have a viable option. Jenny, are there people who are non-diagnosed for diabetes, but have high after-meal blood sugars, able to use insulin when most needed?

Thank you!!


Jenny said...


You wouldn't want to use insulin until you were going into the 200s routinely after eating and your fasting blood sugar was around 125 mg/dl.

Insulin is powerful stuff and requires a lot of work to use it properly. I didn't start using it myself until I was going up to 140 mg/dl eating low carb and my fasting blood sugar had crept into the 120s.

And it is worth noting that insulin doesn't make it easy to eat lots of carbs and indeed the more carbs you try to cover with insulin, the more likely you are to have a serious hypo. It works best when you only try to cover at most 30-50 g (depending on your body size) and I find I do best if I eat one or two low carb meals a day and then one meal with slightly more carbs and insulin.

Anonymous said...

Thank you for these citations and commentary. After a year of maintaining an A1c of between 5.0%-5.2% on 40-60 grams of carbohydrate a day, I'm beginning to see a creep in my FBS and post-meal readings. No change in diet; more physical activity rather than less. Since the creep is close to 15% increase, I'm worried that what I'm doing is no longer working. I'm printing out this blog entry to take to follow up appointment with physician next week and am going to have a chat with her about insulin.

Anonymous said...

"[Insulin] is the only drug that reliably drops blood sugars below the level that cause secondary insulin resistance."

What level is that?

Thanks for your excellent blog, by the way.

Jenny said...

There are differing scholarly opinions about what blood sugar level causes secondary IR. I believe 180 is the number most people think is the boundary. But like everything with blood sugar, it probably isn't binary but is a gradient.

120 mg/dl post meal appears to be normal and the threshold at which most people secrete a big burst of insulin. I suspect over that every bit of increase has an affect on IR until by 180 it is really noticible.

Anonymous said...

Jenny, also just a low carb diet can improve type 2 outcome. I don't think you should be recommending people go straight onto insulin when diagnosed.

Jenny said...


Read what I wrote. I wrote that people should go onto insulin if a low carb diet and exercise did not give them normal numbers.

But it is important to remember than not everyone is capable of adhering to a diet. I don't like the whole puritan punitive attitude that says, "If they can't be self-disciplined and eat a stringent diet they deserve to go blind and lose their limbs."

The typical Type 2 has an A1c between 10 and 12% when first diagnosed--usually after many years of undiagnosed diabetes. If immediate intervention with insulin could leave them with normal blood sugars a year or two later, yes, use insulin.

For those of us capable of dieting the LC diet is worth a shot. But as someone who did LC for 6 years after diagnosis and was NOT able to get normal numbers, I wish I had gone on insulin immediately. I think I'd have more beta function left now. I was over 140 mg/dl too often eating LC because I am insulin deficient and that eroded my fasting blood sugar control.

Zolar1 said...

My last a1c was 5.3, and that was WITH spikes into the 200's, and almost 300.

I asked specifically for insulin. I told them I didn't want to spike that high, and they agreed. Doctor's office suggested pills - I said NO, I HATE TAKING PILLS. So, they gave me my first insulin pen. I love taking shots! I can eat a bit more normally from time to time, like a lousy piece of birthday cake and scoop of ice cream. Or even a small french fry from time to time too.

Pills won't necessarily lower blood sugars fast enough or long enough, and they can cause liver problems years down the road.

I see my endo here in a few days. I will ask for a permanent prescription for insulin and a medical ID bracelet.

And if the doc won't comply, then I'll fire them and get another one who WILL give me insulin. After all, the stuff works wonderfully 95% of the time, and I am the 'customer', not just a patient.

I use the DAFNE method when eating, and if necessary 2 hours later, I will bolus dose.

DAFNE stands for: Dose Ahead For Normal Eating.

I use Humulin R for DAFNE most of the time, and Humalog for bolus doses when needed.

I tried the doctor's suggestion on dosing. Got too many hypos.

Doing it my way, I never had a single Hypo.

I am VERY VERY glad to go on insulin, even though my blood sugars are generally in the normal range, except when I eat certain foods at certain times of the day, and even the temperature causes me to spike highly.

Insulin is also evening out my rapid weight loss problem. I can lose 5 pounds in a single night, and I only weigh around 135 pounds!

Insulin slows the swings.

My glucose meter is my 'food boss', not some doctor or dietitian or educator.

When I say that, I mean that I try to not use insulin if I can under normal conditions.

I don't want to be an "Insulin Junkie" whereby a person would have a 2L of sugar real soft drink in one hand and a bunch of insulin filled syringes in the other. Drink - shoot up, drink-shoot up, repeatedly.

Nope, I don't want to fall into that pitfall, as tempting as it is.

In conclusion, insulin works best for me, my body, and my random lifestyle.

Pills can't anticipate anything.

Jenny said...


You are probably one of the many people for whom the A1c is a poor guide to blood sugar. If you are anemic you can get a low reading(as low as in the 4% range) while having blood sugars (like what you describe) high enough to make you go blind.

Family doctors can be very ignorant about this. The combination of those numbers and your being so thin also suggests you don't have insulin resistant Type 2 so oral drugs wouldnt't be useful.

However, for some people Metformin IS all they need.

P.S. Anyone whose A1c does not match their actual measured blood sugar (as was the case here) must ask for the Fructosamine Test which is accurate in that case.

Unknown said...

since one year I am using PIOGLITAZONE 15 mg at bed time. Now FBS is 126 mg/dl and post-meal is 134. If i start insulin therapy will it gives better outcome for me? does insulin β-cell function and glycaemic control.

Jenny said...

Bhagawan Karki,

Long-acting insulin could bring down your fasting blood sugars. Your post-meal number is not high enough to make fast-acting insulin a safe approach. Ask your doctor about whether long-acting insulin might be helpful.

Anonymous said...

Type two diabetic. Taking novolin R for meals. Getting good post meal glucose reading (120, and 100). Need 60 units to control 50 to 75 grams carb with meals. Am having bad fasting blood sugar (150 to 200). Have heard about novolin N being good to control fasting blood sugar. My primary care physician wants me to go on the expensive basil insulin to control fasting sugar. Can not afford it. Can I get any control of fasting blood sugar by using an Injection place on my body and using novolin R. Or would 70/30 be better?

Jenny said...


Your question is too complex to be answered in a blog comment. I would suggest you read the book, Dr. Bernstein's Diabetes Solution to learn more about how to use the cheap insulins properly. Even if you eat more carbs than he suggests, the information in that book will be very helpful.

Anonymous said...

I do want you to know that I just recently began into internet searching for answers to my type 2 diabetes. And have to say you are doing such a great job. Your website and blog have been a big help for me so far. I will get Dr. Bernstein's Diabetes Solution.

I do understand that poor people with diabetes are in need of better meds and the better insulin. It is just another sad part of this disease. I also am on metformin 2500 mg daily. And know that if I could afford Glucophage I may have much better fasting blood sugar.

I am looking at all the info you have on the website. And will continue to do so. Thank you for being here.

Jenny said...

Generic metformin works fine. I wouldn't obsess about the Glucophage being significantly better. R is good and safe.