Aggressive Insulin Treatment vs. Pills for Diabetes with Athersclerotic Disease: No Difference in Outcomes

When doctors and their patients review the treatment options for diabetes mellitus, a common question is "Why mess around?" If the blood glucose, thanks to a relative lack of the hormone "insulin," is "high," shouldn't the more "natural," tighter and physiologic answer be... insulin? While there are pills that can lower blood glucose levels and patients detest shots, it's unclear if the long term clinical, economic and quality-of-life outcomes favor one approach (oral medications) over another (insulin).

The Disease Management Care Blog thinks that these are the issues that were explored by the important, international and huge multi-center "ORIGIN" Study.  The results have just been published by the New England Journal.

Over 12,000 persons over age 50 years with

1) impaired glucose control or "pre" diabetes (more on this topic here), or

2) just-diagnosed diabetes or,

3) well-controlled diabetes (A1c less than 8 to 9%) on no or just one oral medication

and

4) established heart or atherosclerotic vascular disease

were randomly assigned to insulin glargine (with dosing that aimed for a normal blood glucose of 95 mg.%) or usual care (that relied on physician judgement and local guidelines). 

After one year, 50% of the insulin group hit their targeted blood glucose level and their median A1c (a measure of average blood glucose control over time) was 5.9%.  The usual care group achieved a median glucose level of 123 mg.% and after one year the A1c was 6.2%.  The difference in A1cs persisted over the remainder of the study (Table here)

The mean age of the participants was 63 years with an impressive median follow-up of 6 years that yielded outcome results on 99% of the participants.

Results?  No difference in heart attacks or kidney disease.

When cardiovascular death, nonfatal heart attack, non-fatal stroke were combined, the incidence was the same in both groups - about 3% per year.  There was no difference in kidney outcomes including deterioration in function or need for dialysis.  Hospitalization rates for any cause were the same in both groups.  There was an isolated difference involving angina that, in the DMCB's mind, may have been a statistical fluke.  You can look at the outcomes for yourself here.

There was one important difference.  Among the 1456 persons without formal diabetes (the "impaired" group - see above), persons given the insulin were less likely to progress to a formal diagnosis of diabetes (25% vs. 31%).  Unfortunately, they paid a price, because they had a higher rate of insulin-induced low blood sugar reactions (an incidence of 17 vs. 5 per 100 person-years).

Based on these results, the DMCB thinks:

1. Turning to insulin treatment early in the course of pre or diabetes treatment for persons with heart disease doesn't appear to offer any important difference in macrovascular (heart attack and stroke) disease or kidney disease outcomes. However, that's only true among patients who have achieved an A1c below 7%.  (The DMCB can't figure out what happened to the patients with a baseline A1c in the 8% to 9% range who didn't get to an A1c below 7 - did insulin help them?)

2. What's more, it's possible that driving an A1c lower - once it's below 7% - doesn't offer any additional outcomes advantage.

3.  While early insulin supplementation may prevent the "burn out" of the insulin-producing cells of the pancreas, the price for that is a higher incidence of low blood sugar reactions.  Even though this "cure" of diabetes may seem like a big deal, why bother if there's no difference in survivorship?

4. As the population health management service providers discuss care planning with their patients with diabetes and heart disease, the topic of early aggressive insulin may come up.  Here's an answer to that question.

5. If accountable or risk-assuming organizations believe that early aggressive insulin treatment will lower the direct costs attributable to heart disease among their patients and enrollees with diabetes, the answer is no.

That out-of-ate but compelling image is from a 2003 HHS website on why Prevention Makes Common Cents