Issue 20 / 28 May 2012

TYPE 2 diabetes is one of the fastest growing epidemics in human history. It is not a single disorder in terms of causation and is characterised by progressive pancreatic beta-cell failure.

While lifestyle measures, weight loss and exercise should always be considered as the first-line treatment, these succeed in a limited number of patients. Pharmacological intervention with blood glucose lowering agents is almost always needed.

The multiplicity of causal metabolic defects involved in type 2 diabetes makes it unrealistic to expect that most people with diabetes will, over time, achieve excellent blood glucose control with any single oral hypoglycaemic agent or any combination of agents.

As the progressive nature of type 2 diabetes requires ongoing assessment of metabolic control, intensification of therapy with increasing doses of oral hypoglycaemic agents then a requirement for insulin is not unusual.

Paradoxically, the main recommended plans (algorithms) for treatment, such as the one from the American Diabetes Association and European Association for The Study of Diabetes, may actually set us up for therapeutic failure. Each new recommendation suggested by these algorithms is a result of failure of the previous step.

The clinical pathway finally ends with the administration of insulin, often combined with one or more oral hypoglycaemic agents. A frequent outcome of insulin therapy is weight gain. Obviously, this is counterintuitive for the treatment of patients with type 2 diabetes.

A recent review and meta-analysis in the Canadian Medical Association Journal addressed the issue of initiating insulin in patients with type 2 diabetes. One of its key recommendations is that insulin should be considered early in the course of treatment.

However, experience from North America and the UK in managing type 2 diabetes may differ to that of Australian diabetologists, and many of us are not so “trigger happy” to move to insulin early. It is a subject being debated widely.

Few diabetologists would claim that our present management of type 2 diabetes (and, indeed, type 1 diabetes) is ideal.

In type 1 diabetes, there is absolute insulin deficiency, and multiple daily injections are needed — a basal dose of long-acting insulin and an injection of short-acting insulin before each meal. This regimen attempts to replicate the normal pancreatic response consisting of a continuous basal release of insulin with bursts at each meal in response to food.

Increasingly in patients with type 2 diabetes, as a result of both physician and patient failures to perfect treatment, we are seeing a quick move to insulin therapy. This is often before an adequate analysis of the situation is made, and other options explored.

Decades of managing type 2 diabetes lead me to agree with the list of barriers to initiation of insulin cited by the Canadian authors, ie, fear of hypoglycaemia, the pain from frequent blood testing, weight gain, fear of injections and the associated pain.

Now, while these are important issues for the patient, and often affect the physician’s judgement, it is important to treat every person individually.

Type 2 diabetes is characterised by two key features — pancreatic beta-cell failure and insulin resistance — and the balance between these two varies in individuals, so the natural history of diabetes and its treatment will vary. For example, a person with predominant beta-cell failure may be likely to require insulin earlier in their course of treatment than someone with predominant insulin resistance.

It is very clear that, in a patient with recent onset type 2 diabetes who has lost weight and who has significant and symptomatic hyperglycaemia, insulin should be the first line of treatment. However, there is a large gap between this scenario and the usual presentation of type 2 diabetes, which is often asymptomatic and discovered accidentally.

The jury is still out on the benefits of early institution of insulin in type 2 diabetes, so the debate will continue.

In the meantime, practitioners should promote lifestyle measures and optimal use of oral hypoglycaemic agents. If these measures fail, judicious use of insulin should be considered, and it can often be done as an adjunct to oral agents.

Insulin is safe and can be effective, but each person with type 2 diabetes should receive education and be assessed carefully, considering their compliance to a regimen that requires long-term commitment.

Professor Paul Zimmet is director emeritus of the Baker IDI Heart and Diabetes Institute.

The author has acted as an adviser for Lilly, Novo Nordisk and sanofi Aventis.

Posted 28 May 2012

8 thoughts on “Paul Zimmet: Starting insulin still contentious

  1. JD says:

    Prof Zimmet can you explain why metformin should not be more widely started EARLY in the disease process (at the insulin-resistant pre-diabetic stage)? It is safe, effective, helps weight management and delays the progression to diabetes and subsequent pancreatic failure. Lifestyle measures are important, but my experience (and I’m sure yours) is that patients’ success is very mixed.

  2. Paul Zimmet says:

    This is an excellent question and I agree in principle. Unfortunately metformin is “off label” for people in the “prediabetic” phase. Its use could either prevent diabetes or at least delay its onset. It is a relatively safe drug except in persons with renal impairment and if I look in my “crystal ball”, I think it may well be accepted practice in years to come.

  3. Anonymous says:

    Professor Zimmet.
    At which stage of the disease process would you recommend to patients who are obese (BMI over 30) to access bariatric surgery?

  4. Paul Zimmet says:

    Surgery may be considered as an alternative treatment option for type 2 diabetes in persons with BMI 30 to 35 when diabetes cannot be adequately controlled by optimal medical regimen, especially in the presence of other major cardiovascular disease risk factors such as hypertension and dyslipidaemia

  5. rose says:

    Regarding Metformin being “off-label” for pre-diabetes, it is listed as a general benefit on the PBS. Two issues concern me-the first being the underdiagnosis, and late diagnosis of type 2 diabetes, which is known in Indigenous people, whose HBAIC is not funded by Medicare until they have proven diabetes, which is often a late diagnosis – are we being negligent in not prescribing Metformin when a random BSL is 10 and a fasting GTT is not available ,or shows a 2 hour glucose of 10 due to patient compliance-has any study been done on the validity of a GTT if the patient is not fasting?
    The second concern is that proven diabetic patients, with improved diet , exercise and Metformin may eventually have an HBAIC which falls from a high level to a pre-diabetic level-what happens if we stop Metformin, and their HBAIC returns to a diabetic level, are we negligent for stopping Metformin?
    I would like to see HBAIC funded annually by Medicare for including people with an elevated random glucose, patients with risk factors including elevated BMI, family history, history of gestational diabetes.
    What is the cost of HBAIC versus GTT?

  6. not an adviser says:

    Do any of the drug companies that Prof Paul Zimmet has acted as an adviser for market metformin?

  7. Paul Zimmet says:

    There are now many generic versions of metformin, and combination therapies with metformin. Metformin has been off patent for over a decade. I have been an adviser for Novartis, which has a combination therapy with this compound, and some 15 years ago I was an adviser to Lipha, which was the original developer. Metformin is already recommended by the American Diabetes Association for prediabetes and there appears to be widespread “off-label” use of it even in Australia.

  8. JD says:

    Rose, you may be interested in this article: Ann Intern Med. 2008;149:200-203 which argues against a binary approach (ie you do or don’t have it) for diseases like T2DM. It argues we are doing patients a disservice waiting for them to reach an arbitrary point on what is actually a continuum in a disease process.
    You may be aware the American Diabetes Association has already recommended the use of HbA1c for diagnosis (5.7% – 6.4% = prediabetes, ≥ 6.5% = diabetic).
    As far as “off-label” use goes, is this real a restriction to PBS prescribing?

Leave a Reply

Your email address will not be published.