Treating diabetes

Treatment for diabetes has one aim: to keep blood glucose in a range that avoids harm now and damage later. How that is done differs so completely between type 1 and type 2 that almost nothing useful can be said about treating "diabetes" without naming the type first.

What follows describes what the treatments are and what they do. It is not a guide to which anyone should be taking, which depends on the type, on kidney and heart health, on other medicines, on weight, on occupation and on preference — and is a conversation with a diabetes team rather than something to be settled from a web page.

Type 1: insulin, and nothing instead of it

In type 1 diabetes the pancreas no longer makes insulin, so insulin has to be supplied from outside. There is no alternative to it, there never has been, and this is the single most important fact in the whole subject: nothing taken instead of insulin is safe in type 1 diabetes. Stopping it leads to ketoacidosis within hours to days.

Insulins differ in how quickly they act and how long they last, and treatment usually combines two kinds:

• Rapid-acting insulin, taken with food to cover the carbohydrate in a meal.
• Long-acting or basal insulin, taken once or twice a day to provide the steady background the body would otherwise supply.

That combination is the basal-bolus pattern, which allows meals, exercise and illness to be accounted for separately. Older human insulins and twice-daily mixed insulins are still used, particularly where a simpler routine suits better.

Delivery is by pen, or by pump — a small device giving a continuous background infusion with extra doses at meals — and increasingly by a pump paired with a glucose sensor that adjusts the background automatically. Those systems, and the sensors, are covered under living with diabetes.

Alongside the insulin comes the skill of using it: carbohydrate counting and matching doses to food, which is what structured courses such as DAFNE teach and which makes a substantial difference to how well the whole thing works.

One correction worth making explicitly. The tablets used in type 2 diabetes that work by prompting the pancreas to release more insulin do nothing at all in type 1, because there is no insulin-producing tissue left to prompt. Older articles that list those tablets as helpful for type 1 have it exactly backwards, and the error is not harmless.

Type 2: a sequence that has changed recently

Type 2 diabetes is treated in steps, with treatment added as the condition progresses. Food, activity and weight are part of the treatment throughout rather than a preliminary to it.

Metformin is usually the first medicine. It reduces the amount of glucose the liver releases and improves the body's response to insulin; it does not cause hypoglycaemia; and it has been in use long enough for its effects to be well understood. Its common problem is digestive upset in the first weeks, which often settles and often improves on the slow-release form — worth knowing, because people abandon it unnecessarily.

The significant recent change is that treatment no longer starts from glucose alone. UK guidance now begins with an assessment of cardiovascular risk, and for people with heart failure, established cardiovascular disease, or a high risk of it, an SGLT2 inhibitor is given alongside metformin from the outset.

That reflects something genuinely new. SGLT2 inhibitors, which make the kidneys excrete glucose in the urine, turned out in large trials to protect the heart and the kidneys to a degree that has little to do with their effect on glucose. They are now used for that protection as a primary reason rather than a bonus — which means anyone with diabetes and heart or kidney disease who has been on metformin alone for years may be due a review.

Beyond those, several classes are used singly or in combination:

• DPP-4 inhibitors, which enhance the body's own glucose-lowering hormones. Well tolerated, no hypos, modest effect.
• Pioglitazone, which improves insulin sensitivity.
• Sulfonylureas, which prompt the pancreas to release more insulin. Effective and cheap, and the one group of tablets that can cause hypoglycaemia — which is why anyone taking them is given the same advice about low glucose as someone on insulin, and why they are relevant to the driving rules.
• GLP-1 receptor agonists, given by injection, which slow stomach emptying, reduce appetite and produce weight loss as well as lowering glucose. UK guidance reserves them for specific circumstances, mainly around higher body weight or where insulin would cause particular difficulty.
• Insulin, where tablets are no longer enough.

On that last point: needing insulin in type 2 diabetes is not a failure and not a punishment. Type 2 involves a gradual decline in insulin production, so for many people reaching insulin is simply what the condition does over time. Reluctance to start it — on both sides of the consultation — is common and well documented, and the delay does measurable harm.

Gestational diabetes

Diabetes that appears during pregnancy is actively treated, not watched. Older writing that describes doing nothing as an option has it wrong: raised glucose in pregnancy affects the baby's growth and raises the risk of complications at birth and of low blood sugar in the newborn, which is why it is monitored closely and treated promptly.

It starts with dietary changes and glucose monitoring, with metformin and insulin added where those are not enough. Most of the tablets used in type 2 diabetes are not used in pregnancy. It usually resolves after the birth — and the follow-up glucose test afterwards, then annually, is the thing most often missed, as set out under risk factors for diabetes.

Weight loss, surgery and remission

For type 2 diabetes, weight loss is a treatment rather than an adjunct, and a substantial one. Enough of it, achieved reasonably soon after diagnosis, puts a meaningful proportion of people into remission — the evidence for that, and the limits of it, are under living with diabetes.

Medicines originally developed for diabetes and now also used for weight loss have changed this picture considerably. Two practical points: they are prescription-only medicines with real side effects and real monitoring requirements, and there is a large and unregulated market selling things that claim to be them. Buying injectable medication outside a pharmacy is dangerous, and has caused serious harm.

Bariatric surgery has a larger and more durable effect on type 2 diabetes than any medicine, and is available on the NHS against specific criteria.

Treatments that were withdrawn, and ones that never existed

This section exists because articles on diabetes treatment age badly and a reader has no way of telling how old the advice is.

Rosiglitazone, sold as Avandia, was withdrawn across Europe in 2010 over cardiovascular risk. Any article still recommending it predates that.

There is no oral insulin. Insulin is a protein, and digestion destroys it. Claims that it is "now in widespread use" have been circulating for twenty years and have never been true.

Inhaled insulin was launched in 2006 and withdrawn the following year. A later inhaled product exists in the United States and is not available in the UK.

And a general point about names: UK prescribing uses generic names — metformin, gliclazide, empagliflozin — while older articles tend to use American brand names, several of which refer to products never sold here or no longer sold anywhere. The generic name on the box is the one to go by.

What not to do with this article

Nothing here is a reason to start, stop, substitute or adjust any medicine. Glucose-lowering treatment interacts with illness, with other prescriptions, with alcohol, with exercise and with how much is eaten, and the doses are set individually.

Two things in particular. Insulin is never stopped in type 1 diabetes, for any reason, including during illness or when not eating. And treatments offered as a way of coming off medication — supplements, restrictive regimens, clinics promising withdrawal — belong in the category covered under alternative therapies: weight loss achieves remission in some people with type 2 diabetes, and that is a supervised process with measured results, not a product.

See also diabetes for the rest of this section.