Practice Therapeutics

Newer insulins in type 2 diabetes

BMJ 2012; 345 doi: (Published 11 September 2012) Cite this as: BMJ 2012;345:e4611
  1. Edwin A M Gale, emeritus professor of diabetic medicine
  1. 1University of Bristol, Bristol, UK
  1. Correspondence to: Professor E A M Gale, Diabetes and Metabolism, Learning and Research, Southmead Hospital, Bristol BS10 5NB Edwin.Gale{at}

A 68 year old woman with a body mass index (weight (kg)/(height (m))2) of 31 has a five year history of type 2 diabetes. She is reasonably active but takes treatment for hypertension, hyperlipidaemia, and angina. She has signs of early diabetic background retinopathy. Her current treatment for diabetes is with metformin 850 mg twice daily and gliclazide 160 mg daily. Her haemoglobin A1c (HbA1c) concentration has risen from 0.076 to 0.085 over the past year, and she is keen to improve her glucose control.

After review of her diabetes management, she opts for insulin as the next step in treatment. She mentions that a friend has just started a special new insulin and wonders if that is what she will be taking too. After discussion of the evidence for insulin analogues, she agrees to start taking isophane insulin rather than an analogue.

What are the insulin analogues?

In health, glucose control is achieved by feedback regulation, such that food intake is matched by sharp spikes of insulin, and the glucose concentration remains stable between meals because of regulated basal insulin secretion. By contrast, injected insulin has no feedback regulation, enters the systemic rather than the portal circulation, is delivered into subcutaneous tissue rather than the bloodstream, and diffuses variably into the circulation. Standard, short acting insulins reach the circulation more slowly than food, so blood glucose tends to swing too high after each meal and to drop too low before the next. Longer acting human insulins reach peak absorption 4-16 hours after injection, which means that insulin injected before the evening meal tends to produce a glucose dip in the middle of the night, followed by a rebound as insulin fades before breakfast. Injection therapy thus tends to produce unwanted oscillations in glucose control, with an increased risk of hypoglycaemia before meals and during the night.

Genetically engineered human insulin, introduced in the 1980s, opened the way to subsequent modifications in the molecule designed to produce faster short acting insulins and smoother long acting insulins. Faster absorption is obtained by minor changes in the amino acid sequence. Smoother absorption was traditionally obtained by addition of a retarding agent, such as protamine—as in isophane insulin (also known as human NPH insulin (neutral protamine Hagedorn insulin)). By contrast, insulin glargine forms an amorphous precipitate when exposed to tissue pH at the injection site, whereas insulin detemir has a fatty acid tail, which allows it to bind to albumin, thus creating a reservoir, which buffers release of insulin into the tissues. The insulin analogues are designed to modify the delivery of insulin to its target tissues, but not its action, and controversy revolves around the clinical utility of such pharmacokinetic differences.1 The table lists commonly used insulins.

Commonly used insulins with their relative costs

View this table:

How well do the insulin analogues work?

Discussion of the role of the insulin analogues in type 2 diabetes inevitably reflects a wider discussion on the role of insulin itself. The past two decades have seen a drive towards earlier and more aggressive insulin therapy, but the benefits of this strategy are still open to debate. One reason is that attention has focused on the pursuit of lower thresholds for HbA1c concentrations rather than on the global benefits and disadvantages of different treatment strategies. Recent major outcome trials have therefore related their endpoints to glucose control, and have permitted considerable flexibility in the means used to achieve this. The main exception to this was the United Kingdom Prospective Diabetes Study, which subdivided participants randomised to intensified control according to treatment modality; it found no long term benefit from early randomisation to insulin,2 although the complex study design might leave this conclusion open to some doubt.

Be cautious about making treatment decisions on the basis of HbA1c concentration alone as there are many reasons for suboptimal control, including poor compliance with diet and other aspects of treatment. Furthermore, the undoubted benefits of intensified glucose control are greatest in relatively young individuals with newly diagnosed type 2 diabetes—the group targeted by the United Kingdom Prospective Diabetes Study—but recent analysis and a systematic review suggest that intensified glucose control does not in itself reduce cardiovascular mortality in type 2 diabetes and is associated with diminishing microvascular benefit in older people.3 4 These recent findings do not provide an argument for poor glucose control but may allow for some relaxation of treatment goals in terms of HbA1c concentration, with more emphasis on blood pressure and lipid lowering, whose cardiovascular benefits are more securely established.3

Type 2 diabetes is a heterogeneous condition, typically characterised by high levels of circulating insulin in its earlier stages, later followed by progressive insulin secretory failure. As the relative importance of insulin secretory capacity and insulin sensitivity varies from person to person, as does the rate of β cell failure, each case requires individual consideration and discussion with the patient, and “one size fits all” guidelines based only on the HbA1c concentration should be interpreted in this light.

The use of analogue versus human insulin in type 2 diabetes should be considered against this background. Systematic reviews agree that substitution of an analogue for human insulin does not improve overall glucose control.5 6 7 8 9 The rationale for using the short acting analogues is thus based on convenience as they can be injected closer to meal times without deterioration of post-prandial glucose control. No outcome data, however, underpin this advantage. Compared with human insulin, the longer acting analogues confer a modest reduction of the risk of nocturnal hypoglycaemia, but this benefit has been exaggerated by “treat to target” studies, which encourage investigators to aim for very low fasting glucose concentrations, thus increasing the likelihood of hypoglycaemia during the night. In most studies the analogues do not affect the rate of severe hypoglycaemia.7

How safe are the insulin analogues?

The analogues are designed for improved pharmacokinetics and are almost (but not quite) identical with human insulin in other respects. Their adverse effects are therefore closely similar. Reactions at injection site have been reported in about 1% of those using insulin glargine and in about 0.3% of those using other long acting insulins.8 Theoretical concerns have arisen because of altered binding of the newer insulins to their receptors. In vitro insulin glargine shows greater affinity for the insulin-like growth factor 1 receptor than for human insulin,10 although the molecule rapidly loses this property after injection.11 Some pharmacoepidemiological studies have reported an increased risk of breast cancer with insulin glargine, but the ORIGIN trial reported no excess risk of cancer.12

How cost effective are the insulin analogues?

Prescriptions relating to diabetes represent 8.4% of the total cost of prescribing in the UK,13 and insulin accounts for the highest proportion of the total cost of those prescriptions. The incremental cost to the NHS of prescribing analogue rather than human insulin has been estimated at £625m (€796m; $980m).14 The table shows the relative costs of commonly used insulins. There is striking variation in the cost effectiveness estimates for the insulin analogues.5 7 15 The National Institute for Health and Clinical Excellence (NICE) cites a cost effectiveness ratio for glargine of £33 000 (compared with isophane insulin) rising to £43 000 for pen devices. The estimate is based on the assumption of reduced risk of hypoglycaemia and rises to about £10m if the utility gain from hypoglycaemia is discounted. The NICE appraisal concluded that “the face value results all suggested that human insulin regimens were the only cost-effective approach.”5

What are the precautions?

The insulin analogues are used in the same way as isophane insulin, but their altered pharmacokinetics, such as quicker absorption, should be taken into account. There is, for example, an increased risk of hypoglycaemia with the short acting analogues or premixed preparations if an unexpected delay occurs between the injection and the meal. Weight gain is a common complication of insulin treatment, averaging 5.7 kg (plus a 6 cm gain in waist circumference) over three years in one recent trial; those receiving multiple daily injections gained the most weight.16 Use metformin in addition to insulin whenever possible and re-emphasise the need for diet and exercise.

How are the insulin analogues taken and monitored?

Choice of insulin

NICE recommends that human insulin should be the first line choice for type 2 diabetes. This advice is based on efficacy as well as cost and is supported by the observation that hypoglycaemia is no more common in the first few years of treatment with insulin for type 2 diabetes than it is with sulfonylureas.17 The analogues are the second line option for those who need help with injections; those who can achieve adequate control with once daily insulin analogues plus oral agents; and those with troublesome hypoglycaemia. Premixed analogues may be preferred for those who wish to inject shortly before meals.5

The type of analogue chosen does not seem to be important, as their prices are closely matched and the short acting analogues seem indistinguishable. Insulin glargine produces more weight gain than insulin detemir but requires a lower dose, which has cost implications; in other respects they are closely similar.7 18

Metformin enhances the actions of insulin and restricts weight gain and should therefore be continued alongside insulin. Some physicians continue sulfonylureas when starting insulin, although this strategy may lead to increased weight gain and hypoglycaemia.5


Type 2 diabetes is treated with once or twice daily intermediate or long acting insulin, either taken alone or supplemented by short acting insulin before meals (the short acting insulin may be added before each main meal (basal bolus therapy) or before breakfast and the evening meal only, in which case premixed preparations are often used). Human insulin may be administered as once daily intermediate acting isophane insulin (commonly taken with a bedtime snack) or as twice daily isophane insulin or premixed preparations. Thrice daily short acting insulin is used without a long acting insulin supplement in some mainland European countries. Intensified treatment is delivered as human short acting insulin given thrice daily before meals with isophane insulin at bedtime.

In a systematic review and meta-analysis that combined data from trials with human and insulin analogues, a twice daily premixed preparation or thrice daily short acting insulin led to significantly better control of HbA1c concentration than long acting insulin alone. Use of short acting insulin was associated with significantly greater weight gain and more frequent minor, but not major, hypoglycaemia.19 Falls in HbA1c concentration were greater with basal bolus insulin than with long acting insulin alone in the two trials that reported this comparison.19 The 4T (“Treating To Target in Type 2 diabetes”) study found that basal bolus insulin was significantly more likely than premixed preparations to lead to the target HbA1c concentrations, but the differences were relatively small.16 Note that similar results can be achieved by different strategies of insulin administration, and that none emerges as clearly superior. This allows room for a flexible approach based on a target HbA1c concentration and the patient’s preference.


Draft an individualised treatment plan, agree the objectives of treatment with the patient before insulin is started, and monitor the extent to which these have been achieved. Ideally, offer treatment in the context of “insulin start” groups which offer education plus group support and interaction.

Encourage patients to adjust their own insulin and to make use of the feedback provided by glucose monitoring. Monitor weight as well as hypoglycaemia, as some individuals gain as much as 10 kg without any improvement in glucose control. Such weight gain is most likely to occur in those who eat too much and may indeed have been the underlying reason for the unsatisfactory control that resulted in insulin treatment. Dietary evaluation and support, combined with metformin and avoidance of sulfonylureas or pioglitazone, will minimise the risk of excessive weight gain.

Individuals omitting to administer their treatment is one of the leading reasons for poor glucose control among individuals being treated with insulin, as in those taking oral agents for their diabetes; only about 70% of the insulin dispensed for type 2 diabetes is actually used within the intended time frame.20

Beware harm

Although insulin can and should be a life enhancing treatment, injudicious use can easily do more harm than good. Insulin is one treatment option among several in the earlier stages of type 2 diabetes and should be used accordingly. It is not unreasonable to offer the patient a trial of insulin and to be prepared to try another treatment if this does not bring the expected benefit. On the other hand, when other treatments fail, the introduction of insulin in those who really need it must not be delayed.

How do insulin analogues compare with human insulins?

The insulin analogues are “designer molecules” altered to improve the pharmacokinetics of injected insulin (more rapid or more sustained delivery to its target tissues) without changing its biological properties. These advantages are best seen when very tight control is attempted in insulin deficient (type 1) diabetes, but are less apparent when insulin injections are superimposed on endogenous hyperinsulinaemia in type 2 diabetes. Systematic reviews show that:

  • Analogues do not lead to better glucose control in type 2 diabetes

  • There is no difference in major hypoglycaemia (in most studies)

  • Analogues may result in fewer episodes of minor and nocturnal hypoglycaemia

  • Short acting analogues can be injected closer to meals

  • Analogues cost two to four times as much.

NICE accordingly recommends human insulin as first line therapy in type 2 diabetes and analogues as second line therapy for those with special needs or problems.

Tips for patients

  • People with type 2 diabetes often lose the ability to produce their own insulin over the course of time, and 40-50% are likely to end up having to use insulin (this is just because of the natural course of the disease—there is not much you can do to avoid that)

  • The pros and cons of early insulin treatment are still debated by doctors, and you will have some choice in your treatment. But it can harm your health to delay insulin treatment when this is what your body really needs

  • The insulin analogues are widely advertised and are popular for type 2 diabetes. So if you receive advice favouring an older and cheaper type of insulin (“human insulin”) you might assume the advice has been driven by economy rather than benefit. However, it has become increasingly clear, that most people with type 2 diabetes respond equally well to analogue or human insulin. For those with special problems—such as hypoglycaemia (low levels of blood glucose)—and those who are unable to inject more than once a day, the analogues remain useful alternatives

  • Some people prefer to start insulin treatment with a single overnight injection of human NPH insulin, adding injections of quick acting insulin before meals, according to need (this strategy is known as basal bolus therapy). Basal bolus therapy has advantages for those seeking really tight control and/or a more flexible lifestyle. Others begin with two daily injections of human NPH insulin (isophane insulin) or a premixed preparation, which has the advantages of simplicity and convenience

  • Many people find that they need to increase their insulin dose over time to maintain good glucose control and thereby stay on top of their diabetes


Cite this as: BMJ 2012;345:e4611


  • This is one of a series of occasional articles on therapeutics for common or serious conditions, covering new drugs and old drugs with important new indications or concerns. The series advisers are Robin Ferner, honorary professor of clinical pharmacology, University of Birmingham and Birmingham City Hospital, and Philip Routledge, professor of clinical pharmacology, Cardiff University. To suggest a topic for this series, please email us at practice{at}

  • Contributors: The author is the sole contributor.

  • Competing interests: The author has completed the ICMJE uniform disclosure form at (available on request from the corresponding author) and declares: no support from any organisation for the submitted work; no financial relationships with any organisations that might have an interest in the submitted work in the previous three years; no other relationships or activities that could appear to have influenced the submitted work.

  • Provenance and peer review: Commissioned; externally peer reviewed.

  • Patient consent not required (patient anonymised, dead, or hypothetical).