Managing hyperglycaemia in type 2 diabetes

Type 2 diabetes is a progressive disease and treatment of hyperglycaemia requires ongoing monitoring and modification. Many people will need to take oral antidiabetic agents and eventually most will require insulin. In this NPS News we touch on metformin, review the role of the thiazolidinediones (also known as ‘glitazones’) in type 2 diabetes and provide simple guidelines for starting insulin.

Metformin is the drug of first choice

Start metformin in people with type 2 diabetes who have inadequate glycaemic control (glycated haemoglobin HbA1c > 7%) even after improving their diet and exercising for at least 30 minutes 5 times a week. The sulfonylureas (glibenclamide, glimepiride, gliclazide and glipizide) are an alternative if metformin is contraindicated or not tolerated.

In one of the largest clinical trials involving people with type 2 diabetes (UKPDS), metformin was significantly better than diet, insulin or a sulfonylurea at decreasing all-cause mortality, the incidence of diabetes-related complications and stroke among overweight people.1

Use metformin with caution in the elderly, those with a heavy alcohol intake, among people with ischaemic heart disease and heart failure, and in people with impaired liver function. It should be avoided or discontinued in people with impaired renal function (creatinine clearance < 30 mL/min) or a history of lactic acidosis.2,3 A reduced maximum dose of 1 g daily should be used in people with creatinine clearances of between 30 mL/min and 60 mL/min.2

Glycaemic control tends to deteriorate with time.1,4 When monotherapy fails, try combining metformin and a sulfonylurea to bring HbA1c levels under control.

Insulin or a glitazone?

Consider insulin when combining metformin and a sulfonylurea no longer provides adequate glycaemic control. Insulin reduces the risk of diabetes-related complications4 and has a well defined long-term safety profile.

Consider adding a glitazone (rosiglitazone or pioglitazone) for people who have a severe needle phobia or are very reluctant to use insulin, who require assistance from a third party to administer insulin, or whose employment may be threatened if they use insulin. Start insulin promptly if there is no response to the glitazone by 3 months.5

Adding insulin or a glitazone to metformin and a sulfonylurea results in similar improvements in blood glucose levels. In head-to-head trials among people whose glycaemia was inadequately controlled with metformin and a sulfonylurea, rosiglitazone, pioglitazone and insulin all reduced HbA1c levels by around 1% to 2%.6–10 However, the long-term response and safety profile of glitazones compared with insulin is uncertain because all trials had 12 months or less follow up.

The glitazones may also be trialled with metformin or a sulfonylurea for those who cannot tolerate or have a contraindication to one of these drugs. The glitazone should replace the drug that is poorly tolerated or contraindicated. Insulin should be started promptly if the glitazone fails to control hyperglycaemia by 3 months.5

Glitazones effect on outcomes uncertain

As oral antidiabetic agents will be taken for many years, information on the effect of glitazones on clinical outcomes is vital. However, almost all the clinical trials of the glitazones used surrogate outcomes (e.g. changes in HbA1c) instead of clinical outcomes (e.g. mortality, diabetes-related morbidity).

So far, only 2 trials have provided information on the impact of glitazones on clinical outcomes (Table 1).

PROactive investigated the effect of pioglitazone on cardiovascular morbidity and mortality in people with type 2 diabetes and a history of cardiovascular disease (excluding those with moderate or severe heart failure).11 Pioglitazone added to existing therapies significantly improved a secondary outcome of all-cause mortality, myocardial infarction and stroke compared with placebo (hazard ratio 0.84, 95% confidence interval 0.72 to 0.98, p = 0.027).11 But this should be treated with caution as this secondary outcome was not defined in the original protocol and the primary outcome was not significant.

RECORD is ongoing and aims to show that rosiglitazone combined with metformin or a sulfonylurea does not result in more deaths or hospitalisations due to cardiovascular events, compared with metformin and a sulfonylurea.12 An unplanned interim analysis found no significant difference in the risk of cardiovascular death or hospitalisation after almost 4 years follow up.12

Table 1: Results of the primary, secondary and safety analyses of glitazone outcome trials

Source Study population Comparator Primary outcome ProsA ConsA


Pioglitazone in addition to existing medications

(N = 5 238)

Type 2 diabetes and cardiovascular disease

Placebo plus existing medications

Not significant

58 fewer primary eventsB with pioglitazone

57 fewer secondary eventsC

42 fewer hospital admissions for diabetes 38 fewer cases of angina

115 more cases of heart failure 221 more cases of oedema 20 more cases of pneumonia

Significant weight increase


Rosiglitazone and metformin or a sulfonylurea

(N = 4 447)

Type 2 diabetes and no cardiovascular disease

Metformin and a sulfonylurea

Not significant

15 more primary events (cardiovascular death or hospitalisation) with rosiglitazone

21 more cases of heart failure

A. Only significant differences shown.
B. Primary outcome was a composite of death from any cause, non-fatal myocardial infarction (including silent myocardial infarction), stroke, acute coronary syndrome, leg amputation, coronary revascularisation, or revascularisation of the leg.
C. Secondary outcome was a composite of death from any cause, non-fatal myocardial infarction (including silent myocardial infarction), or stroke.

What are the risks of the glitazones?

Heart failure

People with type 2 diabetes are 2.5 times more likely to develop congestive heart failure than people without the disease.13 Glitazones also increase the risk of heart failure. In a large clinical trial, pioglitazone significantly increased the rate of heart failure compared with those receiving placebo (11% vs 8%, p < 0.0001).11 Among people without any pre-existing cardiovascular disease, rosiglitazone doubled the risk of heart failure compared with the sulfonylurea glibenclamide (1.5% vs 0.6%, p = 0.05) and had a non-significant increased risk compared with metformin (1.5% vs 1.3%, p = 0.52).12,14

Do not use glitazones in people with moderate or severe heart failure (New York Heart Association Class III or IV). In patients who are asymptomatic or have only mild cardiac insufficiency, glitazones may be used cautiously and should be started at the lowest dose.15 Prescribers and patients should be alert for symptoms of developing heart failure. Checking weight daily can provide an early warning of fluid accumulation.

Particular care should be taken when adding insulin to an existing treatment regimen that includes a glitazone. Patients should be started on a low dose and monitored closely for signs of oedema or rapid weight gain.

Myocardial infarction

Rosiglitazone may increase the risk of myocardial infarction.16 Do not prescribe rosiglitazone to people with known ischaemic heart disease, people using insulin or nitrates, and use with caution in people with a high risk for cardiovascular events.16

Pioglitazone does not appear to carry the same risk as rosiglitazone.17 However the meta-analysis from which this conclusion is drawn is limited as it relies heavily on the PROactive study (in which the primary outcome did not reach significance) and many of the other included trials were not designed to assess cardiovascular risk.

Weight gain

Weight gain is common to both rosiglitazone and pioglitazone with gains between 2 kg to 5 kg reported in the large clinical trials.11,14,18 Weight gain appears to continue for as long as people take glitazones: in 1 study, after 5 years of follow-up patients had gained 5 kg on average and their weight was still rising.14


For every 100 women taking a glitazone for 1 year, there will be 1 additional fracture.14,19 In contrast to the vertebral fractures commonly seen among women with osteoporosis, these fractures are most likely to occur in the arms (humerus, forearm, hand, and wrist) or lower leg (foot, ankle, fibula and tibia).14,19

Macular oedema

Macular oedema is a known complication of diabetes which may lead to worsening eyesight. There are reports suggesting an association between the glitazones and the development or worsening of diabetic macular oedema.20

Potential liver toxicity

The first available glitazone, troglitazone, was withdrawn after it was found to cause unpredictable acute liver failure. While this risk appears to be significantly lower for rosiglitazone and pioglitazone, several cases of elevated liver enzyme levels, hepatocellular damage, hepatitis and, rarely, liver failure have been reported.21,22

Do not prescribe glitazones to people with liver disease (including transaminase levels increased by more than 2.5 times the upper limit of normal). Perform liver function tests before starting a glitazone and 2 monthly thereafter, and monitor for signs and symptoms of liver toxicity (nausea/vomiting, jaundice, dark urine, right upper abdominal discomfort)23 — with troglitazone, in some cases, normal enzyme concentrations progressed to irreversible liver failure within 1 month.24

Don’t delay using insulin

The progressive failure of the pancreatic beta-cells responsible for insulin production means that over time oral antidiabetic agents will no longer control hyperglycaemia.25

People with inadequate glycaemic control despite making lifestyle changes and taking high doses of metformin and a sulfonylurea (and/or a glitazone) should start insulin without delay. Patients often see this as a major step and should be provided with encouragement and psychological support (Table 2).26 Using the initiation of insulin as a threat to improve patient adherence to diet and lifestyle modifications or to medication use can worsen a patient’s fear of insulin when the time to confront this choice eventually arises.

Table 2: Issues for discussion with patients about to start insulin26–28

Fear Advice

Insulin means I have failed

Type 2 diabetes is a progressive disease. It is the other treatment options that have failed, not the patient

Fear of injections

Insulin is injected into the fat layer under the skin, not into a vein. The fine, short needles available today do not hurt

Starting insulin is too difficult or complex

Insulin may be as simple as one injection at bedtime in addition to existing oral agents

Fear of hypoglycaemia

Severe hypoglycaemia is rare and occurs much less frequently in type 2 diabetes than in type 1 diabetes

I will gain weight

The long-term benefits of insulin outweigh the risks of modest weight gain. Explore ways of improving diet and increasing activity to counter any weight gain

Improving HbA1c in people with type 2 diabetes requires reductions in average glycaemia. Blood glucose levels before main meals are the most important to target when the HbA1c > 7.3%.29 At this level of glycaemia, blood glucose is commonly elevated throughout the day, both before and after main meals. By improving the pre-meal blood glucose, postprandial values will also fall and this is why recent consensus algorithms for type 2 diabetes focus on basal rather than on bolus or rapid acting insulin.5

One of the simplest and safest ways to initiate insulin is to add night time basal isophane insulin to oral antidiabetic agents.27 A simple approach to starting insulin is set out below (Table 3)

Continue oral therapy

Patients should not be automatically taken off their oral antidiabetic agents when they start insulin. Doing so can lead to a rapid rise in HbA1c as the blood glucose lowering effects of the oral agents wear off. Continuing oral agents often means that patients only require a single, daily injection and allows insulin dosage to be titrated gradually until acceptable HbA1c levels are achieved.27,32

Combining metformin and/or a sulfonylurea with a single, daily injection of isophane insulin results in similar levels of glycaemic control to insulin monotherapy, with less weight gain and lower insulin doses.33

Continue using metformin when initiating insulin — sulfonylureas are often also maintained but can be discontinued if hypoglycaemia develops.32

If a patient is taking a glitazone, consider stopping it once they begin using insulin. Continuing the glitazones after initiating insulin treatment increases the risk of fluid retention, heart failure and weight gain.15 Combining rosiglitazone and insulin may increase the risk of an ischaemic event.34

Table 3: Stepwise guide for initiating and adjusting insulin4,26,27,30

Step 1

ADD 10 Units isophane insulin at bedtime. CONTINUE metformin, a sulfonylurea or both (at the same dosage, but no greater than the maximum recommended dose)

  • If evening blood glucose level is high then use 10 units morning isophane insulin.
  • If both morning and pre evening meal blood glucose levels are high then consider using twice-daily isophane.
Step 2

ADJUST insulin therapy gradually every 3–4 days according to fasting blood glucose (FBG) level until target FBG is reached (usually 4.0–6.0 mmol/L)d

Mean FBG (mmol/L)

> 10




< 4

Insulin dose

Increase by 8 units

Increase by 6 units

Increase by 2 units

No change

Decrease by 2–4 units

Step 3

CHECK overall blood glucose control with HbA1c

Step 4

If FBG and evening blood glucose are on target but HbA1c is not, look for hidden ‘hypers’ — blood glucose peaks that occur during the day, often before lunch or after dinner

Options to correct hidden ‘hypers’ include:

  • changing preceding meal size or composition
  • increasing activity after meals
  • adding acarbose
  • adding a meal time rapid acting insuline
D. A GP, trained practice nurse, credentialled diabetes educator or the educated patient can adjust insulin dose according to this titration schedule. Adjusting the insulin dose gradually can gain patient confidence and reduce the risk of hypoglycaemia.4,31
E. Biphasic or pre-mixed insulin are convenient but require a more strict adherence to the size and timing of meals and dosing can be inflexible leading to blood glucose fluctuation.27


The above guide to initiating insulin suggests adjusting insulin by 6 to 8 units for patients with mean FBG levels ≥ 8 mmol/L. We have received feedback that this varies from other local guidelines. This guidance is just one way to initiate and titrate insulin. Prescribers concerned about the risk of hypoglycaemia can adjust insulin in increments up to 4 units (every 3–4 days, according to mean FBG), rather than 6 or 8 units. Alternatively, they can follow RACGP (adjust by increments of 10% to 20%) or local guidelines.

Evidence supports isophane as initial basal insulin therapy

A basal night-time isophane insulin while continuing oral antidiabetic agents is one of the simplest and safest ways to initiate insulin.27

Basal insulins aim to provide a constant level of insulin between meals without producing hypoglycaemia. Isophane lasts 16 to 24 hours with a peak between 4 and 6 hours after injection. Insulin detemir (Levemir) lasts 12 to 20 hours with a peak at 9 hours. Insulin glargine (Lantus) last for 24 hours and has no peak.2

Isophane, glargine and detemir similarly improve HbA1c levels among people with type 2 diabetes.35,36 The risk of severe hypoglycaemia is similar for all three basal insulins.35,36 The risk of nocturnal hypoglycaemia is lower with glargine and detemir than with isophane.35,36 However, there is a lack of information on the long-term safety of glargine and detemir and as such the use of these newer long-acting insulin analogues should be approached cautiously.35,37 Reserve glargine and detemir for people who have symptomatic or nocturnal hypoglycaemia when using isophane.37

Further information about insulin glargine can be found in the December 2006 NPS RADAR.

Fixed-dose combination insulins (also known as bisphasic or pre-mixed insulins) containing isophane and a rapid acting bolus insulin analogue (NovoMix 30, Humalog Mix 25 or Humalog Mix 50) given daily just before the main evening meal are an alternative to basal night-time isophane insulin, particularly in people where the evening meal is their largest. However, this approach increases the risk of post-prandial hypoglycaemia and titration can be difficult as the fixed combination nature of biphasic insulins means that increasing the basal dose also increases the dose of the rapid acting analogue which further increases the risk of post-prandial hypoglycaemia.27

Adjusting insulin doses

Once insulin is commenced, gradual dosage adjustment should be informed by monitoring of blood glucose levels, any hypoglycaemia, and subsequent HbA1c levels. Increase the dose every 3 to 4 days as per Table 3 until the target fasting blood glucose is reached.27

Expert reviewer

A/Prof Stephen M Twigg, Medical Head of Endocrinology Research Laboratories, Royal Prince Alfred Hospital, Sydney, NSW


Dr James Best, General Pracitioner
A/Prof Nick Buckley, Clinical Pharmacologist, The Canberra Hospital
Ms Jan Donovan, Consumer
Dr John Dowden, Editor, Australian Prescriber
Ms Debbie Norton, Pharmacist
Ms Susan Parker, Head of Medical Affairs, Pfizer Australia
Ms Simone Rossi, Editor, Australian Medicines Handbook

Any correspondence regarding content should be directed to NPS. Declarations of conflicts of interest have been sought from all reviewers. The opinions expressed do not necessarily represent those of the reviewers.

  1. UKPDS Group. Effect of intensive blood-glucose control with metformin on complications in overweight patients with type 2 diabetes. Lancet 1998;352:854–65.
  2. Rossi S, ed. Australian Medicines Handbook. Adelaide: Australian Medicines Handbook Pty Ltd, 2007.
  3. Endocrinology Expert Group. Endocrinology. In: eTG complete CD-ROM]. Melbourne: Therapeutic Guidelines Limited, 2004.
  4. UKPDS Group. Intensive blood glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes. Lancet 1998;352:837–53.
  5. Nathan DM, Buse JB, Davidson MB, et al. Management of Hyperglycemia in Type 2 Diabetes: A Consensus Algorithm for the Initiation and Adjustment of Therapy: A consensus statement from the American Diabetes Association and the European Association for the Study of Diabetes Diabetes Care 2006;29:1963–72.
  6. Ko GT, Tsang PC, Wai HP, et al. Rosiglitazone versus bedtime insulin in the treatment of patients with conventional oral antidiabetic drug failure: a 1-year randomized clinical trial. Advances in therapy 2006;23:799–808.
  7. Rosenstock J, Sugimoto D, Strange P, et al. Triple Therapy in Type 2 Diabetes: Insulin glargine or rosiglitazone added to combination therapy of sulfonylurea plus metformin in insulin-naÏve patients Diabetes Care 2006;29:554–59.
  8. Reynolds LR, Kingsley FJ, Karounos DG, et al. Differential effects of rosiglitazone and insulin glargine on inflammatory markers, glycemic control, and lipids in type 2 diabetes. Diabetes Research & Clinical Practice 2007;77:180–7.
  9. Ovalle F, Bell DSH. Effect of rosiglitazone versus insulin on the pancreatic beta-call function of subjects with type 2 diabetes. Diabetes Care 2004;27:2585–89.
  10. Aljabri K, Kozak SE, Thompson DM. Addition of pioglitazone or bedtime insulin to maximal doses of sulfonylurea and metformin in type 2 diabetes patients with poor glucose control: a prospective, randomized trial.see comment]. American Journal of Medicine 2004;116:230–5.
  11. Dormandy JA, Charbonnel B, Eckland DJ, et al. Secondary prevention of macrovascular events in patients with type 2 diabetes in the PROactive Study (PROspective pioglitAzone Clinical Trial In macroVascular Events): a randomised controlled trial. Lancet 2005;366:1279–89.
  12. Home PD, Pocock SJ, Beck-Nielsen H, et al. Rosiglitazone evaluated for cardiovascular outcomes - an interim analysis. New England Journal of Medicine 2007;357:28–38.
  13. Nichols GA, Gullion CM, Koro CE, et al. The incidence of congestive heart failure in type 2 diabetes: an update. Diabetes Care 2004;27:1879–84.
  14. Kahn SE, Haffner SM, Heise MA, et al. Glycemic Durability of Rosiglitazone, Metformin, or Glyburide Monotherapy. New England Journal of Medicine 2006;355:2427–43.
  15. Nesto RW, Bell D, Bonow RO, et al. Thiazolidinedione use, fluid retention, and congestive heart failure: a consensus statement from the American Heart Association and American Diabetes Association. October 7, 2003. Circulation 2003;108:2941–8.
  16. ADRAC. Emerging cardiovascular concerns with rosiglitazone. Australian Adverse Drug Reactions Bulletin 2007; 26: 22. (accessed 18 Dec 2007).
  17. Lincoff AM, Wolski K, Nicholls SJ, et al. Pioglitazone and risk of cardiovascular events in patients with type 2 diabetes mellitus: a meta-analysis of randomized trials. JAMA 2007;298:1180–8.
  18. Home PD, Jones NP, Pocock SJ, et al. Rosiglitazone RECORD study: glucose control outcomes at 18 months. Diabetic Medicine 2007;24:626–34.
  19. FDA Medwatch. Actos (pioglitazone) Tablets. Rockville, Maryland: Food and Drug Administration, 2007.(accessed 28 August 2007).
  20. European Medicines Agency. Press release: European Medicines Agency: Committee for Medicinal Products for Human Use 11—14 December 2005. 15 December 2005. (accessed 31 October 2007).
  21. Diamant M, Heine RJ. Thiazolidinediones in type 2 diabetes mellitus: current clinical evidence. Drugs 2003;63:1373–405.
  22. ADRAC. The glitazones — early experience. Australian Adverse Drug Reactions Bulletin 2003; 22: 6–7 (accessed 21 December 2007).
  23. Drug-induced liver disease. eTG complete CD-ROM] Therapeutic Guidelines Limited, July 2007.
  24. Tolman KG, Chandramouli J. Hepatotoxicity of the thiazolidinediones. Clin Liver Dis 2003;7:369–79, vi.
  25. UKPDS Group. U.K. prospective diabetes study 16. Overview of 6 years' therapy of type II diabetes: a progressive disease. U.K. Prospective Diabetes Study Group. Diabetes 1995;44:1249–58.
  26. Diabetes Australia, The Royal Australian College of General Practitioners. Diabetes management in general practice 2007/8. 13 ed: Diabetes Australia, 2007.
  27. Phillips PJ. Insulin and type 2 diabetes--a simple guide to prevent 'stuff ups'. Australian family physician 2006;35:975–8.
  28. Funnell MM. Overcoming barriers to the initiation of insulin therapy. Clinical Diabetes 2007;25:36–8.
  29. Monnier L, Benichou M, Charra-Ebrard S, et al. An overview of the rationale for pharmacological strategies in type 2 diabetes: from the evidence to new perspectives. Diabetes & metabolism 2005;31:101–9.
  30. Twigg S, Phillips P. Starting insulin in type 2 diabetes. Medical Observer 4 May 2007 2007. (accessed 7 November 2007).
  31. Cheng AY, Fantus IG. Oral antihyperglycemic therapy for type 2 diabetes mellitus. CMAJ 2005;172:213–26.
  32. Wong J, Yue D. Starting insulin treatment in type 2 diabetes. Australian Prescriber 2004;27:93–6.
  33. Goudswaard AN, Furlong NJ, Rutten GE, et al. Insulin monotherapy versus combinations of insulin with oral hypoglycaemic agents in patients with type 2 diabetes mellitus. Cochrane Database of Systematic Reviews (Online) 2004:CD003418.
  34. FDA Medwatch. Information for Healthcare Professionals - Rosiglitazone maleate (marketed as Avandia, Avandamet, and Avandaryl). Rockville, Maryland: Food and Drug Administration, 2007. (accessed 30 November 2007).
  35. Horvath K, Jeitler K, Berghold A, et al. Long-acting insulin analogues versus NPH insulin (human isophane insulin) for type 2 diabetes mellitus. Cochrane Database of Systematic Reviews (Online) 2007:CD005613.
  36. Tran K, Banerjee S, Li H, et al. Long-acting insulin analogues for diabetes mellitus: meta-analysis of clinical outcomes and assessment of cost-effectiveness. Ottawa: Canadian Agency for Drugs and Technologies in Health, 2007
  37. National Prescribing Centre. The role of newer insulins in diabetes. MeReC Bulletin 2007;17(4). (accessed 25 January 2007).