Commentary: controversies in NICE guidance on management of type 2 diabetes

Publication of the original guidelines (in 2002) from the National Institute for Health and Clinical Excellence (NICE) for type 2 diabetes predated the wholesale change in the delivery of diabetes services in the England and Wales. As a consequence of the national service framework in 2001 and the new general practitioners’ contract in 2003, primary care now delivers much more diabetes treatment, with fewer instances of insulin being started in secondary care. Consequently, the revised guidance¹ will now be judged much more on its relevance to general practice diabetes care, including the drive to hit targets on blood glucose control.

Initial management

It is reassuring that a trial of lifestyle intervention with education is still encouraged before a patient is started on metformin, rather than the immediate prescription of metformin at diagnosis as suggested in the consensus document from the American Diabetes Association and the European Association for the Study of Diabetes.² The place of quality assured structured education programmes is rightly emphasised. However, the delivery of these close to the time of diagnosis represents a major logistical challenge to every primary care trust.

Targets for blood glucose control

HbA_1c target levels have remained much the same in the new guidance, with patients being encouraged to maintain their level below 6.5% and with insulin recommended if the level is above 7.4% after oral treatment has failed. However, the guidelines also note that “a single target figure is unhelpful.” This both acknowledges the difficulties of target setting in the real world and reflects the controversy on pursuing low HbA_1c targets with the abandonment of the ACCORD study. This randomised, multicentre study of 10 251 patients with type 2 diabetes mellitus aimed to test the effects of intensive glycaemic control on major cardiovascular events.³ Tight glycaemic control (aiming for HbA_1c <6.0%) increased the risk of death by 20% compared with the group randomised to a HbA_1c 7-7.9%. Although the cause of the excess mortality was unclear, concern was expressed that aggressive insulin treatment may have been implicated in the group receiving tight glycaemic control.

Another recent report highlights the difficulties of trying to optimise blood glucose levels in patients with poorly controlled type 2 diabetes. The 4T study was designed to investigate how to start insulin treatment in patients with type 2 diabetes who are already taking maximal dose metformin and a sulphonylurea.⁴ It showed that prandial, premixed, or basal insulin resulted in similar HbA_1c levels (7.2%, 7.3%, and 7.6% respectively), but no regimen achieved the study’s own (or NICE’s) desired target of <6.5%. However, hypoglycaemia and weight gain were more frequent with the premixed and prandial regimens than with basal analogue insulin.

Currently it seems reasonable to aim for an HbA_1c level lower than 6.5% only if it is safe and feasible to do so—that is, if it is possible to attain target HbA_1c levels with diet, exercise, and conventional treatment but without intensive insulin treatment.⁵

Insulin treatment and new agents

Long acting analogue insulin glargine has long been marketed in primary care as a useful means of reducing hypoglycaemia. Whether insulin glargine or insulin determir (NICE has yet to appraise the role of the latter) unambiguously benefits most people with type 2 diabetes has been questioned.⁶

NICE has recommended intermediate acting human isophane insulin as the default basal insulin on economic grounds. This change may be difficult to implement in practices that are struggling with the rising numbers of patients with diabetes and trying to get to grips with the new agents affecting the incretin system, such as the glucose-like peptide-1 (GLP 1) analogue exenatide, and the gliptins (dipeptidyl peptidase IV (DPPIV) enzyme inhibitors such as sitagliptin and vildagliptin).

NICE also recommends continuing metformin and sulphonylurea treatment when insulin treatment is started, and also suggests that pioglitazone with insulin may be of value. With the increasing complexity of choice beyond traditional oral hypoglycaemic agents, only a careful integrated commissioning process by primary care trusts—engaging specialist diabetologists and nurses to work alongside primary care staff—is likely to fulfil these NICE recommendations and avoid a resurgence of secondary care referral.

NICE recommends thiazolidinedione treatment as second or third line oral treatment in patients in whom insulin is contraindicated and who are not at risk of heart failure or bone fractures. A 12 month trial of exenatide is also suggested as a third line option for obese patients with type 2 diabetes (body mass index >35) in whom insulin would otherwise be started and in whom high doses of insulin may be required. However, owing to the current controversy around possible cardiovascular risks associated with thiazolidinediones,⁷ emerging data on the use of GLP-1 analogues, and the lack of guidance on gliptins, another revision of this document will probably rapidly become essential.