Answers

Short answers

1 Ascertain what outcomes he hopes for from the consultation, and try to integrate these with your desired outcomes. Reinforce the importance of good management of chronic disease. Offer referrals to structured education and dietetic services.

2 Published evidence supports efforts to lower glycated haemoglobin to 7.0%, and this is reflected by some, but not all, of the national and international guidelines. Evidence for pursuing a more aggressive target of <7.0% is limited, and the two recently published large scale randomised controlled trials studying such an approach gave conflicting results.

3 The patient drives for a living, so for safety reasons he must take extra care to avoid hypoglycaemia. He is currently taking a sulphonylurea (gliclazide), but it might be better to change him to a drug that is less likely to cause hypoglycaemia, such as a thiazolidinedione or a dipeptidylpeptidase-4 (DDPIV) inhibitor. Because he has recently had several hypoglycaemic episodes, he should stop driving until glycaemic control without hypoglycaemia can be achieved.

Long answer 1. Approach
Start by exploring this man’s ideas, concerns, and expectations about his diabetes. Try to understand how he views diabetes and what his priorities are. Only after exploring these matters can the traditional "doctor’s agenda" be meaningfully introduced.

You should make sure that he understands the potential benefits not just of good glycaemic control but also of good control of blood pressure and lipids. Such control reduces the risk of diabetic complications and is central to the management of diabetes, but it can be a difficult message for patients to accept.^{1 2 3 4} The UK Prospective Diabetes Study (UKPDS) graphs of the risk of complications versus glycated haemoglobin are a useful way to illustrate this concept, and they can be downloaded from the University of Oxford Diabetic trials website (www.dtu.ox.ac.uk).

Structured patient education programmes, such as "DESMOND," can help in this regard and can also empower patients to self manage their condition (www.desmond-project.org.uk). Furthermore, recent UK guidance emphasises the importance of structured patient education for newly diagnosed patients and those with established diabetes as "an integral part of diabetes care."⁵ The National Institute for Health and Clinical Excellence (NICE) emphasises that individualised and ongoing dietary advice should be given to all people with diabetes.

The phrase "whether it’s all worthwhile" should raise suspicion of depression. Diabetes is associated with a higher incidence of depression than is seen in the general population,⁶ and in the United Kingdom the General Medical Services (GMS) contract quality and outcomes framework incentivises primary care practitioners to screen for depression in patients with diabetes.

Long answer 2. Lowering of glycated haemoglobin
What do guidelines recommend?
The many national and international guidelines offer different recommendations about glycated haemoglobin targets in type 2 diabetes, and no true consensus exists regarding a single best target. Indeed, UK guidelines do not suggest a single "one size fits all" target. Instead, the NICE blood glucose lowering algorithm advises a target of <6.5% for patients being treated by diet and lifestyle alone or one to two oral hypoglycaemic agents, but it recommends that patients should start insulin, triple oral therapy, or exenatide when glycated haemoglobin rises to 7.5%.⁵ Targets need to be flexible to accommodate individual clinical and social circumstances. The UK primary care GMS contract currently rewards general practitioners whose patients achieve a glycated haemoglobin of 7.5%,⁷ but this will change to 7.0% from April 2009.⁸ The American Diabetes Association/European Association for the Study of Diabetes (ADA/EASD) consensus algorithm recommends that treatment should be intensified when glycated haemoglobin reaches 7.0%,⁹ whereas the International Diabetes Federation advises a target of <6.5%.¹⁰

Our patient’s glycated haemoglobin was <7.5%. Because he lives in the UK, his treatment would not be intensified, whereas if he lived in the United States his treatment would be increased with the aim of keeping his glycated haemoglobin <7.0%.

Evidence to support achieving a glycated haemoglobin of 7.0%
UKPDS showed the benefits of lowering average glycated haemoglobin to 7.0% in the intensive management group compared with 7.9% in the conventional management group.¹ Better glycaemic control significantly improved the following outcome measures: any diabetes related end point, microvascular complications, cataract extraction, retinopathy, and albuminuria. The trial also showed a trend towards reduced myocardial infarction and all cause mortality in the intensively treated group, but these reductions were not statistically significant. Furthermore, intensive treatment with metformin in overweight patients (to achieve similar glycated haemoglobin values to those achieved with intensive treatment with a sulphonylurea or insulin) was associated with significantly reduced risks of any diabetes related end point, diabetes related deaths, all cause mortality, and myocardial infarction.¹¹ Importantly, this benefit of metformin treatment was not associated with a significant increase in weight or risk of hypoglycaemia. This evidence suggests that our patient would benefit from reducing his glycated haemoglobin to 7.0%.

Evidence to support achieving a glycated haemoglobin of <7.0%
The findings of recent trials that aimed for a glycated haemoglobin of well under 7.0% have been contradictory. The ADVANCE trial achieved average glycated haemoglobin values of 6.5% in the intensive treatment group versus 7.3% in the standard treatment group.¹² The intensively treated group showed a significant reduction in the trial’s primary end point—a composite of major macrovascular and microvascular events. This was driven primarily by a reduction in nephropathy in the intensively treated group.

The ACCORD trial aimed for a glycated haemoglobin of <6.0% in the intensive treatment arm,¹³ and it looked at whether or not this ambitious target reduced major cardiovascular events. The trial was stopped prematurely, however, when safety monitoring revealed a significant increase in all cause mortality in the intensively treated group. The reason for this increase is unknown. Hence, at present, we still have insufficient data to support an all encompassing target of <7.0% glycated haemoglobin.

Long answer 3. Problems related to his job
The problem of driving
Our patient drives a taxi for a living. Driving licence regulations differ between countries, but the underlying principle is the same—people with a medical condition that can cause a sudden disabling event at the wheel should not drive. This applies to patients with diabetes who are prone to hypoglycaemia. Furthermore, the medical standards applied to people who drive professionally are obviously much higher.

Our patient should not be driving a taxi at present, and probably should not drive at all. In the UK the Driver and Vehicle Licensing Authority (DVLA) recommends that people with frequent hypoglycaemic episodes that could impair driving should stop driving until control has been re-established.¹⁴ Although "frequent" is not defined, the guidance for drivers with diabetes stipulates that more than one episode of disabling hypoglycaemia within 12 months should be notified to the DVLA. Clinical judgment may be needed to decide the severity of hypoglycaemia likely to impair driving, although any degree of hypoglcaemia has the potential to impair driving.

A change in medication?
It might be wise to change his drugs to ones that are less likely to cause hypoglycaemia. However, not all people taking sulphonylureas who have had hypoglycaemia should have their drugs changed or be advised not to drive. If the cause of hypoglycaemia was avoidable (for example, prolonged fasting) or hypoglycaemia was mild and accompanied by warning signs, sulphonylureas could be continued after educating the patient about the avoidance, recognition, and treatment of hypoglycaemia.

However, drugs such as metformin,¹ thiazolidinediones (glitazones),^{15 16 17 18} and DPPIV inhibitors (gliptins)^{19 20 21 22} all carry a lower risk of hypoglycaemia. Exenatide, an incretin mimetic, also carries a low risk of hypoglycaemia when added to metformin (but a similar risk to insulin when added to metformin and a sulphonylurea). It is also associated with modest weight loss.²³

UK guidelines recommend adding a sulphonylurea as the first line option in patients whose glycated haemoglobin remains 6.5% when treated with metformin, but they also suggest adding a thiazolidinedione instead of a sulphonylurea if hypoglycaemia is a potential problem.⁵ Our patient would certainly fulfil this criterion. UK guidance still in draft form also recommends considering a DDPIV inhibitor in such cases.²⁴ Given our patient’s frustration at having to take so many tablets, combination therapy with a thiazolidinedione or DDPIV inhibitor plus metformin would decrease his current daily tablet burden by six and might improve adherence.²⁵ Although exenatide is a good candidate for our patient, given the relative lack of hypoglycaemia and modest weight loss when added to metformin, he does not fulfil the NICE criteria for starting exenatide. This includes a glycated haemoglobin of 7.5%, despite the use of two oral agents, which would necessitate either triple oral therapy or insulin to improve glycaemic control. In addition, NICE requires patients to have a body mass index of >35 before they start taking exenatide.

Use of the ADA/EASD consensus algorithm would lead to treatment being intensified because his glycated haemoglobin is >7.0%. This algorithm suggests which secondline drug to add to metformin on the basis of their relative merits—sulphonylureas are the cheapest, insulin is the most potent, and thiazolidinediones are least associated with hypoglycaemia—so the addition of a thiazolidinedione to his current regimen would be the best option.

The future
Our patient might have to give up his job if we start him on insulin. We therefore need to explore all other options before starting insulin, but not delay insulin if it is needed. Although the use of insulin may result in loss of occupation, poorly controlled diabetes and the associated risk of complications (especially visual) also pose a risk to his ability to drive safely. Because our patient will want to avoid taking insulin, lifestyle treatment options such as dietary advice and exercise must be emphasised. A large scale randomised controlled trial (RCT) has shown that glycaemic control is more durable with thiazolidinediones than with either metformin or the sulphonylurea glibenclamide.¹⁵ Hence, the most appropriate treatment might be to add a thiazolidinedione to metformin because this combination will probably delay the time until insulin is needed. However, thiazolidinediones are associated with weight gain, so the addition of a DPPIV inhibitor, which is weight gain neutral, is another option. However, it is not clear whether these drugs can provide durable glycaemic control and hence delay the need for insulin as effectively as thiazolidinediones. The thiazolidinedione rosiglitazone has been linked in one meta-analysis with an increased risk of myocardial ischaemia,²⁶ although this has not been confirmed by large scale RCTs.^{15 27} However, an RCT and a meta-analysis of the other thiazolidinedione, pioglitazone, found that this drug had a possible beneficial effect on risk of cardiovascular disease.^{28 29}

Outcome of the consultation
Gliclazide was discontinued and replaced with a twice daily metformin and thiazolidinedione fixed dose combination drug. We chose this combination to minimise the risk of hypoglycaemia, delay the progression to insulin treatment, and decrease tablet burden, with the hope of improving adherence.

Our patient accepted the risk of weight gain and chose this treatment in preference to a DPPIV inhibitor because of the evidence that thiazolidinedione produces durable glycaemic control.

The patient was persuaded to stop driving for eight weeks, after which he declared himself to be free from hypoglycaemic episodes and was advised that he could return to driving. Three months after changing his drugs, his glycated haemoglobin had improved to 7.0%, and his weight was stable.

Cite this as: BMJ 2009;338:b1369