Hyperglycaemia and why it matters

Hyperglycaemia not only has long term implications for the development of complications, but can also pose an immediate threat to life: we aim to clarify precisely what it means, what causes it, what kinds of hyperglycaemia there are, and how they should be managed

he central problem in diabetes is hyperglycaemia, due either to an absolute lack of insulin (type 1 diabetes) or due to a relative lack combined with insulin resistance.

Hyperglycaemia is classed as a plasma blood glucose of 11mmol/l or more. Blood sugar levels between 7 and 11mmol/l are above the homeostatic 'norm' of a non diabetic, and are associated with long term micro and macrovascular complications,^1,2 but they are not normally referred to as hyperglycaemia.

HYPERGLYCAEMIA MATTERS

All hyperglycaemia matters - but sometimes it represents an immediate threat to life and health, while at others it matters less immediately, but is highly significant in terms of the long term risk of micro and macrovascular complications.

Measuring and monitoring blood glucose

Blood sugar is normally measured as plasma blood glucose, and in the UK we measure this in mmol/l (in the US, confusingly, they use g/dl, and one converts from mmol/l to g/dl by multiplying by 18).

Several measures of blood sugar are used in diabetes in order to try to obtain an overall picture of a person's control. These include

• Fasting blood glucose (the level after a fast of at least 8 hours). Diagnosis is based on a fasting plasma glucose of 7.0mmol/l or more
• Random blood glucose (any time)
• Post prandial glucose (the level after eating, conventionally 2 hours after a meal). The International Diabetes Federation recommends a target of 7.8mmol/l and that, in people with normal glucose tolerance, the plasma glucose does not rise above this level.
• HbA1C or glycosylated haemoglobin (the amount of glucose which has stuck to red cells) represents an 'average' snapshot of blood glucose over a three month period and is the measure on which most outcome data are based.^2,3

Glucose is measured in whole blood, plasma or serum. Historically, whole blood values were used, but most laboratories now use serum glucose levels.

What's wrong with sugar?

In the short term, a raised blood sugar may do very little harm, although if it is very raised then symptoms of thirst, itchiness, polyuria and skin and urinary infections may result. However, long term raised blood sugar is harmful. We don't exactly know why, but the more time you spend with your blood sugar above the normal range the more you are at risk of the long term complications of diabetes.^1,2

IMPAIRED GLUCOSE TOLERANCE

The figures below show what happens to blood glucose levels, through the day, when people eat.

In Figure 1, the red line represents the accepted upper level of normal fasting blood sugar, around 7.0mmol/l.

The bottom wavy line represents the normal, non-diabetic individual. That person's blood sugar rises straight after a meal as glucose is absorbed from the diet and enters the circulation. However rapid bolus insulin release from the pancreas acts quickly to lower this. Therefore, although there is a possibility that within the two hours immediately following a meal the blood sugar can rise slightly above 7.0, by two hours after a meal it has fallen again to a baseline well below this.

The middle wavy line represents impaired glucose tolerance. In these individuals the fasting blood sugar levels are still fine but the insulin bolus release, or sensitivity to it, or both, is impaired so the period of raise blood sugar after eating is extended and blood sugar levels may rise well into the diabetic 'zone'.

This is the first step in the development of diabetes.

The area shaded between the curve and the red line represents the time for which these patients are exposed to excess blood glucose and thus exposed to the long-term consequences of this.

These individuals are not considered to have diabetes unless their fasting blood sugar is above 7.0mmol/l or a random reading is above 11.1mmol/l. Otherwise they are classed as having impaired glucose tolerance (IGT). We treat them with dietary measures - not usually drugs - and we do not call them diabetic.

We know that the long-term damage suffered by people with diabetes occurs in the early years of exposure to excess blood sugar levels^4,5 so there is a logic in treating these IGT patients intensively to restore 'normality', not something we normally do. This underlines the importance of building up the whole picture of what is happening in an individual.

DIABETES AND HYPERGLYCAEMIA

In Figure 2, the top wavy line shows the patient with established type 2 diabetes. These patients have an impaired baseline, or fasting, blood sugar as well as prolonged raised periods of blood sugar after meals. The area between the curve and the red line demonstrates that they have a massive exposure to blood glucose, the cause of long-term complications.

The relative proportion of this due to the raised postprandial blood sugar is less,⁶ and in these patients it therefore makes sense to use drugs which lower the whole line, independently of blood sugar levels, first - if possible to an a point when their baseline blood glucose is normal. Then, for fine-tuning, we might address their postprandial surges with glucose dependent products such as glucagon-like peptide agonists (GLP-1s) which won't cause hypoglycaemia but will bring down those peaks.

The usefulness of the fine-tuning is not obvious in patients who have had diabetes for more than 10 years. This is because this group of patients has passed the window in which tight blood sugar control makes a big difference to long-term complications, yet they remain at risk of hypoglycaemia. In these patients we accept a more relaxed HbA1C target of 7.5 mmol/l or more.^7-9 Indeed, the ACCORD trial⁷ suggested that tighter control of diabetes increased mortality - the patients in the ACCORD trial were older, with long established diabetes.

This emphasizes the need for a personalised and individualised approach to treatment.^8,9 Arguably though, to do it we need a better idea of what is happening to individual patients' blood sugar through the day.

GLUCOSE-DEPENDENT DIABETES DRUGS

Some diabetic treatment agents such as GLP-1s only work in the presence of raised blood glucose, while others will lower even a low blood sugar, as their effect is not proportional to blood sugar levels.

Figure 2 shows the consequences of using hypoglycaemic agents that are not glucose dependent, including injected insulin, suphonylureas and glitazones.

If we use too much of these drugs in order to bring down peak blood sugar we risk causing hypoglycaemia, which is also dangerous.

There is an argument, therefore, for quantifying an individual patient's problem accurately.

Some patients, with raised fasting and postprandial sugars, may need to lower their overall blood sugar levels with the 'blunt tools' of insulin, sulphonylureas or gitazones.

However, those with relatively good fasting blood glucose who fail to control their post-prandial glucose excursions by diet alone, might benefit most from GLP-1s and dipeptidyl peptidase-4 inhibitors (DPP-4), the 'gliptins'. Of course these drugs are relatively expensive. Moreover, we can't even tell what's happening to most of our IGT (and early DM) patients because they don't qualify for the blood monitoring kits they need to plot their daily blood sugar patterns.

ACUTE HYPERGLYCAEMIA

Does it matter?

There are several situations in which the practice nurse may encounter hyperglycaemia but not all of these are emergencies.

First to consider is the difference between these two events

1. Random raised blood glucose in well patient

2. Raised blood sugar in unwell patient

HYPERGLYCAEMIA IN THE WELL PATIENT

Patients with diabetes frequently have raised blood sugar when they are otherwise well. They are eating and drinking, but their blood sugar may be well above 11mmol/l - sometimes as high as 25 or 30mmol/l.

The patient who is well but mildly hyperglycaemic may be asymptomatic, although some are aware of subtle changes in the way they feel.

Management of hyperglycemia of this sort is not an emergency, but it does need to be individualised and involves

• working out why the blood sugar is raised
• lowering it in the short term
• improving control in the long term

Some regular testing of pre- and postprandial blood glucose might be helpful in understanding their pattern and improving control.

Treatment

In the short term the patient needs to get some of the excess blood sugar into cells. Using it up is perhaps the healthiest approach. Brisk exercise - a run, a cycle ride or a brisk walk - will lower the circulating blood glucose. The long-term benefits of this approach are obvious. The risk of hypoglycaemia as a consequence is low if the patient self-monitors, and minimal if they are not on glucose-independent medications.

The other options are

• bolus dosing of rapid acting insulin
• general increase in diabetic medication, particularly of that which acts in a glucose-dependent manner.

If you know the patient is running generally high then an overall increase in exercise and/or increase in treatment is needed.

HYPERCLYCAEMIA IN THE POORLY PATIENT

Acute hyperglycemia involving glucose levels that are extremely high is a medical emergency as it can rapidly produce serious complications (particularly fluid loss through osmotic diuresis).

It is most often seen in uncontrolled insulin-dependent diabetes, but may occur in those who are normally well controlled because of intercurrent illness, lack of understanding of diabetes sick day rules, or malfunction of equipment.

Diabetic ketoacidosis

Diabetic ketoacidosis (DKA) consists of hyperglycaemia (above 11.0mmol/l, although it may be very high), ketonaemia (from the breakdown of fats instead of sugar) and metabolic acidosis. It usually occurs when the patient's insulin requirement is not being met. It is more common in people with type 1 diabetics, younger patients and women, and can be the presenting feature of new diabetes. It may occur in patients with insulin pumps if the pump malfunctions, and even experienced type 1 diabetics may assume the symptoms indicate something else. Signs and symptoms of ketoacidosis include an extremely dry mouth, nausea and vomiting, shortness of breath and fruity-smelling breath.

Initially symptoms are of thirst, dry mouth, polydipsia and polyuria but they can progress to nausea and vomiting, abdominal or back pain, confusion and coma. Patients are acutely dehydrated and often toxic. Their breath may smell of ketones (often described as pear drops) and they need urgent admission for fluids to correct their fluid deficit, bicarbonate to correct the acid imbalance, and insulin to get rid of the ketosis. Untreated, DKA can be fatal and indeed in the nineteenth century it almost always was.

Hyperosmolar hyperglycaemic state

Hyperosmolar hyperglycaemic state (HHS) occurs in type 2 diabetes. Patients don't produce ketones in great quantities and don't become acidotic, but are ill. They become dramatically dehydrated, and may develop shock and collapse. Very high blood glucose levels (often over 40 mmol/nl) develop - often due to illness, dehydration, and stopping diabetes medication due to being unwell. HHS is characterised by severe hyperglycaemia with marked serum hyperosmolarity, without evidence of significant ketosis. HHS is a potentially life-threatening emergency, even thought ketosis is absent as the reduced amounts of insulin still present in these patients prevents ketogenesis.

The precipitants are similar to DKA - illness, including infection and myocardial infarction - but some medications can precipitate it including steroids, metformin and illicit drugs such as cocaine and amphetamines. It is also linked to non-compliance.

Patients become very unwell very quickly and may develop stroke-like symptoms and seizures. They are often confused and lethargic - but this can be missed if patients are elderly and on psychotropic drugs as is typically the case. Hypotension and increased respiratory rate are features.

Urine testing shows glycosuria without ketosis, and again, treatment involves urgent hospital admission before a later review with the practice team to identify how this state occurred.

CONCLUSION

All hyperglycaemia matters, but there is a huge difference between the well patient and the ill patient. It is important for all healthcare professionals, particularly those advising patients with diabetes, to understand the difference between life-threatening hyperglycaemia and poor glycaemic control in a well patient.

Beware the unwell, insulin-dependent patient with diabetes and always think hyperglycaemia.

REFERENCES

1. Riddle MC, Yuen KCJ. Reevaluating goals of insulin therapy: Perspectives from Large Clinical Trials. Endocrinol Metab Clin N Am 2012;41:41-56

2. Fowler M J. Microvascular and Macrovascular Complications of Diabetes. Clinical Diabetes 2008;26:77-82

3. Stratton IM, et al. HbA1c, glycated haemoglobin; UKPDS, UK Prospective Diabetes Study. BMJ 2000;321:405-412.

4. The Emerging Risk Factors Collaboration. Diabetes mellitus, fasting blood glucose concentration, and risk of vascular disease: a collaborative meta-analysis of 102 prospective studies. Lancet 2010; 375: 2215-22

5. Age and sex-specific prevalences of diabetes and impaired glucose regulation in 13 European Cohorts. DECODE Study Group. Diabetes Care 2003;26:61-9

6. Monnier L, Lapinski H, Colette C. Contributions of fasting postprandial plasma glucose increments to the overall diurnal hyperglycemia of Type 2 diabetic patients. Diabetes Care 2003;26:881-5

7. Effects of Intensive Glucose Lowering in Type 2 Diabetes, The Action to Control Cardiovascular Risk in Diabetes Study Group (ACCORD). N Engl J Med 2008, 358: 2545-2559

8. NICE Clinical Guideline 87. London. NICE, 2009

9. NICE. Quality standards programme: cost impact and commissioning assessment for diabetes in adults. London: NICE, 2011.