The relationship between impaired symptomatic awareness of hypoglycaemia and an increased rate of severe hypoglycaemia is well established (Hepburn et al., 1990; Gold et al., 1994; Clarke et al., 1995), although affected patients in these studies were not subject to strict glycaemic control. The association between counterregulatory failure and increased risk of severe hypoglycaemia is also well recognised (Ryder et al., 1990). Indeed, counterregulatory failure was proposed as a predictor of risk of severe hypoglycaemia in the subsequent application of intensified therapy (White et al., 1983), and it was not until later that the ability of intensified therapy to cause counterregulatory failure was suggested (Simonson et al., 1985a). It is indeed very important to appreciate that neither asymptomatic nor severe hypoglycaemia are restricted to people using intensified insulin therapy.
Apart from a previous history of severe hypoglycaemia, the greatest risk may be the degree of insulin deficiency, as reflected by the absence of C-peptide (Muhlhauser et al., 1998), as well as the glycaemic control prior to embarking upon intensified therapy and the determination to reach the glycaemic targets (Bott et al., 1994; Muhlhauser et al., 1998). Preservation of endogenous insulin is not affected by intensification of insulin therapy, although there is evidence to suggest that if strict glycaemic control is imposed when diabetes is diagnosed, this may result in more prolonged preservation of endogenous insulin secretion (Shah et al., 1989, The Diabetes Control and Complications Trial Research Group, 1998b). Other factors, related to the patient rather than to treatment, may increase the risk of severe hypogly-caemia, including social class (Muhlhauser et al., 1998) and possibly genetics. In a study from Denmark, much of the risk of severe hypoglycaemia was attributed to ACE genotype (Pedersen-Bjergaard et al., 2003), although this has not been confirmed and has aroused controversy; also, the absence of traditional risk factors in the Danish study is a cause for concern.
The Effects of Intensified Insulin Therapy Upon Risk of Severe Hypoglycaemia
In the DCCT, a clear link was demonstrated between intensified insulin therapy and the frequency of severe hypoglycaemia. In that trial, a three-fold higher rate of severe hypogly-caemia was recorded by the patients in the intensive treatment arm when compared with those on conventional therapy (The Diabetes Control and Complications Trial Research Group, 1991; 1993; 1997). This persisted throughout the entire study, although absolute rates declined gradually in both groups. Furthermore, the risk of severe hypoglycaemia was higher for any given HbA1c, for the people receiving intensive treatment. This phenomenon has not been adequately explained. It is now known that exposure to hypoglycaemia per se can induce defects in counterregulation and loss of subjective awareness of hypoglycaemia (Heller and Cryer, 1991; George et al., 1995; 1997; Davis et al., 1997). It has been assumed that intensive therapy exposes the patient to a greater frequency of mild hypoglycaemia that is sufficient to induce such defects and thereby increase the risk of severe hypoglycaemia by that mechanism. However, methods for delivering intensified diabetes therapy have subsequently improved. Modern methods that focus on transferring skills of insulin adjustment to the patients themselves are reported to achieve improvements in HbA1c with multiple daily injection therapy regimens, without causing more episodes of severe hypoglycaemia, and in their most successful forms achieve a parallel reduction of hypoglycaemia (Jorgens et al., 1993; DAFNE Study Group, 2002; Plank et al., 2004; Samann et al., 2005). The judicious use of insulin analogues in intensified regimens may be associated with slightly less risk of hypoglycaemia (Ashwell et al., 2006), whereas the use of continuous subcutaneous insulin infusion (CSII) with pumps is associated with a much lower frequency of severe hypoglycaemia, and has been used successfully as treatment for patients with problematical hypoglycaemia (Bode et al., 1996, Rodrigues et al., 2005) and in the context of clinical trials (Hoogma et al., 2006).
The Link Between Intensified Insulin Therapy and Risk of Severe Hypoglycaemia
Patients describe symptoms of hypoglycaemia at a wide range of blood glucose concentrations. In an individual patient, the main determinant of the blood glucose concentration at which protective responses commence is probably the recent prevailing range of blood glucose concentration to which the patient has been exposed. For example, when patients with poorly-controlled type 2 diabetes were studied with a controlled hypoglycaemic challenge after blood glucose had been normalised overnight, their epinephrine responses to hypoglycaemia were triggered at higher blood glucose values than in well-controlled patients (Korzon-Burakowska et al., 1998).
As mentioned earlier, the first indication that strict glycaemic control might cause abnormal responses to hypoglycaemia was observed when controlled hypoglycaemia was induced in a small group of patients with type 1 diabetes before, and after, they had been treated with intensified insulin therapy (Simonson et al., 1985a). Following the improvement in glycaemic control, the magnitude of the counterregulatory hormonal response to an abrupt lowering of blood glucose to 2.8 mmol/l was significantly less than observed previously. This study had been planned to investigate the potential of better glycaemic control to restore some of the defects of normal counterregulation that develop in people with type 1 diabetes (see Chapter 6), so these results were unexpected. The importance of these preliminary observations was underlined by a subsequent study in which patients with type 1 diabetes receiving intensified insulin treatment were found to have impaired glucose counterregulation (Amiel et al., 1987). During an intravenous infusion of insulin, most patients were unable to maintain arterialised plasma glucose above 3.0 mmol/l, in contrast with conventionally-treated diabetic patients whose glycaemic control was not as good (as demonstrated by higher glycated haemoglobin concentrations) or non-diabetic volunteers. The intensively-treated diabetic patients were less symptomatic, and although the rise in their plasma epinephrine was of similar magnitude to the other groups, this occurred only when the hypoglycaemia was more profound. Further studies of hypoglycaemia, using a stepped glucose clamp to produce a controlled reduction of blood glucose, confirmed that the symptomatic and hormonal responses started at lower blood glucose concentrations in patients with strict glycaemic control, and were delayed in onset and diminished in magnitude for any given blood glucose concentration (Amiel et al., 1988) (Figure 8.1).
The delayed onset and diminished vigour of symptomatic and hormonal responses to hypoglycaemia in strictly-controlled diabetic subjects offers a partial explanation for the increased
Time -40 -20 0 20 40 60 80 100 120 140 160 180 200 220 240 mins
Figure 8.1 The effect of intensified diabetes therapy (IRx) on epinephrine responses to a slow reduction in plasma glucose over four hours. Copyright © 1988 American Diabetes Association. From Amiel et al., 1988. Reprinted with permission from The American Diabetes Association
Time -40 -20 0 20 40 60 80 100 120 140 160 180 200 220 240 mins
Figure 8.1 The effect of intensified diabetes therapy (IRx) on epinephrine responses to a slow reduction in plasma glucose over four hours. Copyright © 1988 American Diabetes Association. From Amiel et al., 1988. Reprinted with permission from The American Diabetes Association occurrence of asymptomatic biochemical hypoglycaemia. The risk may be particularly manifested when the glycated haemoglobin concentration is reduced to within, or just above, the non-diabetic range (Box 8.1). This was shown in a study of 34 subjects with type 1 diabetes who had a wide range of total HbA: values (Kinsley et al., 1995). They were subjected to a stepped glucose clamp to lower arterialised blood glucose to 2.3 mmol/l and the responses were compared with a non-diabetic control group. Symptomatic responses (particularly autonomic) and some counterregulatory hormonal responses were diminished in the seven diabetic subjects who had a total HbA: of 7.85% or less, i.e., glycaemic control that was within their local non-diabetic range of total HbA:. A very similar study by Pampanelli etal. (1996) produced identical observations in 10 of 33 subjects, whose HbA1c was within the local non-diabetic range, and in whom it was also noted that the onset of some aspects of cognitive dysfunction was delayed. Current evidence would suggest that it is the increased exposure to episodic hypoglycaemia, associated with the treatment strategy that is promoting the problem. Most importantly, a series of studies has shown that hypoglycaemia awareness and counterregulatory hormone responses can be restored in well-controlled diabetic subjects by avoidance of blood glucose concentrations below 3.0 mmol/l in daily life, confirming the circular link between hypoglycaemia exposure and impaired awareness of hypoglycaemia (Fanelli etal., 1993; Cranston etal., 1994).
Thus, although impaired awareness of hypoglycaemia is a major problem in clinical practice, it is by no means exclusively confined to intensified therapy. Although the risk remains greater with lower mean glucose and glycated haemoglobin concentrations, impaired awareness is reversible, at least in the setting of carefully controlled research studies, by scrupulous avoidance of even modest hypoglycaemia in daily life (Fanelli et al., 1993; Cranston et al., 1994). Although this may result in a deterioration of glycaemic control as the problem was reversed, with a rise in mean HbA1c from 6.9% to 8.0% in one small study of seven patients with impaired awareness of hypoglycaemia (Liu et al., 1996), this is not inevitable (Cranston et al., 1994). It is possible for avoidance of hypoglycaemia to result in an improvement of glycated haemoglobin, as post-hypoglycaemia hyperglycaemia is eradicated.
Much less work has been done in type 2 diabetes, although there is increasing evidence that in patients with insulin-treated type 2 diabetes of long duration, the prevalence of severe hypoglycaemia is not greatly different from people with type 1 diabetes (see Chapter 11). Modern trends of starting insulin earlier in type 2 diabetes, when insulin deficiency is not
Box 8.1 Effects of strict glycaemic control in type 1 diabetes
• Reduction in microvascular and macrovascular complications.
• Potential increase in risk of severe hypoglycaemia.
• Diminished counterregulatory and symptomatic responses to hypoglycaemia.
• Altered glycaemic thresholds for activation of responses (i.e., lower blood glucose required).
• Promotion of increased frequency of exposure to hypoglycaemia which exacerbates impaired awareness of hypoglycaemia.
• Tendency to weight gain.
severe, are likely to reduce the overall risk of hypoglycaemia in patients with insulin-treated type 2 diabetes. Recent studies using bedtime basal insulin as the first line of intensifying diabetes treatment for type 2 patients who are not achieving glycaemic targets, have reported a low risk of severe hypoglycaemia, even when using conventional insulins (Yki-Jarvinen et al., 2006). However, caution is indicated when patients require conversion to full insulin therapy. In a small study of poorly-controlled patients treated with oral medication, in which responses to hypoglycaemia were measured before and after improving glycaemic control with insulin, counterregulatory responses and the blood glucose thresholds at which these were initiated were modified, as occurs in type 1 diabetes (Korson-Burakowska et al., 1998).
Was this article helpful?