Management of hyperglycaemia in people with acute coronary syndromes (NICE Clinical Guideline 130): uncertainty persists

Cardiovascular disease is the leading cause of mortality in the UK, with approximately 124 000 myocardial infarctions annually.1 An estimated 20% of patients presenting with an acute myocardial infarction are known to have diabetes mellitus, a further 25% with undiagnosed diabetes and 40% with impaired glucose tolerance.2 Thus, up to 85% of patients with an acute coronary syndrome (ACS) have some degree of dysglycaemia at presentation, which persists in a significant proportion of patients at three months.3 The admission blood glucose level following an ACS is a strong predictor of in-hospital morbidity and mortality. A 1mmol/L increase in blood glucose above the normal range correlates with a 4% increase in mortality for non-diabetic patients and 5% for known diabetic patients.4 However, despite this epidemio-logical association, it is unclear whether intensive treatment to lower elevated blood glucose in ACS will reverse this increased mortality. Uncertainty also exists about the optimal method to lower blood glucose in ACS, and whether, how and when patients with ACS-induced hyperglycaemia should be screened for diabetes mellitus. Unsurprisingly, the lack of good quality, randomised controlled trial evidence to answer these fundamental questions has been associated with lack of consensus and substantial variations in practice.5 It is with these issues in mind that the National Institute for Health and Clinical Excellence (NICE) published the short clinical guideline, ‘Management of hyperglycaemia in people with acute coronary syndromes’ (CG130) in October 2011, which is targeted at secondary and tertiary health care professionals treating patients with and without known diabetes who have hyperglycaemia, defined as a blood glucose >11.0mmol/L, in association with ACS.6 The guideline was developed by conducting a systematic review of the evidence, and then evaluating the quality of the evidence using GRADE methodology.7, 8 Where evidence was deficient, recommendations were developed utilising the consensus opinion of the Guideline Development Group (GDG).7 Unusually for a NICE guideline, there were no formal health economic assessments for this guideline as the GDG felt there was a lack of evidence to support the use of intensive insulin therapy, which would clearly be more expensive than standard care, thus rendering an economic comparison inappropriate.7 The guideline contains a total of seven recommendations which are summarised in Table 1. Prior to the publication of this guideline, there was only limited guidance on management of hyperglycaemia in ACS. The Scottish Intercollegiate Guidelines Network (SIGN) guidance 116 (2011) recommends that diabetic patients with marked hyperglycaemia in ACS (>11.0mmol/L) should receive immediate intensive blood glucose control (although there is no specific method stipulated for doing this) and this should be continued at least for 24 hours.9 The American Diabetes Association's Standards of Medical Care in Diabetes (2010) recommend initiating insulin therapy in critically ill patients with hyperglycaemia with blood glucose >180mg/dl (>10mmol/L) and to target a blood glucose range of 140–180mg/dl (7.8–10.0mmol/L).10 European Society of Cardiology (ESC) and European Association for the Study of Diabetes (EASD) guidelines on diabetes, pre-diabetes, and cardiovascular diseases (2007) recommend initiation of an insulin infusion in diabetic patients with a significantly raised blood glucose, who are admitted with acute myocardial infarction, in order to reach normoglycaemia as soon as possible.11 These guidelines also advocate the use of oral glucose lowering agents to manage such patients admitted with a relatively normal level of blood glucose.11 However, the threshold glucose values to define ‘significantly raised’, ‘relatively normal’ and ‘normoglycaemic’ are not specified. The first two recommendations of this NICE guideline (see Table 1) suggest that hyperglycaemia should be controlled but not ‘intensively’ (although the GDG does not define intensive control). The balance of the evidence reviewed by the GDG suggests an overall ‘signal’ towards improved outcomes using intensive control of blood glucose in ACS, with most studies showing neutrality or benefit.6 For non-diabetic patients, observational data from the UK Myocardial Ischaemia National Audit Project (MINAP) and two randomised controlled trials were reviewed.12-14 The MINAP data showed a statistically significant reduction in 7-day and 30-day mortality associated with intensive insulin among ST-segment-elevation myocardial infarction (STEMI) patients, and one randomised controlled trial demonstrated significantly reduced 30-day mortality in Killip class 1 patients.13, 14 Nevertheless, as the quality of the evidence appraised was graded as being low or very low, the GDG appears to have taken the view that this constituted ‘absence of evidence’ which it considered equivalent to ‘evidence of absence’, and recommended against intensive control of blood glucose. In favour of a more conservative approach to management is a perceived risk of hypoglycaemia in ACS, but this was only identified as a problem in the two DIGAMI studies which were considered to be of very low quality,15, 16 and if this guideline is implemented in a coronary care or high dependency setting, intensive observation ought to minimise the risk of iatrogenic hypoglycaemia. The definition of hyperglycaemia is taken as a blood glucose level >11mmol/L, but it is unclear why this value was chosen as some studies considered hyperglycaemia at lower levels of blood glucose, and blood glucose values from 7–11mmol/L cannot be considered euglycaemic. In contrast, a lower threshold for blood glucose in ACS remains undefined by the GDG. As most diabetologists consider 4mmol/L as ‘the floor’, the use of this value would likely have been acceptable to most clinicians. There are no firm recommendations made for the actual method of controlling hyperglycaemia, with a statement saying that intravenous insulin use should not be routine but be considered with problematic control. None of the appraised studies reviewed intervention methods other than intravenous insulin. Among diabetic patients, mortality as an inpatient and for up to 3.4 years after the index event was reduced with intensive insulin control, although not reaching statistical significance.15 Significance was demonstrated, though, in a subgroup analysis of low risk insulin naive patients, with mortality at 1 year (RR 0.48, 95% CI 0.25–0.92) and 3.4 years (RR 0.54, 95% CI 0.35–0.84) reduced compared with standard care.16 This was considered very low quality evidence by the GDG because the study was under-powered for subgroup analysis.15 Intensive insulin therapy is deemed cost ineffective based upon incremental costs of £85.15 in patients without diabetes and £103 among those with diabetes.6 Questions remain as to how the GDG concluded that newly diagnosed diabetic patients in the intensive insulin group required 12–24 glucose strip tests daily compared to 2–4 per day for standard care (40 additional tests over 48 hours, estimated at £86.11 for intensive insulin therapy, versus £0.96 over 48 hours for standard therapy); and for patients with pre-existing diabetes on intensive insulin therapy 12–24 strip tests daily compared with 8–12 for standard care (estimated at £163.49 over 48 hours and £60.46 over 48 hours, respectively).6 The definition of standard care for people with pre-existing diabetes included pre-filled insulin pens, diabetes specialist nurse time and an intravenous cannula but no consideration of any patients being managed on oral hypoglycaemics.6 Also, diabetes specialist nurses were not considered necessary for potential newly diagnosed patients with diabetes on standard therapy. In patients presenting with ACS and hyperglycaemia but without known diabetes, regular formal testing for diabetes is considered to be a primary care matter, with pre-discharge and then annual assessment of fasting glucose and HbA1c.6 As such, the guideline effectively advocates HbA1c as a screening and diagnostic tool for diabetes. There appears to be no evidence base for this in an ACS population. The initial aims of this NICE short clinical guideline were: (1) to optimise hyperglycaemic management through standardisation of treatment thresholds and provide direction on when and how to use intensive insulin therapy (intravenous insulin); and (2) to guide investigation and advice to ACS patients at high risk of developing diabetes. It is disappointing that this guideline was unable to fulfil these aims because of a paucity of appropriate evidence. Indeed, the fact that the only research recommendation suggested by the GDG is to ascertain the optimal management of hyperglycaemia in ACS, the very crux of this guideline, begs the question: was this an appropriate topic for a NICE guideline at this time? The GDG seems to have had an unenviable task working with such limited data and the recommendations are likely to do little to reduce variability in practice in the management of hyperglycaemia in ACS. There are no conflicts of interest declared.