ADS Guidelines for Routine Glucose Control in Hospital (Text Version)

Australian Diabetes Society

Guidelines for Routine Glucose Control in Hospital

2012

Contents

Introduction Page 3

Section 1 Methodology and Process Page 5

Section 2 What glucose target should be aimed for in acute myocardial infarction? Page 6

Section 3 What glucose target should be aimed for in acute stroke? Page 8

Section 4 What are appropriate glucose targets for patients in general hospital wards?

Page 9

Section 5 What special measures need to be undertaken for people on enteral or parenteral nutrition?

Page 11

Section 6 How is steroid‐induced hyperglycaemia best managed? Page 13

Section 7 What is the optimal means of achieving and maintaining glycaemic control in hospitalised patients who are not critically ill??

Page 15

Section 8 How should patients on insulin pump therapy be managed in hospital? Page 16

Section 9 What is appropriate glucose control in end of life situations? Page 18

Section 10 At what level is hyperglycaemia in hospital predictive of diabetes and how should patients with newly discovered hyperglycaemia be followed up?

Page 20

Section 11 What is the role of a specialist diabetes inpatient team? Page 22

Section 12 What routine measures should be undertaken for people with diabetes admitted to hospital?

Page 23

Appendices Page 24

Contributors Page 59

Glossary Page 60

References Page 61

Introduction

Diabetes is estimated to affect 7.4% of the Australian population1, and is increasing annually by 0.8%2. People with diabetes have a higher utilisation of both primary and tertiary health services. In 2004‐05, 9% of all hospital admissions were recorded as having diabetes3. This is likely to be an underestimate as clinical audits from Australia and overseas have found hospital rates of diabetes of 11‐25%4‐9 and furthermore, many cases are undiagnosed at the time10. Australian data indicate that the proportion of people with diabetes as a diagnosis in hospital has been increasing, with a 35% increase in numbers between 2000‐01 and 2004‐053. They also have longer lengths of hospital stay, being about 2 days longer than people without diabetes3,9. The Australian Institute of Health and Welfare has estimated the cost of diabetes to hospital services in 2004‐05 was $371M3.

Diabetes and hyperglycaemia has been shown in a number of observational studies to be associated with poorer outcomes and are markers of morbidity and mortality. Reasons for the increased morbidity and mortality may be related to poor immune response, delayed healing, inflammation and thrombosis associated with hyperglycaemia as well as a higher rate of co‐morbid conditions in this patient group11.

Independent of diabetes, hyperglycaemia per se is also associated with worse hospital outcomes. This is the case whether the person has diabetes or not, but the relationship is stronger for people who do not have diabetes. The relationship between hyperglycaemia and adverse hospital outcomes, in particular mortality, has been clearly demonstrated in many different hospital settings, including myocardial infarction, stroke, general medical and surgical wards, trauma, cardiothoracic surgery, TPN, intensive care, and emergency admissions. For hyperglycaemic people who are not known to have diabetes, it is unclear if the higher mortality is due to the hyperglycaemia, or if the hyperglycaemia is but a marker of underlying critical illness. Most of the high quality studies demonstrating benefit of tight glycaemic control have come from critical care situations, and even these have produced conflicting results.

For patients with hyperglycaemia that is newly discovered in hospital, there is a high probability of undiagnosed diabetes, or future diabetes. However, at present follow‐up is often haphazard, and the opportunity for early diagnosis and treatment of diabetes and thereby prevention of acute and long‐term complications may be missed.

The aim of this document is to provide guidance for the management of hyperglycaemia in a range of hospital situations. The ADS has focused on the management of hyperglycaemia in patients with myocardial infarction and stroke, on general hospital wards, receiving enteral and parenteral nutrition, with steroid‐induced or exacerbated hyperglycaemia, and in end of life situations. The optimal means of achieving tight control, the role of the specialist inpatient diabetes team, inpatient management of insulin pump therapy, and general measures for diabetes management have also been examined. We also provide guidance for the follow‐up of patients with newly discovered hyperglycaemia. The recommendations were based on evidence obtained from systematic reviews where trials had been performed; otherwise they were made by consensus.

It is not the intention of these guidelines to deal with screening for diabetes, the management of diabetic emergencies such as diabetic ketoacidosis, hyperglycaemic hyperosmolar state, and hypoglycaemia, nor do they cover paediatrics, obstetrics or intensive care. Otherwise they should provide guidance for the management of patients with hyperglycaemia in the majority of hospital wards, and are complementary to the Australian Diabetes Society Perioperative Diabetes Management Guidelines.

We sought to achieve concordance in our recommendation to a single target glucose level for the majority of clinical situations, although there are some differences in the limited data for different scenarios. The overall recommendation is that for most hospital patients with hyperglycaemia, treatment should be instituted to achieve and maintain blood glucose (BG) levels below 10 mmol/L, but because of the potential dangers of hypoglycaemia, treatment should not aim to lower glucose levels below 5 mmol/L.

Section 1: Methodology and Process

Systematic reviews were conducted to provide the best possible evidence base for the recommendations. PICO searches of the Cochrane Database for Systematic Reviews, and Pubmed Clinical Queries were undertaken. Systematic reviews, meta‐analyses and existing guidelines relating to our questions were reviewed by a member of the Writing Group, and summarised (Appendix 1). Key cited articles were also reviewed. Where systematic reviews were not available, general searches of the literature were undertaken. The evidence was discussed in an ADS workshop comprising an expert panel of Endocrinologists and Diabetes Educators, held in July 2011. At this workshop, recommendations for each section of the guidelines, and overall recommendations were agreed upon. Where there was little or no evidence, then the committee relied on experience, judgment and consensus to make their recommendations. Issues arising from the discussion, for which there is no evidence base, are included as practice points. The Writing Group drafted this document, which was circulated for further feedback from the participants of the Workshop, and others who were unable to attend.

Section 2: What Glucose Target Should be Aimed for in Acute Myocardial Infarction?

Hyperglycaemia and Cardiac Mortality

Hyperglycaemia is common with myocardial infarction. Data from numerous observational studies show a clear and consistent association between the initial admission glucose level and infarct outcomes, in particular mortality. A meta‐analysis by Capes et al12 showed that amongst patients without diabetes, those with an admission blood glucose level (BGL) ≥6.1‐8.0 mmol/L had a 3.9 fold (95%CI 2.9‐5.4) higher risk of death than those with lower BGL. For patients with diabetes, those with a BGL ≥10‐11.0 mmol/L had a 1.7 fold (95%CI 1.2‐ 2.4) increased risk of death. The majority of studies in this publication were performed in the pre‐thrombolytic era, but newer publications show similar results (Appendix 2, Table 2.1). Virtually all have shown a dose relationship and a glucose threshold for increased mortality of around 6‐8 mmol/L. In addition, there are observational data demonstrating a relationship between glucose levels in the first 24 hours after myocardial infarction and mortality (Appendix 2, table 2.2). These indicate that persistent hyperglycaemia, even if mild, is also associated with increased mortality following myocardial infarction.

Hypoglycaemia

Most studies have concentrated on the relationship between hyperglycaemia and increased mortality. There are also some data that hypoglycaemia is associated with adverse outcomes, with a U‐shaped relationship being described in several observational studies15,23,25. The increased risk was seen in patients with admission BGLs ranging from <3.3 to <7 mmol/L. In the DIGAMI Study where there was active lowering of glucose, there was no increase in mortality or other major outcomes amongst subjects who developed hypoglycaemia <3 mmol/L, after adjustment for confounding variables31.

Clinical Trial Data and Existing Recommendations

Five systematic reviews with specific analysis (in some cases subanalysis) of whether tight glucose control in myocardial infarction improves survival were identified32‐36. One older systematic review which predated a number of the more recent trials with negative results, found a reduction in mortality with tight glucose control32. A more recent review suggested that tight glycaemic control can reduce mortality but did not make this conclusion on the basis of a meta‐analysis35, whilst another one decided that the evidence is inconclusive34. Two recent high quality systematic reviews concluded that tight glycaemic control did not reduce mortality33,36, but one included cardiac conditions other than myocardial infarction.

Four of the randomised controlled studies identified in the systematic reviews had set specific glucose targets for their intervention (Appendix 2, Table 2.4)28,31,44,48. There was improvement in survival in the intensive treatment arm only in the oldest of these studies, where the glucose target was 7‐10 mmol/L31. It has been postulated that the failure to demonstrate an effect in the more recent studies may be due to i) failure to

achieve a large enough differential in glucose levels between the arms of the study, ii) glucose levels in the control arm being only minimally elevated, iii) the advent of modern treatments for AMI (PTCA, thrombolysis, anti‐platelet therapy, beta‐blockade, statin therapy), overwhelming any benefit of glucose control53. Existing guidelines covering glucose control in myocardial infarction have given diverse recommendations (Appendix 2, Table 2.5)54‐57. Two of the 4 guidelines did not have specific recommendations for myocardial infarction, but encompassed myocardial infarction within broader guidelines for hospital glucose control55,57. Two of the guidelines recommended target BGs <10 mmol/L55,56, one recommended “normal” levels54, and one recommended against tight control57.

Conclusions

Observational data indicate a clear association between hyperglycaemia and mortality in myocardial infarction. However, only one RCT of patients with myocardial infarction has shown a benefit of glycaemic control, with a glucose target of 7‐10 mmol/L. In the other studies, no mortality benefit of tight control was seen. Despite this, most professional organizations have recommended a glucose target of <10 mmol/L, provided that this can be achieved safely.

Recommendations and Practice Points

1. Patients admitted to hospital with myocardial infarction who have hyperglycaemia, should be treated to achieve and maintain glucose levels less than 10 mmol/L.

2. Hypoglycaemia must be avoided. It would be prudent to avoid treatment which lowers the glucose below 5 mmol/L.

3. Insulin infusion therapy may allow for tighter targets but this requires frequent monitoring and high level staff training.

Section 3: What Glucose Target Should be Aimed for in Acute Stroke

Hyperglycaemia and Stroke Mortality Data from numerous observational studies show an association between initial glucose levels and outcomes of stroke, in particular mortality. Another meta‐analysis by Capes et al showed that amongst patients without diabetes, those with an admission BGL ≥6.1‐8.0 mmol/L had a 3.07 fold (95%CI 2.50‐3.79) higher risk of death than those with lower BGL58. There was no increase in risk amongst patients with diabetes at these levels (RR 1.3, 95%CI 0.49‐3.43) increased risk of death. Mortality from haemorrhagic stroke mortality was not associated with admission hyperglycaemia. More recent publications show similar results (Appendix 3, Table 3.1). Observational data also indicate that there is a relationship between glucose levels during the first 24 hours after stroke and mortality or infarct size (Appendix 3, Table 3.2). Clinical Trial Data and Existing Recommendations The 3 systematic reviews examining studies of tight glucose control in stroke came to divergent conclusions (Appendix 3, Table 3.3)36,75,76. Although none of the studies reviewed demonstrated a benefit of glucose control, one review recommended insulin therapy if glucose levels exceed 10 mmol/L75. There were 7 randomised controlled trials of tight glycaemic control for stroke. One had a large sample size but was discontinued early due to slow recruitment and failed to demonstrate a benefit of glucose control78. Most of the other trials were more of a pilot nature (Appendix 3, Table 3.4). An additional recent Australian study where there was a glucose control target of 4‐8 demonstrated a 16% reduction in mortality with the intervention arm85. However glucose control was only one of 3 factors in the intervention package (the others being management of swallowing and fever), and it is difficult to determine the contribution of glucose control to the outcome. This study had not been included in any of the above systematic reviews. Two sets of stroke guidelines which provided some recommendations regarding glucose control were identified (Appendix 3, Table 3.4). Both suggested aiming to keep BGs below a level around 10 mmol/L, but admit that the evidence for this is weak. Conclusions Observational data indicate a clear association between hyperglycaemia and mortality in acute thrombotic stroke. There is a lack of clinical trial evidence regarding appropriate glucose targets in stroke, and the recommendation is made on the basis of extrapolation from other clinical situations, and consensus. Recommendations and Practice Points

1. Patients admitted to hospital with acute thrombotic stroke who have hyperglycaemia, should be treated to achieve and maintain glucose levels less than 10 mmol/L.

2. Hypoglycaemia must be avoided, and therefore it would be prudent to avoid treatment which lowers the glucose below 5 mmol/L.

Section 4: What are Appropriate Glucose Targets for Patients in General Hospital Wards?

Hyperglycaemia and Complications in General Hospital Wards

A number of observational studies have demonstrated an association between glucose levels and adverse outcomes in patients in general hospital wards. These have shown a higher risk of adverse outcomes above a random glucose level of 8‐12.2 mmol/L (Appendix 4, Table 4.1). The adverse outcomes include infection, mortality, and longer length of stay. There is also a dose relationship between glucose levels and mortality91‐93. The relationship between hyperglycaemia and mortality in the general wards is much stronger among those with newly discovered hyperglycaemia than among those with known diabetes.

Systematic Reviews and Existing Guidelines

Three systematic reviews have examined clinical trials of tight glycaemic control outside of the intensive care situation, and not specifically focusing on myocardial infarction or stroke (Appendix 4, Table 4.2). Most studies included in these reviews were in the perioperative context, or included subjects with myocardial infarction. The findings have been mixed, with one review finding a reduction in mortality with tight glycaemic control with cardiac surgery94, one finding no benefit in the non‐ICU or peri‐operative settings36, and a third finding a reduction in infection rate only95. There is a recent study in general surgical patients which found that treating to a pre‐meal glucose target of <7.8 mmol/L with basal, bolus and supplemental insulin resulted in better glycaemic control and fewer wound infections and total complications than using sliding scale insulin with the same target104. However this study was designed to compare the 2 insulin regimes, rather than the effect of treating to their target. No trials have as their primary objective, examined the effect of treating to specific glucose targets in general medical wards.

Three existing guidelines for glucose control in non‐critically ill hospital patients have recommended glucose levels below 10 mmol/L (Appendix 4, Table 4.3)55,105,107. A fourth guideline maintains that there is no evidence for strict control in non‐ICU patients106. The American Association of Clinical Endocrinologists / American Diabetes Association and Endocrine Society of America guidelines also recommend pre‐meal glucose levels of 3.9‐7.8 mmol/L, without giving the rationale for different pre‐meal and random glucose targets55,107. The caveat that these should only be the targets if they can be safely achieved has also been stated.

Conclusions

As the evidence is limited, our recommendations are based on existing guidelines and extrapolations from other clinical situations. Having the same glucose targets as for myocardial infarction and stroke was considered important for uniformity across the hospital, and to avoid confusion. Although one would not regard glucose levels as being in the hypoglycaemic range until they are below 4 mmol/L, active intervention

should not aim to reduce the glucose levels below 5 mmol/L, which allows for an added margin of safety. If aiming for tight glycaemic control, frequent glucose testing is required.

Recommendations and Practice Points

1. Most patients in general hospital wards with hyperglycaemia should be treated to achieve and maintain glucose levels less than 10 mmol/L.

2. Hypoglycaemia must be avoided. It would be prudent to avoid treatment which lowers the glucose below 5 mmol/L.

3. To achieve tight glucose control safely, frequent glucose monitoring is recommended

Section 5: What Special Measures Need to be Undertaken for People on Enteral or Parenteral Nutrition?

Hyperglycaemia and Enteral and Parenteral Feeding

Hyperglycaemia is a common occurrence in patients receiving nutritional support either in the form of enteral or parenteral nutrition. The specific effect of hyperglycaemia on clinical outcomes in patients receiving nutrition support has only been reported by one observational study. A retrospective study of 111 patients receiving total parenteral nutrition (TPN) found that increased blood glucose levels were associated with an increased risk of cardiac complications, infection, sepsis, acute renal failure and death91. Those receiving TPN with mean glucose levels >9.1 mmol/l had a 10‐fold greater risk of mortality than those with mean glucose

levels ≤6.9 mmol/l. This association was independent of age, sex and presence of pre‐existing diabetes. This adds further weight to the overwhelming evidence of a clear relationship between high blood glucose levels and adverse outcomes in critically ill or hospitalised patients, as reviewed in the earlier sections of this guideline.

A major goal in the management of patients with diabetes receiving nutritional support is the achievement of good glycaemic control, avoiding both hyperglycaemia and hypoglycaemia, with their associated risks of fluid imbalance and dehydration, ketoacidosis and hyperosmolar coma, infection and neurological events. However, how best to achieve good glycaemic control in these patients remains unclear. A critical factor for consideration is where the patient will be cared for: in the ICU or general ward. Other important considerations include the method of nutritional therapy (enteral vs parenteral) and composition of the feeds particularly carbohydrate/dextrose content. In general, diabetic enteral formulas (low carbohydrate high monounsaturated fatty acid formulas) are preferable to standard high carbohydrate formulas in patients with diabetes107. Close monitoring of BGLs and review of diabetes management is essential when enteral/parenteral feeds cease and oral intake resumes.

Clinical Trials

No studies investigating the effects of oral glucose lowering agents on blood glucose levels and outcomes in patients receiving enteral or parenteral nutrition were identified. There are 2 studies, both of poor quality and at high risk of bias, which have investigated the effects of different insulin regimens in patients receiving enteral nutrition (Appendix 5, Table 5.1), but none in the situation of parenteral nutrition. One compared the effects of sliding scale insulin to sliding scale insulin and regular subcutaneous glargine insulin, showing no differences in blood glucose levels, adverse outcomes or length of stay108. However, a significant proportion of the patients in the sliding scale alone group also received NPH insulin during follow up. This suggests that a basal insulin on top of a correctional insulin regimen, has a role in achieving adequate glycaemic control in patients receiving enteral nutrition. A second (nonrandomized) pilot study with a retrospective control group found that a basal bolus insulin protocol achieved lower mean glucose levels than a variable dose preprandial

insulin regime, at the expense of a small increase in hypoglycaemia109. The nurse led insulin protocol was implemented in the ICU setting which limits its generalisability.

Conclusions

On the balance of the limited evidence, insulin therapy is likely to be the most effective agent for immediate control of blood glucose levels in patients receiving enteral and parenteral nutritional support. The recommendations made are based on experience and consensus.

Recommendations and Practice Points

1. Individualised nutritional plans should be provided as insulin therapy will depend on the nature of the feeding cycle.

2. Sliding scale insulin should not be used alone to optimize glucose control in patients receiving enteral or parenteral nutrition.

3. Insulin therapy should include regular basal insulin (intermediate or long acting insulin) with prandial and correctional insulin if required.

4. Perform BG testing 4‐6 hourly. With bolus enteral or parenteral nutrition perform BG testing before each bolus is given.

6. Patients with unstable metabolic control or variable parenteral feeding may benefit from an intravenous insulin infusion therapy.

7. Close liaison with the dietitian or team managing the enteral or parenteral nutrition is critical particularly if calorie intake is changing, as insulin doses will need to be adjusted.

Section 6: How is Steroid‐Induced Hyperglycaemia Best Managed?

Prevalence and risk factors

Hyperglycaemia is common amongst inpatients who are receiving glucocorticoids (GC), with reported incidences of 64‐71%110,111. Risk factors for development of hyperglycaemia amongst inpatients include a pre‐ existing diagnosis of diabetes110,112, higher HbA1c113, increasing age111, steroid dose114, and family history of diabetes115, 116.

There is little data on temporal BG profile of individuals receiving GC. An open prospective observational trial performed on acute hospital wards examined the interstitial glucose profiles of pts admitted with COPD treated with at least prednisone 20mg/day as compared to pts with COPD, not known to have diabetes, admitted for another indication who did not receive GC117. Patients receiving GC in the morning had higher BGLs in the afternoon and evening, as compared to those not receiving GCs (with the greatest elevation seen in those with known diabetes). A rise in fasting glucose is also seen when extremely high dose GC (e.g. methylprednisone 250‐1000mg/day) are administered113. Based on ambulatory data, the effect of GC on BG profile is rapid, with a change seen within 2‐3 hours of administration of GC118,119. This is also rapidly reversible, in that lower glucose levels are seen on GC free days in patients who receive alternate day GC120.

Screening for development of hyperglycaemia and monitoring in those with DM

Prior to or upon the initiation of GC, it is prudent to exclude the presence of undiagnosed diabetes through measurement of serum glucose (see section 11). Screening for development of steroid‐induced hyperglycaemia by afternoon fingerprick BG assessment is likely to detect the development of most cases of hyperglycaemia112, and twice daily GC induced hyperglycaemia should still be detected. Reliance on fasting glucose is inadequate. If hyperglycaemia is detected, BG monitoring should occur as per the general diabetes protocol.

Management of glucocorticoid induced hyperglycaemia

There are no prospective trials on the use of any anti‐diabetic medication for the management of GC related hyperglycaemia. The limited observational data are outlined in Appendix 6, Table 6.2. Sulphonylureas have a limited role in the treatment of steroid‐induced hyperglycaemia in hospital. There are reports of thiazolidinedione use in the setting of organ transplantation, but these agents are also unsuitable for most patients in hospital. The management of new onset diabetes after transplantation has been addressed in other guidelines140 and will not be further discussed in this document.

Although there are no trials of its use in steroid‐induced hyperglycaemia, insulin is considered to be the agent of choice for the management of steroid‐induced hyperglycaemia in hospital. Benefits provided by insulin include greater dose flexibility, more rapid onset of action and titration and that there is usually no dose ceiling as compared to other glucose lowering agents. Insulin dose requirements will always need to be individualised, and require pre‐emptive titration as the GC dose is adjusted, usually on a daily basis. The insulin

regimen should predominantly target post‐prandial control, and with morning GC administration, the afternoon hyperglycaemia. The use of isophane insulin for management of steroid‐induced hyperglycaemia has been advocated, with the initial dose determined according to GC dose and patient weight124,139. Isophane type insulin can be supplemented with ultraquick insulin analogue with meals139. With twice, thrice or 4 times a day GC regimens, isophane insulin twice daily with prandial rapid acting analogue can be initiated. A regime that controlled glycaemia on previous occasions can be re‐initiated, e.g. when cyclical GCs are required, as long as there has been no major interval change in weight or renal function. For those with pre‐existing insulin requiring diabetes, a pre‐emptive increase in insulin will be required, and further adjustment based on blood glucose response.

Recommendations and Practice Points

1. In patients receiving glucocorticoids, undiagnosed diabetes should be excluded. Those free of diabetes should be screened for the development of hyperglycaemia by random blood glucose monitoring performed in the afternoon following morning administration of GC.

2. Hyperglycaemia is best managed with insulin: basal insulin as isophane type insulin, and rapid acting analogue with meals as required.

3. In individuals already on insulin the likely need for increased insulin should be recognised. Dose requirements need to be individualised and require daily review.

Section 7: What is the Optimal Means of Achieving and Maintaining Glycaemic Control in Hospitalised Patients who are not Critically Ill?

This section examines the optimal methods for achieving and maintaining good routine glycaemic control in hospital. It does not discuss the use of insulin infusion therapy, or perioperative management. For the latter, we refer the reader to the Australian Diabetes Society Perioperative Diabetes Management Guidelines141.

There is a paucity of data in the non‐critically ill patient group as to the best method of maintaining glycaemic control. This group of patients differs greatly from those critically ill as they are often eating. Intensive insulin therapy has been shown to be beneficial in a critically ill patient population, but there have been no studies evaluating outcomes in general medical wards. The main adverse event with intensive subcutaneous insulin therapy is hypoglycaemia which can be quite severe.

Intensive insulin therapy requires frequent monitoring and should not just be reactive to changes in glucose loads, e.g. food. Its application requires a specific skill set for staff to maintain. Traditionally sliding scales have been used to maintain blood glucose levels in non‐critical hospitalized patients. This method of injecting a set dose of insulin at set times is often reactive to high levels of blood glucose. BGs often fluctuate from high to low, which can potentially be detrimental. Sliding scale administration of insulin is not recommended, and American guidelines recommend that an insulin regimen with basal, nutritional and supplemental (correction) components be utilized for hospitalised patients with diabetes or stress hyperglycaemia142.

There are few studies that have examined different subcutaneous insulin regimens in non‐critical hospitalised patients (Appendix 7). Most studies have moderate to high risk of bias and outcome measures have been inconsistent between the different studies. Basal bolus regimens have been shown to be superior to sliding scale regimens for glucose control102,104, and sliding scale insulin alone has been no more effective than continuation of the patient’s usual diabetes medication101. Effective use of basal bolus insulin requires frequent and regular blood glucose monitoring (at least 4 and preferably 6‐8 times daily). Based on clinical expertise, current practices and the limited literature, the following consensus recommendations were made.

Recommendations and Practice Points

1. Sliding scale insulin should not be used to optimise glucose control in the inpatient general medical or surgical setting.

2. Oral hypoglycaemic agents or pre‐mixed insulin can be used in certain stable hospitalised patients who are eating regularly. Supplemental insulin should be written up in addition.

3. Insulin therapy in hospitalised patients should otherwise consist of a basal insulin, prandial and supplemental insulin.

Section 8: How Should Patients on Insulin Pump Therapy be Managed in Hospital?

Continuous Subcutaneous Insulin Infusion Therapy in Hospital

Continuous subcutaneous insulin infusion (CSII) or insulin pump therapy is used in the management of growing numbers of patients with Type 1 diabetes in Australia. Anecdotal reports suggest that patients established on CSII usually prefer to continue on their pumps during hospital admissions. Hospital health care providers will increasingly be faced with the issue of how to manage such inpatients. A number of publications have detailed guidelines regarding inpatient management of patients previously established on CSII144‐147. Whilst there are no data from randomised trials available, observational reports indicate that patients admitted to hospital continued on CSII who are managed with best‐practice consensus protocols fare at least as well as those changed over to subcutaneous insulin injections and managed by the endocrinology team. The data regarding hypoglycaemia is conflicting with one study indicating a lower incidence in those inpatients continued on CSII which was not confirmed with a subsequent study148,149. A caveat is that these reports have stemmed from tertiary academic medical centres in the United States and their applicability to a spectrum of hospitals (including community hospitals) in Australia is yet to be determined. The recommendations below are based upon a consensus opinion.

Management of CSII in Hospital

General recommendations for CSII therapy in hospital are outlined in Appendix 8, Table 8.1. In appropriate circumstances, CSII may be the preferred method of insulin delivery. However, device operating menus and programs vary not only according to the manufacturer but also from model to model. It is highly unlikely that non‐specialised medical and nursing staff will be familiar with the operation of all available devices. We therefore recommend that CSII therapy is be continued in hospital only in those situations where the patient (or guardian) has the ability to safely self‐manage their insulin dosing and the pump. The competency requirements are outlined in Appendix 8, Table 8.2. If these criteria are not met CSII must be substituted with either a subcutaneous insulin regimen or an intravenous insulin infusion. Contraindications to CSII therapy are listed in Appendix 8, Table 8.3. All aspects of CSII management should be documented (Appendix 8, Table 8.4) and it is recommended that the Endocrine team be involved.

CSII and Surgery

Surgery itself is not an absolute contraindication to continuation of CSII. If CSII is to be continued intra‐ operatively this decision must be made in conjunction with the anaesthetist, surgeon/proceduralist, and endocrinology team with the documented consent of the patient or their guardian. CSII and an intravenous insulin infusion should not be used simultaneously for any extended period150. The situations appropriate for intraoperative CSII or for its substitution with an intravenous insulin infusion are outlined in Appendix 8, Table 8.5. When CSII is being used intraoperatively, it is important for there is a protocol for its management (Appendix 8, Table 8.6.). Appropriate overlap and timing is important when switching a patient from CSII to insulin infusion or multiple subcutaneous insulin injections, and vice versa (Table 8.6.).

Recommendations and Practice Points

1. In general, CSII should be continued in hospital where the patient can competently and safely self‐ manage the pump and self‐dosing.

2. Details of pump therapy should be documented, and supported by the endocrine team

3. CSII may be continued for short operative procedures if those responsible for the patient’s intraoperative care are comfortable with its use.

Section 9: What is Appropriate Glucose Control in End of Life Situations?

Diabetes and End of Life

For patients with diabetes and advanced disease limiting their life expectancy there is no body of evidence available regarding the impact of tight glycaemic control on outcomes. Life‐limiting disease includes, but is not limited to, cancer and includes any disease process such as advanced dementia, end stage cardiac and respiratory failure, which is incurable and significantly shortens the patient’s life expectancy. As the patient with diabetes approaches the end of their life the guidelines regarding glucose monitoring and glycaemic targets detailed earlier in this document may no longer be appropriate with a potential for discomfort, inconvenience and significant morbidity relating to hypoglycaemia. Tight glycaemic control is questionable benefit and the avoidance of long‐term complications is no longer relevant. Conversely it is important to maintain a level of glycaemia to prevent hyperglycaemia associated thirst, dehydration, polyuria associated urinary frequency, altered conscious state and symptomatic hypoglycaemia. Treatment regimens need to be individualised according to the patient’s circumstances.

Palliative care is defined as medical or comfort care that reduces the severity of a disease or slows its progress rather than providing a cure. Currow et al151 have described 4 phases in the end of life pathway: Stable, unstable, deteriorating, and terminal (see Appendix 9, Table 9.1 for details). Palliative patients may be admitted to hospital for management of an acute illness, either intercurrent or related to their primary underlying disorder or for terminal care. There is an absence of level I data though there are a number of valuable consensus based guidelines addressing the global management of palliative patients with diabetes151‐ 153. The following represents a consensus of opinion in the absence of a suitable evidence base, and is in part based on the 2010 Guidelines for Managing Diabetes at the End of Life152. This consensus document focuses on the inpatient management of hyperglycaemia in those patients with diabetes deemed as requiring palliative care. As management should be individualised to each patient’s needs this document provides general principles for the inpatient management of palliative care patients with diabetes and detailed protocols cannot be provided.

Glucose Management in End of Life Situations

Glucose management during inpatient admissions will depend on the type of diabetes and the phase of the end of life pathway (see Appendix 9, Table 9.2 for details). In general, in the earlier stages of end of life, the person’s usual diabetes medication would be continued, with adjustments based on the many situational factors which would affect glycaemic stability (Appendix 9, Table 9.3). The decision to simplify and rationalise treatment regimes and targets would need to be made on an individual basis. As the person progresses through the phases of end of life, the emphasis shifts towards maintenance of comfort, with corresponding reductions in medication and glucose testing, and some liberalisation of food restriction. This does not imply a nihilistic approach in the metabolic management of palliative patients. Avoidance of marked hyperglycaemia is still relevant, particularly in hospital, to avoid symptoms of hyperglycaemia, and improve wound healing and

resistance to infection. Hypoglycaemia must also be avoided. With type 1 diabetes, ketoacidosis is likely to precipitate death, so it should be prevented until a decision is made to withdraw all treatment in the terminal phase. Therefore until then, some glucose testing and insulin administration may remain necessary. It is reasonable to continue on insulin pump therapy in those patients established on these devices.

The views of the patient and their family need to be determined. They may require advice and counseling regarding the management of the patient’s glucose levels as many years may have been spent where glucose levels have been diligently maintained in a target range. The realisation that long‐term survival is no longer a viable proposition and that maintenance of tight glycaemic control is of dubious value and could even adversely impact quality of life can be confronting. Ultimately the decision of the patient and their family will take precedence. The status of the patient may be evolving continuously requiring the ongoing reassessment of glycaemic management strategies by the medical team.

Recommendations and Practice Points

1. Palliative care patients may still benefit from a level of glucose control in hospital so diabetes treatment remains relevant.

2. The level of intervention would generally be less intensive than for other hospital patients, and needs to be individualised, depending on the phase of end of life, and other situational factors.

Section 10: At What Level is Hyperglycaemia in Hospital Predictive of Diabetes and How Should Patients with Newly Discovered Hyperglycaemia be Followed up?

Stress Hyperglycaemia

Patients with a known history of diabetes commonly have hyperglycaemia in hospital, but patients without a history of diabetes may also be found to have elevated blood glucose levels. Hyperglycaemia in patients not known to have diabetes may be secondary to stress or to undiagnosed diabetes. It is often difficult to distinguish the cause of hyperglycaemia in a short hospital stay.

Stress hyperglycaemia most commonly occurs in patients with acute or critical illness and is more likely to occur in a more critically ill patient. Hyperglycaemia is postulated to be mediated through cytokines, the sympathetic nervous system and the hypothalamic‐pituitary‐adrenal axis155. It is not clear whether patients who manifest stress hyperglycaemia have an underlying impairment in their glucose metabolism, but in the long term, inpatient hyperglycaemia may herald undiagnosed diabetes or the development of diabetes in the future. The prevalence of undiagnosed diabetes varies in different inpatient settings and can be up to 60% (Appendix 10, table 10.1). It is important to diagnose patients with diabetes early to ensure appropriate management, both lifestyle and medication to prevent the development of long term complications.

There is limited literature to guide the level of hyperglycaemia predictive of diabetes or to suggest an appropriate algorithm for detection of diabetes in the acute hospital setting. The American Association of Clinical Endocrinologists / American Diabetes Association consensus recommendations defines a BSL >7.8mmol/L as inpatient hyperglycaemia and suggest an HbA1c may assist in diagnosis of diabetes. HbA1c >6.5% (48 mmol/mol) is strongly suggestive of underlying diabetes55,160. However, there is considerable heterogeneity amongst studies looking at predictors of diabetes in inpatient populations (Appendix 10, table 10.1). Different glucose values have been used to define hyperglycaemia. HbA1c levels used to define a diagnosis of diabetes and the populations studied have also been quite variable. Whilst HbA1c has not been ratified for the general diagnosis of diabetes in Australia, there is no doubt that for a patient with hyperglycaemia, it is a strong indicator of underlying diabetes.

Whilst in hospital, patients with newly diagnosed diabetes should be referred to the Specialist Diabetes Inpatient Team (section 12) or the Endocrine Team for management. Irrespective of whether diabetes is definitively diagnosed in hospital, patients with inpatient hyperglycaemia should receive follow‐up to ensure that the diagnosis is clarified, and appropriate counseling and management instituted. Notification of the general practitioner is vital to this process.

A suggested algorithm for the approach for the diagnosis and follow‐up of an inpatient with newly discovered hyperglycaemia is given in Appendix 11, Figure 11.1.

Recommendations and Practice Points

1. All inpatients with newly discovered hyperglycaemia (random plasma glucose >7.8mmol/L) should have an HbA1c performed.

2. All inpatients who are newly diagnosed with diabetes should be managed appropriately for diabetes. If there is diabetes expertise available, an early referral should be made.

3. All patients with abnormal glucose metabolism detected in hospital should have adequate follow up arranged, and the findings should be communicated to the usual care practitioner.

Section 11: What is the Role of a Specialist Inpatient Diabetes Team?

Improving glycaemic control has been shown to reduce the risk of adverse outcomes associated with hyperglycaemia, but the evidence for these clinical benefits have been obtained in and limited to specific individual clinical units. Translating these improved outcomes to a whole hospital is more challenging and requires a different approach. Rather than focusing on improved clinical outcomes, or on specific blood glucose targets, hospital‐wide approaches have focused on reducing the difference in length of stay for people with diabetes by improving overall diabetes management. The drivers for this approach are not so much an improvement in quality of care or clinical outcomes, but rather reductions in associated costs and improved bed utilisation. The factors contributing to increased length of stay and poorer outcomes associated with diabetes that are potentially modifiable include blood glucose control, inappropriate diabetes management and delayed involvement of specialist diabetes services.

Different approaches to this problem have been utilised, with varying levels of evidence to support the intervention. These vary from the traditional consultative service, to systematic hospital wide diabetes programmes, to the newer concept of the Specialist Diabetes Inpatient Management Team (Appendix 11, table 11.1). There has now been one randomised controlled trial164 and a number of comparative studies which have demonstrated improved outcomes with the latter approach (Appendix 11, Table 11.2).

These teams usually comprise dedicated Diabetes Inpatient Specialist Nurses (DISN), usually led by a consultant in diabetes. The role of such teams has included improving diabetes management expertise throughout the hospital, the development and implementation of diabetes management protocols, direct management of diabetes with specific referral criteria, ward liaison, troubleshooting, management advice, and discharge planning (Appendix 11, Table 11.3). DISNs are currently involved in 30‐50% of UK hospitals171, with Diabetes UK recommending a ratio of one Diabetes DISN for every 300 beds172. The NHS (UK) has adopted this approach to improve diabetes inpatient management through the whole health system, resulting in reductions in adverse outcomes and length of hospital stay9. In Australia, the introduction of Specialist Diabetes Inpatient Management Teams will require additional resources, but the long‐term economic argument is compelling. The literature suggests that hospitals which have introduced these teams have realised shorter lengths of stay and significant cost‐savings165,166,167,170. Health administrators need to invest in such teams which should result in better inpatient diabetes care, shorter lengths of hospital stay, and cost‐ savings to the health system. Forward planning is also needed for the training of the specialised workforce required for Diabetes Inpatient Management Teams.

Recommendation

1. Hospitals should consider the introduction of Specialist Diabetes Inpatient Management Teams

Section 12: What Routine Measures Should be Undertaken for People with Diabetes Admitted to Hospital?

Effective inpatient diabetes management should be provided early and continuously throughout the hospital admission. To support optimal glycaemic control in hospital and diabetes management after discharge, it is important to have established routine processes and protocols for the care of people with diabetes in hospital. These recommendations are generally based on good general hospital practice, experience, and common sense. General recommendations include: clear identification of diabetes in the medical record, blood glucose monitoring, a hypoglycaemia management protocol, HbA1c testing, a multidisciplinary team approach, dietetic assessment, diabetes self‐management education when appropriate, and discharge planning142. Insulin is a common source of medication error171,172, and must be minimised by mechanisms such as staff education, pharmacist oversight, and dedicated insulin prescription charts173.

Blood Glucose Monitoring

Where tight glycaemic control is desired, and particularly for patients on insulin, it is important for blood glucose monitoring to occur before and after meals. This is critical to facilitate appropriate adjustments to the patient’s insulin dosage, and monitor for hypoglycaemia. Additional testing at bed‐time and overnight is often also helpful. For stable patients, or those where tight glucose control is not an aim, testing can be reduced accordingly.

Discharge Planning and Diabetes Education

Whilst this document focuses on the management in hospital, it is important to take the opportunity to improve the post‐discharge management of diabetes as well. Liaison with the general practitioner is an important component of this. Not only might this improve patient outcomes, but it may reduce the need for readmission to hospital. The various team members participating in inpatient management also have a role in promoting and facilitating better diabetes care post‐discharge (Appendix 12, Table 12.1). Appropriate diabetes education is a critical component of inpatient patient care and discharge planning. A focus on the continuity of care where the patient is the central member in the management of diabetes is important, and their family members may need to be brought into the discussion.

Recommendations and Practice Points

1. Ensure clear processes and protocols are implemented in the hospital for routine diabetes care.

2. Ensure discharge planning which facilitates optimal long‐term diabetes management.

Appendices

Appendix 1: Search Methodology of Systematic Reviews

Table 1.1. PICO search questions and search terms used for each of the chapters.

Question Search Terms What glucose target should be aimed for in acute myocardial infarction?

hyperglyc(a)emia, diabetes, intensive glucose control, tight glucose control, intensive glyc(a)emic control, tight glyc(a)emic control ,myocardial infarction, acute coronary syndrome, with the outcomes of mortality or death.

What glucose target should be aimed for in acute stroke?

hyperglyc(a)emia, diabetes, intensive glucose control, tight glucose control, intensive glyc(a)emic control, tight glyc(a)emic control ,myocardial infarction, stroke, cerebrovascular accident, with the outcomes of mortality or death.

What are appropriate glucose targets for patients in general hospital wards?

intensive glucose control, tight glucose control, intensive glyc(a)emic control, tight glyc(a)emic control, hospital, surgery, medicine

What special measures need to be undertaken for people on enteral + parenteral nutrition?

Diabetes and (enteral nutrition or parenteral nutrition)

How is steroid‐induced hyperglycaemia best managed?

(Metformin or sulphonylurea or incretins or Dipeptidyl‐Peptidase IV Inhibitors or thiazolidinediones or insulin) and (glucocorticoids or prednisone) and (hyperglycaemia or diabetes)

What is the optimal means of achieving routine glucose control in hospital?

hyperglyc(a)emia, diabetes, intensive glucose control, blood glucose/sugar control, intensive glyc(a)emic control, tight glyc(a)emic control, hospital, inpatient

How should patients on insulin pump therapy be managed in hospital?

diabetes, guidelines, hyperglyc(a)emia, hypoglyc(a)emia, hospital admission, acute care, inpatient care, perioperative management, CSII, insulin pump, insulin pump therapy, IPT

What is appropriate glucose control in end of life situations

diabetes, guidelines, hyperglyc(a)emia, hypoglyc(a)emia, hospital admission, inpatient care, end‐of‐ life, palliative care, terminal illness, advanced cancer, hospice, insulin, oral hypoglyc(a)emic agents, sliding scale, blood glucose, therapy, and management

How should patients with newly discovered hyperglycaemia be followed up?

hyperglyc(a)emia, diabetes, intensive glucose control, blood glucose/sugar control, intensive glyc(a)emic control, tight glyc(a)emic control, hospital, inpatient

What is the role of a specialist diabetes inpatient team?

Diabetes, hospital, inpatient

What routine measures should be undertaken for people with diabetes admitted to hospital?

Consensus only

Appendix 2: Literature reviewed for “What Glucose Target Should be Aimed for in Acute Myocardial Infarction?”

Table 2.1. Recent studies examining the relationship between admission glucose levels and mortality following myocardial infarction

Study Subjects Characteristics Elevated admission glucose predictive of mortality?

Threshold level for effect?

Comments Methodology

Wong 200413 158 STEMI Yes 8 mmol/L Similar relationship for both inpatient and 6 month mortality Relationship between BG and death present with and without reperfusion therapy

Clinical cohort study

Stranders 200414

846 Any AMI Yes 11.1 mmol/L for non‐diabetics

Above 11.1 mmol/L, non‐diabetics had same risk as those with diabetes.

Retrospective clinical cohort study

Timmer 200415

356 STEMI with PTCA or reperfusion

Yes 7.8 mmol/L Also association with larger infarct size and reduced LV function

Post‐hoc subanalysis of a clinical trial cohort

Kosiborod 200516

141680 Age >65 Yes 6.1 mmol/L for non‐diabetes 13.3 mmol/L for diabetes

Similar results for 30 day and one year mortality

Analysis of database

Straumann 200517

978 All had PTCA Yes 7.8 mmol/L Similar results for 30 day and longer‐term mortality

Analysis of database

Meier 200518 227 All AMI Yes 7.4 mmol/L for non‐diabetics, 7.9 mmol/L for diabetes

Survival >3.5 years was assessed Clinical cohort study

Goyal 200619 1469 Subanalysis of CARDINAL Trial

Yes (only for non‐ diabetics)

Lower mortality amongst non‐diabetics where there was a greater drop in BG over 24 hrs

Post‐hoc subanalysis of a clinical trial cohort

Bhadriraju 200620

9020 Subanalysis of OPUS‐TIMI trial

Yes 5.6 mmol/L Relationship stronger for non‐diabetics. Results also validated in subanalysis of TACTICS‐TIMI trial

Post‐hoc subanalysis of a clinical trial cohort

Naber 200921 5866 Non‐diabetic STEMI (ACOS Registry)

Yes 8.3 mmol/L Inpatient and 1 year mortality Cohort study

Sinnaeve 13526 Global registry Yes 6.9 mmol/L Random BGL associated with inpatient Analysis of

200922 (random) 5.6 mmol/L (fasting)

mortality only, fasting BGL associated with both inpatient and 6 month mortality.

database

Ishihara 200923

3750 Within 48 hrs of AMI

Yes 7 mmol/L U‐shaped curve for patients with diabetes, increased mortality if BGL<7 or >11 mmol/L

Cohort study

Dziewiercz 200924

763 Non‐STEMI treated conservatively

Yes 5 mmol/L Relationship stronger for non‐diabetic subjects

Analysis of database

Goyal 200925 30536 CREATE ECLA and OASIS‐6 cohorts

Yes 7.8 mmol/L Hypoglycaemia < 3.3 also predicted mortality

Post‐hoc analysis of clinical trial cohorts

De Mulder 201026

1185 Both preinvasive and PTCA eras

Yes 11 mmol/L Each mmol/L increase corresponded to a 7% increased mortality (adjusted HR 1.07, 95% CI 1.04‐1.10)

Cohort study

Timmer 201127

4176 Non‐diabetic STEMI

Yes 8.2 mmol/L 30 day and 1 year mortality assessed. U‐ shaped curve for mortality with increased mortality for those with BGL ≤6.9 mmol/L

Cohort study

Table 2.2. Observational data of a relationship between average glucose levels or glucose levels achieved in the first 24 hours after myocardial infarction and mortality.

Study Subjects Characteristics Glucose parameter

Elevated glucose predictive of mortality?

Threshold level for effect?

Comments Methodology

Cheung 200628, 200829

240 Myocardial infarct with known diabetes or admission BG ≥7.8 mmol/L

1‐2 hourly capillary BGs

Yes 8 mmol/L Post‐hoc subanalysis of a clinical trial cohort

Kosiborod 200830

7820 All AMI Mean glucose measurements

Yes 6.1 mmol/L Mortality lower in insulin treated patients

Analysis of database

Table 2.3. Systematic reviews of randomized controlled trials of tight glucose control in myocardial infarction, where the primary outcome was death.

Review Search Method Selection Question Studies Subjects Result/Conclusion Comment Pittas 200432

Medline, Cochrane Controlled Clinical Trials Register

Studies of Insulin in critically ill hospitalized adult patients. Subanalysis: those aiming for glucose control

AMI subanalysis: 8 studies. Davies 199137, Malmberg 199531, Scott 199938, Lazar 200039, Szabo 200140, van den Berghe 200141, Groban 200242, Smith 200243

Subanalysis: 2772

Subanalysis: 29% reduction in mortality (RR 0.71, 95%CI 0.54‐0.93)

Subanalysis included studies of coronary surgery and ICU patients

Zhao 201033

Medline, CENTRAL, EMBASE

RCTs of GIK or insulin‐ glucose. Subanalysis: insulin‐ glucose only

AMI subanalyis: 3 studies. Malmberg 199531, Malmberg 200544, Cheung 200628

Subanalysis: 2113

Subanalysis: No reduction in mortality (RR 1.07, 95%CI 0.85‐1.36)

Devine 201034

Medline and Embase

Randomised trials of ACS with hyperglycaemia comparing insulin infusion or GIK with active controls which assessed mortality and morbidity

Malmberg 199531, Diaz 199845, Ceremuzynski 199946, Malmberg 200544, CREATE‐ECLA 200547, Cheung 200628

Current evidence that insulin therapy reduces mortality and morbidity in ACS is inconclusive

Meta‐analysis not done

Lipton 201135

Pubmed Trials with insulin in patients with unstable angina or AMI. Subanalysis: those aiming for glucose normalization

AMI subanalysis: 3 studies. Malmberg 199531, Malmberg 200544, Cheung 200628

Intensive glucose lowering insulin therapy can reduce mortality

Meta‐analysis not done

Kansagara 201136

MEDLINE, Cochrane Database of Systematic Reviews, ClinicalTrials.gov

RCTs using insulin to achieve strict glycaemic control. Subanalysis: AMI.

AMI subanalysis: 6 studies Malmberg 199531, Van der Horst 200348, Malmberg 200544, Cheung 200628, Rasoul 200749, Oksanen 200750,

Subanalysis: 4007

Non‐ICU: 9 studies, no reduction in short‐term mortality (RR 1.0, 95%CI 0.94‐ 1.07) AMI: No mortality reduction

6‐fold risk of hypos (BGL<2.2 mmol/L) in all settings RR 6.0, 95%CI 4.06‐8.87, p<0.001). Oksanen50=study of subjects following VF

Bold=studies with specified objective of intensive glucose control in AMI

Table 2.4. Randomised controlled trials of myocardial infarction with a specific glucose target.

Trial Subjects Entry Criteria Insulin Regimen Glucose Target

Primary Outcome

Secondary Findings

Comments

DIGAMI, Malmberg 199531, 199751

620 Myocardial infarct and admission BG >11.0 mmol/L.

Variable rate glucose‐insulin solution for at least 24 hrs.

7‐10 mmol/L

Reduced one year mortality in insulin infusion group (18.6% vs 26.1%, p=0.027).

Greatest benefit to patients with low premorbid cardiovascular risk profile.

Amongst first 327 subjects, blood glucose at 24 hours lower in insulin than in control subjects (9.2±2.9 vs 12±4.4 mmol/L). Insulin group received regular subcutaneous insulin after discharge, which may have contributed to better outcomes.

GIPS, van der Horst 200348

940 Within 24 hrs of ST‐ segment elevation infarct (all had PTCA).

20% glucose‐ potassium solution at 3mls/kg/hr with insulin at variable rate.

7‐11 mmol/L.

No significant reduction in 30 day mortality (4.8% vs 5.8%).

Median BG at 16 hours 7.7 mmol/L for GIK group and 8.1 mmol/L for controls (NS).

DIGAMI‐2, Malmberg 200544

1253 Myocardial infarct and either known type 2 diabetes or admission BG >11.0 mmol/L.

Variable rate glucose‐insulin solution for at least 24 hrs.

7‐10 mmol/L

No reduction in mortality with insulin infusion.

BG at 24 hours in insulin treated groups only 0.9 mmol/L lower than for conventional treatment group (9.1±3.0 and 9.1±2.8 vs 10±3.6 mmol/L, p=0.0001)

HI‐5, Cheung 200628

240 Myocardial infarct with known diabetes or admission BG ≥7.8 mmol/L.

Variable rate insulin with 5% dextrose 80 mls/hr.

4‐10 mmol/L

No reduction in mortality with insulin infusion.

Mortality higher in subjects with mean 24 hour blood glucose level >8.0 mmol/L.

Mean 24 hour BG in insulin treated group only 0.7 mmol/L lower than for conventional treatment group (8.3±2.2 vs 9.0±2.8 mmol/L, NS)

Other trials of insulin glucose therapy where there were no glucose targets were excluded from consideration: ECLA (1998)45, POL‐GIK (1999)46, CREATE‐ECLA (2005)47, GIPS II (2006)52

Table 2.5. Guidelines regarding glucose control in myocardial infarction

Guideline Population Recommendation ESC and EASD guidelines on diabetes, prediabetes and cardiovascular disease,

People with diabetes and AMI

There is reasonable evidence to initiate glucose control by means of insulin infusion in diabetic patients who are admitted for AMIs with significantly elevated

200754 blood glucose levels in order to reach normoglycaemia as soon as possible (Class IIa, Level B)

AACE/ADA consensus statement on inpatient glycemic control, 200955

All hospitalized patients

Insulin infusion should be used to control hyperglycaemia in the majority of critically ill patients in the ICU setting, with a starting threshold of no higher than 10 mmoml/L. Target 7.8 – 10 mmol/L, and greater benefit may be realized at the lower end of this range. For majority of non‐critically ill patients, premeal <7.8 mmol/L, random < 10 mmol/L. as long as this can be safely achieved.

ACC/AHA guidelines for the management of patients with ST elevation myocardial infarction, 200956

ST elevation AMI Reasonable to use insulin based regimen to achieve and maintain BG <10 mmol/L whilst avoiding hypoglycaemia

ACP guidelines for intensive insulin therapy for the management of glycemic control in hospitalized patients, 201157

All hospitalized patients

Do not use intensive insulin therapy to strictly control glucose in non‐ICU patients with or without diabetes mellitus

Appendix 3: Literature reviewed for “What Glucose Target Should be Aimed for in Acute Stroke?”

Table 3.1. Recent studies examining the relationship between admission glucose level and stroke outcomes.

Study Subjects Characteristics Relation between admission glucose and mortality?

Relation between admission glucose and other outcome?

Threshold level for effect?

Comment

Baird 200359 25 Ischaemic stroke N/A Yes, infarct size, NIHSS and mRS

7 mmol/L

Allport 200460 31 Acute ischaemic stroke

N/A Yes, insular cortical ischaemia

N/A

Alvarez‐Sabin 200461

138 MCA territory treated with tPA

N/A Yes with NIHSS and mRS N/A Detrimental association greatest with early reperfusion therapy.

Farrokhnia 200562

447 Acute stroke Yes (only for non‐ diabetic)

N/A Diabetes: 10.3 mmol/L, non‐ diabetics: 6.3 mmol/L

Threshold determined from ROC

Stollberger 200563

992 All acute stroke Yes N/A 9.2 mmol/L for non‐diabetics

Gentile 200664 960 Ischaemic stroke Yes N/A 7.2 mmol/L

Yong 200865 748 Received tPA for acute hemispheric stroke

Yes Yes, with BI, mRS, 7 day neurological improvement

7.8 mmol/L Effect not seen among subjects with known diabetes

Fuentes 200966 476 Acute ischaemic stroke

Yes Yes with mRS 8.6 mmol/L Threshold determined from ROC

Stead 200967 447 Acute ischaemic stroke

Yes Yes with stroke severity and functional impairment

7.2 mmol/L Relationship stronger for patients without diabetes

Poppe 200968 1098 Acute ischaemic stroke treated with t‐PA

Yes Yes, with symptomatic intracerebral haemorrhage and mRS

N/A

Ntaios 201069 1446 Ischaemic stroke N/A Yes, with NIHSS and mRS

<3.7 and >7.2 mmol/L

Ahmed 201070 16049 Ischaemic stroke treated with thrombolysis

Yes Yes, with NIHSS and mRS

6.7 mmol/L similar threshold for diabetes and non‐diabetics

Dziedzic 201071 302 Ischaemic stroke Yes N/A N/A

Saposnik 201172 8223 Acute ischaemic stroke in registry

Yes N/A 7.5 mmol/L

Kimura 201173 97 Received tPA within 3 hours of stroke onset

N/A Infarct volume larger and worse mRS

7.2 mmol/L Relationship not present amongst those with successful early recanalisation.

Hu 201274 774 Acute stroke Not reported Yes, with NIHSS, BI and mRS

Diabetes: 8.9 mmol/L, non‐ diabetics: 6.8 mmol/L

NIHSS = National Institutes of Health Stroke Scale, BI = Barthel Index, mRS = modified Rankin Score

Table 3.2. Studies examining the relationship between mean glucose levels and stroke outcomes:

Study Subjects Characteristics Glucose parameter

Elevated glucose predictive of mortality?

Relation between admission glucose and other outcome?

Threshold level for effect?

Baird

200359

25 Ischaemic stroke Mean capillary and mean CGMS

N/A Yes, infarct size, NIHSS and mRS

7 mmol/L

Farrokhnia,

200562

447 Acute stroke Mean capillary Yes N/A 10.3 mmol/L for diabetes, 6.3 mmol/L for non‐diabetics (based on ROC)

Yong

200865

748 Received tPA for acute hemispheric stroke

Glucose at 24 hrs Yes Yes, BI. mRS, 7 day neurological recovery.

7.8 mmol/L

Fuentes

200966

476 Ischaemic stroke Maximum capillary glucose

Yes Yes, mRS 8.6 mmol/L

NIHSS = National Institutes of Health Stroke Scale, BI = Barthel Index, mRS = modified Rankin Score

Table 3.3. Systematic reviews of randomized controlled trials of tight glucose control in stroke, where the primary outcome was death or a measure of disability.

Review Search Method Selection Question Studies Subjects Result/Conclusion Comment

Kansagara

201136

MEDLINE, Cochrane Database of Systematic Reviews, ClinicalTrials.gov

RCTs using insulin to achieve strict glycaemic control. Subanalysis: stroke and acute brain injury

Walters 200677, Gray

200778, Azevedo

200779, Bruno 200880,

Yang 200981

Non‐ICU setting: 9 studies, no reduction in short‐term mortality (RR 1.0, 95%CI 0.94‐ 1.07) Stroke and acute brain injury: No reduction in mortality

Increased risk of hypoglycaemia in all settings.

Kruyt

201075

Not stated Studies investigating the feasibility and efficacy of tight glycaemic control in patients with ischaemic stroke

Walters 200677, Gray

200778, Bruno 200880

BG >10 mmol/L should trigger insulin administration.

Review not restricted to RCTs

Bellolio

201176

Cochrane Stroke Group Trials Register, CENTRAL, MEDLINE, EMBASE, CINAHL, Science Citation Index, Web of Science, Scopus

RCTs comparing intensively monitored insulin therapy versus usual care in adult patients with acute ischaemic stroke.

Vinychuk 200582,

Walters 200677, Gray

200778, Staszewski

2007*, Bruno 200880,

Kreisel 200983,

Johnston 200984

1296 No difference in death or disability and dependence (OR 1.00, 95% CI 0.78 to 1.28) or final neurological deficit (SMD ‐ 0.12, 95% CI ‐0.23 to 0.00).

Increased hypoglycaemia with intervention. Some studies in this review were not designed to assess neuro outcomes.

Bold=studies of stroke only, *Unpublished

Table 3.4. Randomised controlled trials of stroke with a specific glucose target.

Trial Subjects Entry Criteria Insulin Regimen

Glucose Target

Glucose Achieved

Primary Outcome

Secondary Findings

Comments Hypos

Vinychuk 200582

128 Within 24 hours of ischaemic stroke onset, admission BG 7‐16 mmol/L

Glucose potassiu m insulin infusion

<7 mmol/L

Diabetes: 7 vs 11.2 mmol/L at 12‐24 hrs No diabetes: 5.8 vs 8.1 mmol/L

Improvemen t in NHISS compared to

4 arms to study Not reported

Walters 200677

25 Within 24 hours of ischaemic stroke onset, admission BG 8‐20 mmol/L

Variable rate insulin infusion

5‐8 mmol/L

In target 87% of the time vs 71% of the time (p<0.001)

‐ 1 death insulin group, 0 in control group

A pilot study 1 in insulin group, 0 in controls

GIST‐UK, Gray 200778

933 Within 24 hours of stroke onset, admission BG 6‐17 mmol/L

Glucose potassiu m insulin infusion

4‐7 mmol/L

GKI group 0.57 mmol/L lower (p<0.001)

No reduction in death at 90 days with GKI

No reduction in residual disability or functional recovery

Trial discontinued early due to slow recruitment. BG dropped spontaneously in control group.

41.2% GKI patients had BG <4 mmol/L and 15.7% required rescue iv glucose

Bruno 200880

46 Within 12 hours of cerebral infarct and BG ≥8.3 mmol/L

Variable rate insulin infusion for 72 hrs

<7.2 mmol/L

7.4 vs 10.5 mmol/L

Diff in glucose achieved

2 death insulin group, 0 in control group

A pilot study 35% vs 13% had hypos <3.3

Kreisel 200983

40 Within 24 hours of ischaemic stroke onset

Variable rate insulin infusion

4.4‐6.1 mmol/L

6.5 vs 8.0 mmol/L, p<0.0005

Diff in glucose achieved

No difference in death or functional outcome

Not powered to detect difference in death and disability

25 vs 2 hypo events (p<0.05)

Johnston 200984

74 Within 12 hours of cerebral infarct and BG ≥6.1 mmol/L

Variable rate insulin infusion

3.9‐6.1 mmol/L

6.2 vs 8.4 mmol/L

Diff in glucose achieved

No difference in death or functional outcome

3 arms to study Not powered to detect difference in death and disability

30% vs 4% had at least one hypo

Middleton 201185

1126 Within 48 hours of acute stroke

Variable rate insulin infusion if BG >=11 mmol/L if diabetes, >=16 for non‐ diabetics, up to 72 hours

4‐8 mmol/L or local guidelin es once insulin infusion comme nced

7.0 vs 6.8 mmol/L, p=0.02 (Note small difference only)

Difference in mortality or dependency 42% vs 58% (p=0.002)

No difference in Barthel index, but higher SF‐36 Physical Health Score in intervention

Intervention was package of glucose, fever, and swallowing management. Not possible to determine contribution of glucose control.

Not reported.

Table 3.4. Guidelines regarding glucose control in stroke

Guideline Population Recommendation

AHA/ASA guidelines for the early management of adults with ischemic stroke, 200786

Ischaemic stroke Serum glucose concentrations (possibly >7.8 to 10.3 mmol/L) probably should trigger administration of insulin (Class IIa, Level of Evidence C).

European Stroke Guidelines, 200887 Ischaemic stroke Treatment of serum glucose levels > 10 mol/L with insulin titration is recommended (Class IV evidence). Severe hypoglycaemia [<2.8 mmol/L should be treated with intravenous dextrose or infusion of 10–20% glucose (Class IV evidence)

Appendix 4: Literature Reviewed for Question “What are Appropriate Glucose Targets for Patients in General Hospital Wards?”

Table 4.1. Observational studies examining the relationship between hyperglycaemia and outcomes in hospital outside of the situations of intensive care, myocardial infarction and stroke.

Study Subjects Finding Comment

Pomposelli 1999 Study88

97 patients undergoing general surgery

Single BGL >12.2 mmol/L predictive of nosocomial infection

Golden 199989

411 CABG patients Mean capillary BGL ≥11.5 mmol/L associated with increased infection

Umpierrez, 200290

2030 admissions to a community teaching hospital

2x Fasting BGL ≥7.0 mmol/L or rBGL ≥11.1 mmol/L associated with increased mortality, need for ICU and longer LOS

Risk higher for new hyperglycaemia than diabetes

Cheung 200591

122 subjects on TPN Mean BGL ≥7.9 mmol/L associated with increased infection Mean BGL ≥9.1 mmol/L associated with increased mortality

Baker 200892 903 patients in general medical ward

Admission BGL ≥5.6 mmol/L associated with increased mortality

No association between BG levels and mortality amongst people with known diabetes. HbA1c also predictive of mortality in non‐diabetics.

Cheung, 200893

6187 admissions to a teaching hospital

Admission BGL >8.0 mmol/L predictive of mortality and longer LOS

Risk higher for new hyperglycaemia than diabetes

Table 4.2. Systematic reviews of studies examining trials of tight glycaemic control outside of the intensive care setting, and not specifically focusing on myocardial infarction or stroke

Review Selection Question Studies Total subjects

Result/Conclusion Comment

Haga 201194

Effects of “tight” versus “normal” glycaemic control, peri and post‐operatively, in patients undergoing cardiac surgery.

3 studies: Groban 200242, Lazar 200496, Ingels 200697

686 48% reduction in mortality (OR 0.52, 95%CI ‐0.3 ‐ ‐0.91, p=0.02). May be some benefit to tight glycaemic control during and after cardiac surgery.

Kansagara 201136

RCTs using insulin to achieve strict glycaemic control. Subanalysis: Non‐ICU studies Subanalysis: Perioperative

9 non‐ICU studies: Malmberg 199531, Van der Horst 200341, Butterworth 2005, Li 2006, Cheung 200628, Oksanen 200750, Azevedo 200779, Yang 200981, 5 perioperative studies: Smith 200243, Lazar 200439, Butterworth 200598, Li 200699, Barcellos 2007100

Non‐ICU: 2677

Non‐ICU setting: no reduction in short‐term mortality (RR 1.0, 95%CI 0.94‐1.07). Perioperative: no reduction in short‐term mortality.

No trials in general medical wards Perioperative studies mostly poor quality RR hypoglycaemia 6.0, 95%CI 4.06‐ 8.87, p<0.001).

Murad 201295

Observational or randomized studies that compared the effect of intensive glycaemic control to a control group seeking less aggressive normalization of glycaemic levels. Intensive care setting excluded.

19 studies: RCTs – Malmberg 199531, Dickerson 2003101, van der Horst 200348, Malmberg 200531, Cheung 200628, Walters 200677, Umpierrez 2007102, Umpierrez 2009103, Umpierrez 2011104

Varied for different analyses

No association between intensive glucose control and risk of death, myocardial infarction or stroke. Association with reduced risk of infection (RR 0.41, 95%CI 0.21‐ 0.77). Trend to increased hypoglycaemia.

Inclusion of observational studies is questionable.

Table 4.3. Existing guidelines for glucose targets in non‐critically ill patients in hospital.

Guideline Population Recommendation

AACE/ADA, 200955 All hospitalized patients

Insulin infusion should be used to control hyperglycaemia in the majority of critically ill patients in the ICU setting, with a starting threshold of no higher than 10 mmol/L. Target 7.8 – 10 mmol/L, and greater benefit may be realized at the lower end of this range.

For majority of non‐critically ill patients, premeal <7.8 mmol/L, random < 10 mmol/L, as long as this can be safely achieved.

Society for Ambulatory Anesthesia, 2010105

Pre and intra‐ operative

Insufficient data to recommend the level of preoperative fasting blood glucose above which elective ambulatory surgery should be postponed. No evidence that any particular blood glucose level is either beneficial or harmful for patients undergoing ambulatory surgical procedures. Suggest that in patients with well controlled diabetes, intraoperative blood glucose levels be maintained <10 mmol/L.

ACP, 2011106 All hospitalized patients

Do not use intensive insulin therapy to strictly control glucose in non‐ICU patients with or without diabetes mellitus.

Endocrine Society, 2012107 Recommend premeal target <7.8 mmol/L, random target <10 mmol/L, but modify according to clinical status. Reassess therapy if BG values fall below 5.6 mmol/L. Modify therapy when BG values are <3.9 mmol/L.

Appendix 5: Literature Reviewed for Question “What is the best method to maintain glycaemic control in a hospitalized patient who is receiving parenteral or enteral nutrition?”

Table 5.1. Clinical trials of glucose control among patients receiving enteral or parenteral feeding.

Paper Design Quality Level of evidence

Statistical precision Size and direction of effect

Relevance

Koryt‐ kowski 2009108

Sliding scale insulin (4‐6 hourly if BGL >7.1 mmol/l) vs sliding scale insulin and glargine. Target glucose range (5.6‐10 mmol/l). Non‐critically ill hospitalized patients with ≥2 BGLs over 7.2 mmol/l (with or without prior diagnosis of diabetes) Receiving enteral nutrition therapy formula and delivery at discretion of nutrition team with majority receiving ≥50% carbohydrate.

Selection bias, randomization process not described. Information bias, open label study. Financed by Sanofi Aventis; clearly disclosed and stated that sponsor did not influence study design conduct. Inclusion/exclusion criteria described. Baseline characteristics similar between groups. Large percentage (55%) of patients in SSI arm also received NPH insulin.

II No difference between the treatment groups in mean study glucose levels p=NS.

No effect. Both regimens effective in lowering mean glucose levels. Similar rates of hypoglycaemia on BG measures, total adverse events and LOS. Similar total daily insulin doses.

Inpatient population receiving enteral nutrition therapy.

Grainger 2007109

Variable dose preprandial insulin (standard of care) vs nurse led insulin protocol (variable dose lispro, regular + correctional dose) and fixed dose glargine (weight dependent). Critically ill hospitalised patients with known type 2 diabetes or FBG> 11.1. Receiving bolus enteral nutrition (TwoCal HN or Nepro renal failure) q4 hours to provide 6 feeds/day.

Selection bias, high. Information bias, high. Inclusion/exclusion criteria described. Used historical controls from before protocol implementation. Excluded NIRDM patients. Baseline characteristics, few reported but similar.

III‐3 Lower mean glucose in intervention cf control group p<0.001. Greater proportion in intervention group achieved goal glucose range p<0.01, + shorter time to achieve glucose control 21 vs 60 hrs, p not reported. Increased hypoglycaemia p=0.02.

Insulin protocol increased proportion of patients achieving target blood glucose range. Modest increase in hypoglycaemia.

ICU inpatient population receiving enteral nutrition therapy.

Appendix 6: Literature reviewed for the Question “How is steroid‐induced hyperglycaemia best managed?”

Table 6.1. Incidence of steroid‐induced hyperglycaemia.

Author Study Design Steroid Incidence of steroid‐induced hyperglycaemia

Risk factors for steroid‐induced hyperglycaemia

Comment

Fajans 1954115 steroid given prior to a GTT

CRH or cortisone N/A First‐degree relative with diabetes

Gurwitz 1994114 Case control study

Any oral glucocorticoids N/A Steroid dose (odds ratio rose from 1.77 for 39 mg/day hydrocortisone equivalent, rising to 10.34 for 120mg/day or more

Hyperglycaemia based on prescription of oral antidiabetic agent

Feldman‐Billard 2005113

Retrospective audit

Pulse methylprednisolone

64% had at least one BG ≥14 mmol/L

Higher HbA1c

Donihi 2006110 Retrospective audit

At least prednisone 40 mg/day or equivalent for 2 days

Overall 64% had at least one BG ≥11.1mmol/L. 56% amongst non‐diabetics.

Pre‐existing diabetes 21% patients not screened

Fong 2011111 Prospective audit

At least prednisone 25 mg/day for 2 days

71% had at least one BG ≥10mmol/L

Age

Burt, 2011112 Prospective study

At least prednisone 20 mg/day for 2 days

53% had at least one BG ≥10mmol/L

Pre‐existing diabetes Used CGMS

Table 6.2. Evidence regarding the effectiveness of anti‐diabetic medication in the management of steroid‐induced hyperglycaemia.

Medication Class

Intervention Studies Other Literature and Comments

Metformin No trials identified Limitations to its use in the inpatient setting may include the presence of risk factors for lactic acidosis, or variable oral intake

Sulphonyl‐ ureas (SU)

No prospective trials. Studies in ambulatory setting only. A retrospective review of 40 ambulatory pts with diabetes at baseline who had received prednisone (dose range 5‐40mg/day for management of non‐endocrine conditions such as COPD and SLE) described 5 patients who had received repaglinide. However insulin was also required to achieve glycaemic control in 4 of the 5121. Kasayama reported 3 adult ambulatory patients with immunological conditions, initially requiring prednisone 20‐ 40mg/day, and maintained on 5‐10mg/day, who developed hyperglycaemia after 1‐2 yrs of maintenance GC therapy122. All 3 were able to be controlled with glimepiride monotherapy (1‐ 3mg), achieving Hba1c <7%.

Although predominant effect of GCs is a reduction in insulin sensitivity, individuals who develop hyperglycaemia with GC administration, have a reduced insulin secretory capacity123. Hence insulin secretogogues are not an ideal choice. Risk of hypoglycaemia as GC tapered if long acting agents are used124, where there is meal omission125 or if renal function is reduced126. SU use has been suggested to be limited to those with milder degrees of GC induced hyperglycaemia, where fasting BG <7mmol/L126. Since inpatients usually receive high GC doses and may have variable oral intake or renal function, SUs have a limited role.

Thiazolidine‐ diones

Published reports of use for steroid‐ induced hyperglycaemia in setting of organ transplantation only. Pioglitazone improved glycaemic control as an adjunct to insulin (mean HbA1c falling by 1.28%) in a series of renal transplant recipients managed with prednisone (4‐20mg/day) and sirolimus or mycophenolate, 6 of the 10 patients had diabetes predating their transplant127. Efficacy of other thiazolidinediones for post transplant diabetes or GC related DM has also been described128‐130.

Delayed onset of action makes this group generally unsuitable for acute inpatient management of hyperglycaemia.

Incretins No completed trials identified. A trial of vildagliptin in the management of post transplant diabetes is in progress131.

Potential role has been explored by van Raalte132.

Insulin No trials of insulin regimens in management of steroid‐induced hyperglycaemia in hospital other than intravenous insulin133. Observational report suggests that use of morning isophane insulin versus long acting insulin (either insulin with rapid acting insulin at mealtime) achieved glycaemic control more quickly134. Preliminary report suggests that early use of basal insulin in the

Where marked fasting hyperglycaemia (>10mmol/L) is present, insulin is thought to provide better management than oral agents126,137. Insulin provides greater dose flexibility, more rapid onset of action and titration and there is usually no dose ceiling. Insulin dose requirements need to be individualised, due to variations in insulin sensitivity, insulin secretory capacity, GC regimen and dosing, oral intake, renal function

setting of renal transplantation, prevents the development of subsequent new onset diabetes135. In response to prednisone 60mg/day a series of 10 patients with T1DM managed with subcutaneous insulin pump required an increase of between 30‐100%136.

and prior control. Pre‐emptive increases and decreases in insulin required as GC dose is adjusted. This usually requires daily adjustment. Generally the insulin regimen or adjustments to a pre‐existing regimen should predominantly target post‐prandial control, and with morning GC administration, the afternoon hyperglycaemia. Clore has advocated the use of isophane insulin for management of steroid hyperglycaemia124, the initial dose determined according to GC dose and patient weight (e.g. prednisone 10mg daily requires 0.1unit/kg/day as isophane insulin; prednisone 40mg daily requires 0.4 unit/kg/day), and titrated according to response. For a patient already on insulin, this may be added to the existing regimen. This titration schedule is easier to continue upon discharge than more complex regimens, and more likely to be successful if patients have a consistent routine and carbohydrate consumption from day to day138. Isophane insulin can be supplemented with ultraquick insulin analogue with meals139. Basal plus prandial insulin is likely to be required in patients receiving high dose GC (eg >50mg prednisone/day) where prior glycaemic control was poor, GCs had been initiated without pre‐emptive consideration of glycaemia or GC are given as split dose. With multiple daily dose GC regimens, isophane insulin twice daily with prandial rapid acting analogue can be initiated. A regime that controlled glycaemia on previous occasions can be re‐initiated when cyclical GCs are required, as long as there has been no major interval change in weight or renal function.

Appendix 7: Literature Reviewed for Question “What is the best method to maintain glycaemic control in a hospitalized patient who is not critically ill?”

Table 7.1. Randomised controlled trials comparing insulin regimes for glucose control in hospital.

Author Study design Quality and risk of Bias Level of evidence

Statistical precision and significance

Size and direction of effect Relevance

Dickerson 2003101

153 subjects. Randomised trial comparing glycaemic control in medical patients with type 2 diabetes receiving sliding scale insulin versus routine diabetes medications.

Medium–high risk of bias. Bias that population chosen from convenience sample. Randomisation procedure clearly described. Open label. Clear exclusion criteria. Baseline characteristics mostly similar. May have been treatment changes during study.

III‐1 Statistical calculations included in study design. Hyperglycaemic and hypoglycaemic events.

Primary outcome was frequency of hyperglycaemia (>16.7) and hypoglycaemia (<2.8). No difference between the two groups. No difference in secondary outcome which was LOS.

General medical patients, similar to Australian general medical wards

Yeldandi 2006143

94 patients. Randomised trial comparing once‐ daily glargine insulin with twice daily NPH/regular insulin for control of hyperglycaemia in inpatients post‐ cardiovascular surgery.

Medium risk of bias. Randomisation procedure not described. Open label. No exclusion criteria. Baseline characteristics similar.

III‐1 No calculations were done to show power. Mean BG 124mg/dL v 131mg/dL p=0.065 Hypoglycaemia less with glargine p=0.036.

Minimal effect. Calculated % of BGs within target range. Similar BGs achieved with both regimens in whole cohort. Subgroup with diabetes: bd NPH had lower mean BG than glargine. Subgroup without diabetes: no difference in mean BGs. % of BGs within target range similar between the groups. Hypoglycaemia less common with glargine p=0.036.

Restricted to post operative cardiac surgery. Patients could have diagnosed diabetes or hyperglycaemia. Caloric intake very diminished therefore patients not requiring a prandial insulin component.

Umpierrez 2007 (RABBIT 2 Trial)102

130 patients. Randomised trial of basal‐bolus insulin therapy compared to sliding scale insulin in the inpatient

Medium risk of bias. Not clearly defined what the randomization process was. Open label. Inclusion and exclusion criteria detailed. Baseline

III‐1 Unclear if study appropriately powered. FBG p<0.001, mean random blood glucose

Large effect. Significant difference in favour of BBI in regards glycaemic control, p<0.001, lower mean daily glucose, mean random BG and FPG. No difference in

Tertiary hospital. Similar patient population and treating teams. Difference is ceasing all oral

management of general medical patients with type 2 diabetes.

characteristics same. Unrestricted educational event from Sanofi Aventis

p<0.001, mean glucose throughout hospital stay p<0.001.

hypoglycaemia. No difference in secondary endpoints (LOS and mortality). Significant difference in the units of insulin used.

hypoglycaemic agents on admission. Sliding scale not appropriate comparator.

Umpierrez 2009103

130 subjects. Randomised trial comparing regimens with detemir and aspart versus neutral protamine hagedorn and regular in medical patients with type 2 diabetes.

Medium risk of bias. Selection bias, randomisation process unclear. Information bias, open label study. Inclusion and exclusion criteria detailed. Baseline characteristics the same. Financed by Novo‐ Nordisk (disclosed and had no influence on study design).

III No difference between the treatment groups, p=NS

No effect. Both regimens equally as effective in lowering mean BGs and lead to similar rates of hypoglycaemia. BG <140mg/dL was achieved in 45% of BBI group and 48% in split mixed group.

Tertiary hospital. Similar patient population and treating teams. Difference is ceasing all oral hypoglycaemic agents on admission.

Umpierrez 2011 (RABBIT 2 Surgery)104

211 subjects. Randomised trial of basal‐bolus insulin therapy compared to sliding scale insulin in the inpatient management of patients with type 2 diabetes undergoing general surgery

Medium risk of bias. Computer generated randomisation. Open label. Inclusion and exclusion criteria detailed. Baseline characteristics the same. Unrestricted educational event from Sanofi Aventis

II Appropriately powered. FBG p=0.037, mean BG p<0.001, BG <140mg/dL p<0.001. Secondary outcomes p=0.003, wound infection p=0.05.

Large effect. BBI better than SSI in treating hyperglycaemia. Significant difference between groups in regards glycaemic control, p<0.001, lower mean daily BG and FPG. More hypoglycaemia in BBI group. Difference in secondary endpoints favour BBI, with reduced wound infections and ICU stay. No difference in LOS or mortality. Significant difference in units of insulin used.

Tertiary hospital. Similar patient population and treating teams. Difference is ceasing all oral hypoglycaemic agents on admission. Sliding scale not appropriate comparator.

Appendix 8: How Should Patients on Insulin Pump Therapy be Managed in Hospital?

Table 8.1. General recommendations for diabetic patients who continue continuous subcutaneous insulin infusion therapy in hospital.

• CSII therapy is to be continued in hospital only in those situations where the patient or their guardian have the ability to self‐manage their insulin dosing and the pump (button pushing and set changes) safely.

• CSII should never substitute for an intravenous insulin infusion to treat patients with diabetic ketoacidosis.
• In a metabolically stable patient, who is able to eat, CSII may be more appropriate than an intravenous insulin infusion or a basal + top up injection

regimen.
• Regardless as to whether CSII is to be continued or ceased during the patient’s hospitalization it is strongly recommended that an endocrine service

consultation (if available) is obtained for all patients at the time of admission. The endocrinologist usually responsible for the care of the patient should be notified at the time of admission.

• The patient will be responsible (in consultation with the endocrine team) for setting basal rates, determining bolus doses administered with meals or to correct elevated glucose levels and for set changes.

• Comprehensive documentation all aspects of CSII management is required.
• CSII therapy must be substituted with either a subcutaneous insulin regimen or an intravenous insulin infusion if:

1/ The patient or guardian is not able to demonstrate that they are able to safely and reliably manage the insulin pump. 2/ A severe acute illness is present. 3/ A procedure or investigation is planned potentially rendering CSII therapy ineffective or the anaesthetic staff are not confident regarding the management of the pump. 4/ There are concerns regarding a malfunction in the pump.

• Should there be concerns regarding the technical functioning of the pump manufacturer’s help line should be contacted.
• CSII therapy should never be discontinued without first ensuring the provision of insulin via an alternative route (IV infusion or subcutaneous

injection)
• The patient’s admission to hospital should be used as an opportunity to review all aspects of CSII management by the Endocrinology team.

Table 8.2. Minimum patient competency requirements for continued inpatient CSII therapy.

• Ability to operate the management menu of the device to alter basal rates.
• Ability to operate the management menu of the device to modify parameters of and operate the bolus calculator (including a basic level of

proficiency in carbohydrate counting)
• Ability to perform a set change and manage line occlusions or leaks and have relevant supplies to implement a set change.

Table 8.3. Contraindications to continued inpatient CSII therapy.

• Patient is unable to demonstrate a basic level of competency in the operation of their insulin pump.
• Impaired or fluctuating conscious state.
• Major psychiatric disorder (psychosis)
• Severe acute illness (including diabetic ketoacidosis) requiring an insulin infusion
• Lack of supplies (infusion sets, batteries and other equipment required to continue the patient on CSII therapy)
• Extensive skin infections or inflammation.
• Concerns regarding technical malfunction of the pump.
• Numerous radiological procedures (CT and MRI). The pump should be suspended and disconnected prior to the patient entering a CT or MRI

scanner.
• Patient undergoing lengthy or complicated surgery, or serious medical illness likely to be accompanied by significant metabolic disturbance.

Table 8.4. Hospital documentation recommended for inpatients continuing CSII.

• The model of the pump and duration of CSII.
• Date current pump purchased.
• Details of insulin delivery line.
• The name of the insulin infused with an indication that it is being delivered via a pump.
• Insulin delivery parameters including basal rates, insulin to carbohydrate ratios, correction factors, duration of insulin action and glucose targets.
• Any changes to the pump insulin delivery settings should be clearly documented at the time they are implemented. It should also be documented

that these changes have been clearly conveyed to and confirmed by the patient or their guardian.
• The date and time of set changes. A follow‐up fingerprick glucose should be performed 2 hrs later and documented.
• Fingerprick glucose readings. At least 4 (pre‐meal and bed‐time) and preferably 7 (pre‐meal, 2 hour post meal and bed‐time) finger‐prick glucose

readings are recommended. These should be documented on the glucose monitoring chart.
• Ketone readings. Blood ketones are preferable to urinary ketone measurements.
• A signed agreement with the patient that clearly documents the patient’s responsibilities with regard to inpatient CSII management is

recommended.

Table 8.5. Intra‐operative conditions appropriate for CSII or switch to temporary insulin infusion

Situations appropriate for intra‐operative CSII Indications for intraoperative intravenous insulin infusion

• Procedure of short duration (e.g. D&C).
• Medical and anaesthetic staff that are familiar with pumps.
• Patient awake and alert intraoperatively.
• Patient metabolically stable.
• Patient alert and to resume eating shortly after completion of the

procedure.

• Prolonged and complicated procedure (eg coronary bypass surgery).

• Medical and anaesthetic staff unfamiliar with CSII.
• Patient critically unwell and metabolically unstable.
• Prolonged post‐operative recovery period.

Table 8.6. Recommendations for perioperative management of CSII.

Situation Recommendations

Preoperative
• Perform a set change on the morning or afternoon of the day prior to the procedure. Ensure that the insertion site is well away from the operative field.

• In fasting patients consider infusing insulin at a reduced temporary basal rate eg 70%.
• IV dextrose (eg 5% at 80ml/hr) should be infused to prevent ketosis.
• No boluses are necessary unless they are used to correct elevated glucose levels.
• Monitor glucose levels and ketones with increased frequency as per the hospital’s established protocols.
• If CSII is to be continued intra‐operatively consent must be obtained from the patient or their guardian.
• If CSII is to be continued intra‐operatively a label which is clearly visible must be attached to the patient stating that they have Type 1

diabetes and are using an insulin pump. Intraoperative
• Ensure that the anaesthetist is aware that CSII is to be continued during surgery and is able to disconnect the pump if necessary.

• Ensure that the pump is accessible to the anaesthetist intraoperatively.
• Glucose levels should be monitored frequently (at least hourly) and ketones as determined by the anaesthetist.
• In the event of an unexplained elevation in glucose levels or frank ketosis an IV insulin infusion should be commenced and the insulin

pump suspended and disconnected.
• In the event of intraoperative hypoglycaemia, the pump should be suspended immediately and a bolus of IV dextrose administered.

Once hypoglycaemia has been abolished the insulin pump can be recommenced at a reduced basal rate. Alternatively the IV dextrose can be infused at a greater rate. In the face of unstable glucose levels, and an anaesthetist with limited CSII experience, CSII should be ceased and a formal IV insulin infusion commenced.

• Bipolar diathermy is contraindicated. Unipolar diathermy can be used. Ceasing and Recommencing CSII

• When commencing patients managed with CSII on a subcutaneous insulin regimen the first injection(s) should be given at the time CSII is ceased and should include a long acting insulin analogue.

• An intravenous insulin infusion should be commenced within 2 hours of cessation of CSII.
• In hospitalized patients where CSII has been ceased with the patient managed on an insulin infusion or multiple daily injections, when

recommencing CSII is preferable that CSII be recommenced in the morning and with the insertion of a new line.
• In those managed with multiple daily injections while an inpatient, if they are on a long acting analogue administered in the evening,

half the dose should be given on the night with CSII commenced the next morning.
• CSII should be re‐commenced immediately prior to cessation of an insulin infusion.

Appendix 9: Appropriate Glucose Control in End of Life Situations

Table 9.1. Phases of end of life pathway

Phase Description

Stable The person’s symptoms are adequately controlled on their established management plan but interventions to maintain symptom control and quality of life have been planned. This phase may last for several years.

Unstable The person develops a new unexpected problem or a rapid increase in the severity of existing problems.

Deteriorating The person’s existing symptoms gradually worsen or they develop new but unexpected problems.

Terminal Death is likely in a matter of days and no acute intervention is planned or required.

Table 9.2. Suggested inpatient management of type 1 and type 2 diabetes in the phases of the end of life pathway

Phase Type 1 diabetes Type 2 diabetes Stable Inpatient management of glycaemia as per standard care. Insulin

should not be ceased. Hospitalisation may provide an opportunity to review the appropriateness of the patient’s current insulin regimen, glycaemia targets, need for any non‐diabetes medications exacerbating glycaemia and general diabetes education. Stable Phase palliative patients are usually discharged home from hospital. Post‐discharge follow‐up should be organized.

Inpatient management of glycaemia as per standard care. Hospitalisation may provide an opportunity to review the appropriateness of the patient’s current diabetes regimen, glycaemia targets, need for any non‐diabetes medications exacerbating glycaemia and general diabetes education. Stable Phase palliative patients are usually discharged home from hospital. Post‐discharge follow‐up should be organised.

Unstable As for “Stable Phase”. If the patient is to be discharged home consider simplifying their insulin regimen if appropriate. Liberalisation of the patient’s diet should be considered.

As for “Stable Phase”. If the patient is to be discharged home from hospital, a review of the current medication regimen should be undertaken aiming for simplicity, minimisation of the risk for hypoglycaemia and other adverse effects associated with some oral agents. Liberalisation of the patient’s diet should be considered.

Deteriorating Adjustments to usual insulin regimen are likely to be required at the time of hospitalisation particularly if nutritional intake is reduced with cachexia, renal and hepatic impairment, delirium or altered conscious state, increasing pain, mucositis, nausea and vomiting. Insulin should not be withdrawn completely unless it is at the request the patient or their family. Consider simplification of

If the patient is on oral hypoglycaemic agents at the time of hospitalisation a decision is required as to whether these are still appropriate and the need for each reviewed. If hyperglycaemia is present and responsible for symptoms consider commencing a simple insulin regimen e.g. a single dose of basal insulin to promote comfort. If the patient is already on insulin, simplification of the

the insulin regimen e.g. single basal insulin injection with top‐ups of short acting insulin analogue to maintain comfort. Less frequent BG monitoring (1‐2 per day) and ketone checks are recommended. Aim for BGs between 5.0‐15.0 mmol/L. Remove food restrictions. Review the need for any non‐diabetes medications exacerbating hyperglycaemia or hypoglycaemia. If patient is to be discharged home from hospital, consider modifying insulin regimen aiming for simplicity and minimisation of the risk for hypoglycaemia. Ensure follow‐up and support of the patient post discharge.

current insulin regimen should be implemented if possible with less frequent BG monitoring (1‐2 per day). Aim for BGs between 5.0‐15.0 mmol/L. Remove food restrictions. Review the need for any non‐ diabetes medications exacerbating hyperglycaemia or hypoglycaemia. Ensure follow‐up and support of the patient post discharge.

Terminal The patient’s preferences or those of their carer take precedence. The primary objective is to maintain patient comfort. A single daily injection of basal insulin administration may be required to maintain comfort by addressing severe hyperglycaemia and to prevent frank ketoacidosis. Consider minimising/ceasing all glucose and ketone monitoring after the appropriate discussion with the patient or their carer.

The primary objective is to maintain comfort. Consider ceasing all glucose monitoring. Consider ceasing all insulin and oral hypoglycaemic agents. If severe hyperglycaemia and the patient symptomatic from hyperglycaemia consider commencing a daily injection of basal insulin.

Table 9.3. Factors potentially influencing management of glycaemia.

Anorexia and weight loss. Confusion and altered conscious state. The stress response to pain, anxiety, infection and unrelated intercurrent illness. Disturbance in glucose metabolism resulting from some malignant tumours. Use of corticosteroids and other diabetogenic medications. Metabolic derangement including renal and hepatic dysfunction.

Appendix 10: How Should Patients with Newly Discovered Hyperglycaemia be Followed‐Up?

Table 10.1. Incidence of newly diagnosed diabetes at various glucose and HbA1c screening thresholds

Study Aim Glucose threshold and diabetes prevalence

Subjects above Threshold

Quality Level of evidence

Risk of Bias

Measures of accuracy for HbA1c

Relevance

Krebs 2000156

Prevalence study. Retrospect‐ ive trial.

22% prevalence of undiagnosed diabetes at random plasma glucose ≥7.8 mmol/l.

88 Subjects not described

No declaration of conflict of interest. Inclusion / exclusion criteria unclear.

IV Mode‐ rate

At plasma glucose of ≥7.8mmol/L and HbA1c ≥6.0%. Sensitivity 47%. Specificity unable to calculate from the data.

New Zealand hospital. Diabetes diagnosed by various methods.

Greci 2003157

Study of diagnostic yield.

60% prevalence of undiagnosed diabetes at random plasma glucose ≥6.9 mmol/l.

35 Subjects were described

No declaration of conflict of interest. Small number of patients. Inclusion / exclusion criteria detailed.

IV High Sensitivity and specificity at random BG ≥6.9 mmol/l and HbA1c>6.0% were 57% and 100% respectively. Sensitivity and specificity at random BG ≥6.9 mmol/l and HbA1c 5.2% were 100% and 50% respectively. PPV at >6% 100%, NPV at <5.2% 100%.

Similar patient population. Diabetes diagnosed on basis of fasting BG x 2 or GTT. Better HbA1c assay available (not HPLC)

Gray 2004158

Study of diagnostic yield, but also prevalence study.

21% prevalence of undiagnosed diabetes at random plasma glucose ≥6.1 mmol/l.

62 Subjects were described

Declarations of conflict of interest. Inclusion / exclusion criteria detailed previously.

IV Mode‐ rate

With random BG of ≥6.1 mmol/l and HbA1c ≥6.2% Sensitivity=86%, specificity=94% PPV 80%, NPV 96%

Patient group confined to those with acute stroke. UK setting. Diabetes diagnosed on basis of GTT. Hba1c ≥6.2%, HbA1c assay HPLC

George 2005159

Prevalence study.

33% prevalence of undiagnosed diabetes at random plasma glucose ≥7.0

36 Subjects not described

No declarations of conflict of interest. Inclusion / exclusion criteria

IV Mode‐ rate

HbA1c not done. UK hospital emergency department. Diabetes diagnosed on basis of fasting plasma glucose x2

mmol/l.

detailed.

Wong 2010155

Prevalence substudy of RCT of tight glucose control for myocardial infarction.

8% prevalence of undiagnosed diabetes at random plasma glucose ≥7.8 mmol/l.

55 Subjects were described

Declaration that there were no conflicts of interest. Inclusion / exclusion criteria detailed previously.

IV Mode‐ rate

HbA1c not reported. Australian population. Patient group confined to hyperglycaemic patients with myocardial infarction. Diabetes diagnosed mainly on basis of GTT.

Valentine 2011160

Study of diagnostic yield, but also prevalence study.

11% prevalence of undiagnosed diabetes at random plasma glucose ≥5.5 mmol/l.

2672 Subjects were described

Declaration of conflict of interest from authors. No inclusion / exclusion criteria

IV Mode‐ rate

No measures of accuracy. Similar patient population, general medical patients in Australian setting. Diabetes diagnosed on basis of HbA1c ≥6.5% (HPLC) only. Poor uptake of oGTT post discharge.

De Mulder 2011161

Study of diagnostic yield, but also prevalence study.

25% prevalence of undiagnosed diabetes at random plasma glucose ≥7.8 mmol/l.

109 Subjects were described

Declaration that there were no conflicts of interest. Inclusion / exclusion criteria detailed.

IV Mode‐ rate

At random BG of ≥7.8mmol/L and HbA1c ≥6.5%. Sensitivity 29%, specificity 100%, PPV 100%, NPV 71%

Dutch hospital. Patient group confined to hyperglycaemic patients with acute coronary syndrome. Diabetes diagnosed by GTT.

Figure 11.1. Suggested approach to diagnosis of diabetes and follow‐up of inpatient with newly discovered hyperglycaemia.

Elevated random blood glucose >7.8mmol/L

Determine HbA1c

≥6.5%* (48 mmol/mol)

<6.5% (48 mmol/mol)

Manage as diabetes

• Early referral to inpatient specialist diabetes team if available

• Appropriate outpatient follow up

#oral GTT as an outpatient or repeat

HbA1c

#Ensure appropriate communication to the GP

*International expert committee

Appendix 11: The Specialist Diabetes Inpatient Management Team.

Table 11.1. Hospital approaches to diabetes management

Approach Description Evidence Consultant Service.

The traditional hospital model of care, whereby specialised diabetes services are invited, at the discretion of the admitting team, to assist with specific patients’ diabetes management.

There is no evidence that improving this model has resulted in any substantial benefits. Anecdotal evidence suggests that this is akin to “shutting the gate once the horse has bolted”.

Systematic Hospital‐wide Diabetes Programme

These programmes aim to improve the identification of patients with diabetes and to enhance the diabetes management skills of all staff, by education and implementation of diabetes management and prescription guidelines. The responsibility of managing the patient’s diabetes remains with the admitting team.

The evidence supporting such an institution‐wide approach in improving diabetes‐related outcomes is limited to one comparative study162 which demonstrated a reduction in length of stay of 1.8 days for patients with primary diabetes following the intervention.

Specialist Diabetes Inpatient Management Team

This involves a multidisciplinary team approach, with the role of the Inpatient Diabetes Management Team varying from an advisory function to active management of the patient’s diabetes, for all patients with diabetes and usually commences at the time of the patient’s admission.

Several comparative trials (4‐9) have shown reductions in ALOS of 0.26‐5.6 days following intervention by an inpatient diabetes management team, primarily involving a specialist diabetes nurse (some with prescribing capabilities). See table 11.2.

Table 11.2. Studies examining effectiveness of Specialist Diabetes Inpatient Teams

Study Design Subjects Team Finding Comment Setting Koprosky 1997163

RCT of Diabetes Team Care vs standard care

179 Diabetes nurse educator and Endocrinologist

Primary diagnosis diabetes: median LOS 5.5 days (Team) vs 7.5 days (control) Secondary diagnosis diabetes: median LOS 10 days (Team) vs 10.5 days (control) Post‐discharge control better with Team Care: 75% good control vs 46%. Readmission lower with Team Care: 15% vs 32%, p=0.01.

Failure to achieve significance due to small sample size?

Single US medical centre

Davies 2001164

RCT of Diabetes Specialist Nurse Care vs standard care

300 Diabetes specialist nurse

Median LOS 8 days (intervention) vs 11 days (control), p<0.001. No difference readmission rate. Intervention group more satisfied with diabetes care (p<0.001). Control group £436 more expensive per patient.

Single UK teaching hospital

Cavan Retrospective 1811 Diabetes Median LOS decreased from 11 to 8 days (p<0.001) Net saving of 4171 Single UK

2001165 analysis of diabetic admissions pre and post intervention

specialist nurse

amongst medical patients and 8 to 5 days (p<0.001) in surgical patients

bed days over one year

hospital

Sampson 2006166

Retrospective analysis of diabetic admissions pre and post intervention

14722 Diabetes inpatient specialist nurse service

Mean excess bed days for diabetes admissions reduced from 1.9 days to 1.2 days after introduction of the service

Estimated 700 bed days saved per 1000 inpatients with diabetes

Single UK teaching hospital

Newton 2006167

Random chart audit pre and post intervention

Not reporte d

Diabetes clinical nurse specialist, Diabetes nurse case managers, medical director

Reduction in mean glucose levels from 9.8 mmol/L to 8.4 mmol/L (p<0.0001). Reduction in LOS for patients with diabetes from 6.01±0.32 to 5.75±0.38 days (p=0.01).

Estimated saving of $2.2M per year for the hospital

Single US tertiary care hospital

Courtney 2007168

Prospective study comparing diabetic admissions with intervention vs historical controls

452 Diabetes Specialist Nurse with prescribing rights

Reduction in medication errors from median 6 to 4 (p<0.01). Reduction in LOS from median from 9 to 7 days (p<0.05)

6 wards in UK general hospital

Flanagan 2008169

Retrospective analysis of diabetic admissions pre and post intervention

28,016 5 specialist nurses, consultant and specialty registrar

LOS fell from 8.3 0±18 to 7.7±0.10 (p=0.002). Effect predominantly in medical patients. Emergency admissions fell from 9.7±0.2 to 9.2±0.2 (p<0.001). No change in elective admissions.

Single UK teaching hospital

Brooks 2011170

Retrospective audit 1140 Endocrinologist, diabetes nurse specialist, junior doctor

Reduction in average LOS for all patients with diabetes from 9.39 to 3.76 days. No change in average LOS for patients with primary diagnosis of diabetes

Estimated cost saving £132,500 over 3 months

Single UK district hospital

Table 11.3. The role of the Specialist Diabetes Inpatient Team

Improving diabetes management expertise throughout the hospital, Development and implementation of specific diabetes management protocols, Direct management of diabetes with specific referral criteria, Ward liaison, troubleshooting, management advice, Discharge planning,

Appendix 12: What Routine Measures Should be Undertaken for People with Diabetes Admitted to Hospital?

Table 12.1: Role of various team members in ensuring optimal routine diabetes care in hospital and after discharge.

Team member Role in hospital management Role in discharge planning

Diabetes Educator

Ensure appropriate blood glucose testing and quality control of glucose testing kits, support to ward nursing staff.

Provide clinical leadership and continuing stable care in an environment of transitional and rotating medical, nursing and allied health workforce.

Report back to the diabetes team if input should be offered to the referring unit.

Assess and consolidate knowledge and skills regarding eating plan, physical activity, self‐monitoring, medication usage, insulin adjustment, sick day management, foot care.

Qualified professionals are "ADEA Credentialled Diabetes Educators"176. If available, the services of a diabetes educator are useful in the early stages and a continuing liaison can be established.

Dietitian Ensure appropriate diet in hospital and nutritional needs are met.

Liaise with Diabetes Team regarding changes in diet particularly in the situation of enteral and parenteral nutrition.

Assess readiness to change eating behaviour, and dietary counseling that is innovative and specific to the requirements of the individual.

Provide dietetic intervention – recommendations that include consistency in day‐to‐ day carbohydrate intake, substitution of sucrose‐containing foods, usual protein intake, cardio protective nutrition interventions, weight management strategies, regular physical activity

Pharmacist Medicines review Home medicines review on discharge for patients with co – morbidities.

Podiatrist Podiatric advice and initial management of high risk foot

Organise follow up management of high risk foot. Note Medicare Enhanced Primary Care item available for GP referrals.

Aboriginal Health Worker

Providing culturally appropriate and practical support and counseling for ongoing care, thus improving patient understanding and adherence to treatment programs.

Social Worker Liaison with family and social services Without ensuring other aspects of the patient and his/her family welfare are catered for, good diabetes management may be challenging to achieve.

Contributors to the ADS Guidelines for Routine Glucose Control in Hospital

Writing Party

A/ Prof N Wah Cheung

Dr David Chipps

Ms Shirley Cornelius*

Dr Barbara Depczynski

Ms Kristine Heels*

Dr David O’Neal

Dr Jennifer Wong

A/Prof Sophia Zoungas

*Representing Australian Diabetes Educators Association

Expert Consultative Panel and Additional Contributors

Prof Frank Alford

Ms Jan Alford

A/Prof Sofianos Andrikopoulos

A/Prof Mark Boughey

Prof Lesley Campbell

Dr Jennifer Conn

Prof Timothy Davis

A/Prof Michael D’Emden

Dr Shirley Elkassaby

Dr P Gerry Fegan

A/Prof Jeff Flack

A/Prof Jenny Gunton

A/Prof Sanghamitra Guha

Dr Jane Holmes‐Walker

A/Prof Alicia Jenkins

A/Prof Maarten Kamp

Ms Jen Kinsella

Dr Margaret Layton

Ms Fiona McIver

Dr Alison Nankervis

Ms Ann O’Neill

Dr Glynis Ross

A/Prof Ashim Sinha

Dr Stephen N Stranks

Prof Helena Teede

Dr Vincent Wong

A workshop with the Expert Consultative Panel was held with support from the National Diabetes Services Scheme in July 2012.

GLOSSARY OF TERMS USED RELATING TO INSULIN

Sliding scale insulin

The prescription of insulin given in a variable dose depending on the glucose level at the time. Often this is the sole insulin prescribed.

Supplemental insulin

The administration of variable dose insulin to correct hyperglycaemia, given in conjunction with appropriate adjustments to the patient’s scheduled anti‐diabetic therapy.

Correctional insulin This term is used interchangeably with “supplemental insulin”

Basal insulin

Intermediate or long acting insulin providing background insulin (eg

Protaphane®, Humulin NPH®, Lantus®, Levemir®)

Prandial insulin

Rapid acting insulin given at mealtimes (eg Novorapid®, Humalog®, Apidra®)

Basal bolus insulin Insulin regime comprising the combination of a basal insulin with multiple daily doses of prandial insulin

Ispohane insulin Intermediate acting insulin (eg Protaphane®, Humulin NPH®) Continuous subcutaneous insulin infusion (CSII)

Insulin administered by continuous subcutaneous infusion via an patient self‐managed insulin pump

REFERENCES

1) Dunstan D, Zimmet P, Welborn T et al. The rising prevalence of diabetes and impaired glucose tolerance: The Australian Diabetes, Obesity and Lifestyle Study. Diabetes Care 2002; 25: 829‐34.

2) Barr E, Maliano D, Zimmet P et al AusDiab 2005. The Australian Diabetes, Obesity and Lifestyle Study. Tracking the accelerating epidemic: its causes and outcomes. Report. Melbourne: International Diabetes Institute, 2006.

3) Australian Institute of Health and Welfare 2008. Diabetes: Australian facts 2008. Diabetes series no. 8. Cat. no. CVD 40. Canberra: AIHW.

4) Aubert R, Geiss L, Ballard D, Cocanougher B, Herma. We Diabetes‐related hospitalization and hospital utilization. Diabetes in America. 2nd Ed. Bethesday, MD: National Institutes of Health; 533‐559, 1995.

5) Levetan C, Passaro M, Jablonski K, Kass M, Ratner R. Unrecognised diabetes among hospitalized patients. Diabetes Care 21:246‐9, 1998.

6) Abourizk NN, Vora CK, Verma PK. Inpatient diabetology: The new frontier. J Gen Intern Med 2004; 18: 466‐71.

7) Rando L, Keith C, Sardjono DA, Robertson MB, Colman PG. Diabetes Ward management – Room for Improvement. J Pharm Pract Res 2004; 34: 95‐99.

8) Wallymahmed ME, Dawes S, Clarke G, Saunders S, Younis N, MacFarlane IA. Hospital in‐patients with diabetes: increasing prevalence and management problems. Diabet Med 2005; 22: 107‐9.

9) NHS Institute for Innovation and Improvement. Focus on: Inpatient care for people with diabetes. Coventry: 2008.

10) Umpierrez GE, Isaacs SD, Bazargan N, You X, Thaler LM, Kitabchi AE. Hyperglycemia: An independent marker of in‐hospital mortality in patients with undiagnosed diabetes. J Clin Endocrinol Metab 2002; 87: 978‐982.

11) Clement S, Braithwaite SS, Magee MF et al. Management of diabetes and hyperglycemia in hospitals. Diabetes Care 2004; 27: 553‐591.

12) Capes SE, Hunt D, Malmberg K, Gerstein HC. Stress hyperglycemia and increased risk of death after myocardial infarction in patients with and without diabetes: a systematic overview. Lancet 2000: 355: 773‐778.

13) Wong V, Ross DL, Park K, Boyages S, Cheung NW. Hyperglycemia following acute myocardial infarction is a predictor of poor cardiac outcomes in the reperfusion era. Diabetes Res Clin Pract 2004; 64: 85‐91.

14) Stranders I, Diamant M, van Gelder RE et al. Admission blood glucose level as risk indicator of death after myocardial infarction in patients with and without diabetes mellitus. Arch Int Med 2004; 164: 982‐988.

15) Timmer JR, van der Horst ICC, Ottervanger JP et al. Prognostic value of admission glucose in non‐ diabetic patients with myocardial infarction. Am Heart J 2004; 148: 399‐404.

16) Kosiborod M, Rathore SS, Inzucchi SE et al. Admission glucose and mortality in elderly patients hospitalized with acute myocardial infarction. Circulation 2005; 111: 3078‐86.

17) Straumann E, Kurz DJ, Muntwyler J et al. Admission glucose concentrations independently predict early and late mortality in patients with acute myocardial infarction treated by primary or rescue percutaneous coronary intervention. Am Heart J 2005; 150: 1000‐1006.

18) Meier JJ, Launhardt V, Deifuss A et al. Plasma glucose at hospital admission and previous metabolic control determine myocardial infarct size ans survival in patients with and without type 2 diabetes. Diabetes Care 2005; 28: 2551‐2553.

19) Goyal A, Mahaffey KW, Garg J et al. Prognostic significance of the change in glucose level in the first 24h after acute myocardial infarction: results from the CARDINAL study. Eur Heart J 2006; 27: 1289‐ 1297.

20) Bhadriraju S, Ray KK, DeFranco AC et al. Association between blood glucose and long‐term mortality in patients with acute coronary syndromes in the OPUS‐TIMI 16 trial. Am J Cardiol 2006; 97: 1573‐1577.

21) Naber CK, Mehta RH, Junger C et al. Impact of admission blood glucose on outcomes of nondiabetic patients with acute ST‐elevation myocardial infarction (from the German Acute Coronary Syndromes [ACOS] Registry). Am J Cardiol 2009; 103: 583‐87.

22) Sinnaeve PR, Steg GS, Fox KAA et al. Association of elevated fasting glucose with increased short‐term and 6‐month mortality in ST‐segment elevation and non‐ST‐segment elevation acute coronary syndromes. Arch Int Med 2009; 169: 402‐09.

23) Ishihara M, Kojima S, Sakamoto T et al. Comparison of blood glucose values on admission for acute myocardial infarction in patients with versus without diabetes mellitus. Am J Cardiol 2009; 104: 769‐74.

24) Dziewierz A, Giszterowicz D, Siudak Z et al. Impact of admission glucose level and presence of diabetes mellitus on mortality in patients with non‐ST‐segment elevation acute coronary syndrome treated conservatively. Am J Cardiol 2009; 103: 954‐58.

25) Goyal A, Mehta SR, Diaz R et al. Differential clinical outcomes associated with hypoglycemia and hyperglycemia in acute myocardial infarction. Circulation 2009; 120: 2429‐37.

26) De Mulder, Cornel JH, van der Ploeg T, Boersma E, Umans VA. Elevated admission glucose is associated with increased long‐term mortality in myocardial infarct patients, irrespective of the initially applied reperfusion strategy. Am Heart J 2010; 160: 412‐19.

27) Timmer JR, Hoekstra M, Nijsten MWN et al. Prognostic value of admission glycosylated haemoglobin and glucose in non‐diabetic patients with ST‐segment elevation myocardial infarction treated with percutaneous coronary intervention. Circulation 2011; 124: 704‐11.

28) Cheung NW, Wong V, McLean M. The Hyperglycemia: Intensive Insulin Infusion In Infarction (HI‐5) Study ‐ A randomised controlled trial of insulin infusion therapy for myocardial infarction. Diabetes Care 2006; 29: 765‐70.

29) Cheung NW, Wong VW, McLean M. What glucose target should we aim for in myocardial infarction? Diabetes Res Clin Pract 2008; 80: 411‐15.

30) Kosiborod M, Inzucchi SE, Krumholz HM et al. Glucose normalization and outcomes in patients with acute myocardial infarction. Arch int Med 2009; 169(5): 438‐46.

31) Malmberg K, Ryden L, Efendic S et al. Randomised trial of insulin‐glucose infusion followed by subcutaneous insulin treatment in diabetic patients with acute myocardial infarction (DIGAMI Study): effects on mortality at 1 year. J Am Coll Cardiol 1995; 26: 57‐65.

32) Pittas AG, Siegel RD, Lau J. Insulin therapy for critically ill hospitalized patients. Arch Intern Med 2004; 164: 2005‐11.

33) Zhao YT, Weng CL, Chen ML et al. Comparison of glucose‐insulin‐potassium and insulin‐glucose as adjunctive therapy in acute myocardial infarction: a contemporary meta‐analysis or randomized controlled trials. Heart 2010; 96: 1622‐26.

34) Devine MJ, Chandrasekara WM, Hardy KJ. Review: Management of hyperglycaemia in acute coronary syndrome. Br J Diabetes Vasc Dis 2010; 10: 59‐65.

35) Lipton JA, Can A, Akoudad S, Simoons ML. The role of glucose normalization and insulin therapy in patients with acute coronary syndrome. Neth Heart J 2011; 19: 79‐84.

36) Kansagara D, Fu R, Freeman M, Wolf F, Helfand M. Intensive insulin therapy in hospitalized patients. A systematic review. Ann Intern Med 2011; 154: 268‐82.

37) Davies RR, Newton RW, McNeill GP, Fisher BM, Kesson CM, Pearson D. Metabolic control in diabetic subjects following myocardial infarction. Scott Med J 1991; 36: 74‐76.

38) Scott JF, Robinson GM, French JM, O’Connell JE, Alberti KG, Gray CS. Glucose potassium insulin infusions in the treatment of acute stroke patients with mild to moderate hyperglycemia. Stroke 1999; 30: 793‐799.

39) Lazar HL, Chipkin S, Philippides G, Bao Y, Apstein C. Glucose‐insulin‐potassium solutions improve outcomes in diabetics who have coronary artery operations. Ann Thorac Surg 2000; 70: 145‐150.

40) Szabo Z, Arnqvist H, Hakanson E, Jorfeldt L, Svedjeholm R. Effects of high‐dose glucose‐insulin‐ potassium on myocardial metabolism after coronary surgery in patients with type II diabetes. Clin Sci (Lond) 2001; 101: 37‐43.

41) Van den Berghe G, Wouters P, Weekers F, et al. Intensive insulin therapy in critically ill patients. N Eng J Med 2001; 345: 1359‐1367.

42) Groban L, Butterworth J, Legault C, Rogers AT, Kon ND, Hammon JW. Intraoperative insulin therapy does not reduce the need for inotropic or antiarrhythmic therapy after cardiopulmonary bypass. J Cardiothorac Vasc Anesth 2002; 16: 405‐412.

43) Smith A, Grattan A, Harper M, Royston D, Riedel BJ. Coronary revascularization: a procedure in transition from on‐pump to off‐pump? J Cardiothorac Vasc Anesth 2002; 16: 413‐420.

44) Malmberg K, Ryden L, Wedel H et al. Intense metabolic control by means of insulin in patients with diabetes mellitus and acute myocardial infarction (DIGAMI 2); effects on mortality and morbidity. Eur Heart J 2005; 26: 650‐661.

45) Diaz R, Paolasso EC, Piegas LS et al. Metabolic modulation of acute myocardial infarction: the ECLA glucose‐insulin‐potassium pilot trial. Circulation 1998; 98: 2227‐2234.

46) Ceremuzynski L, Budaj A, Czepiel A et al. Low‐dose glucose‐insulin‐potassium is ineffective in acute myocardial infarction: results of a randomized multicenter Pol‐GIK trial. Cardiovasc Drugs Ther 1999; 13: 191‐200.

47) The CREATE‐ECLA Trial Group Investigators. Effect of glucose‐insulin‐potassium infusion on mortality in patients with acute ST‐segment elevation myocardial infarction. JAMA 2005; 293: 437‐446.

48) Van der Horst ICC, Zijlstra F, van’t Hof AWJ et al. Glucose‐insulin‐potassium infusion in patients treated with primary angioplasty for acute myocardial infarction. J. Am. Coll. Cardiol. (2003) 42: 784‐791.

49) Rasoul S, Ottervanger JP, Timmer JR et al. One year outcomes after glucose‐insulin‐potassium in ST elevation myocardial infarction. The Glucose‐insulin‐potassium study II. Int J Cardiol 2007; 122: 52‐55.

50) Oksanen T, Skrifvars MB, Varpula T et al. Strict versus moderate glucose control after resuscitation from ventricular fibrillation. Intensive Care Med 2007; 33: 2093‐100.

51) Malmberg K. Prospective randomized study of intensive insulin treatment on long term survival after acute myocardial infarction in patients with diabetes mellitus. Br. Med. J. (1997) 314: 1512‐15.

52) Timmer JR, Svilaas T, Ottervanger JP et al. Glucose‐insulin‐potassium infusion in patients with acute myocardial infarction without signs of myocardial infarction: The Glucose‐Insulin‐Potassium Study (GIPS)‐II. J Am Coll Cardiol 2006; 47: 1730‐31.

53) Cheung NW. Glucose control during acute myocardial infarction. Intern Med J 2008; 38: 345‐48.

54) Ryden L, Standl E, Bartnik M, et al. Guidelines on diabetes, pre‐diabetes, and cardiovascular diseases: executive summary. Eur Heart J 2007; 28: 88‐136.

55) Moghissi ES, Korytkowski MT, DiNardo M et al. American Association of Clinical Endocrinologists and American Diabetes Association Consensus Statement on inpatient glycemic control. Diabetes Care 2009; 32: 1119‐1131.

56) 2009 focused updates: ACC/AHA guidelines for the management of patients with ST‐elevation myocardial infarction (updating the 2004 guideline and 2007 focused update) and ACC/AHA/SCAI guideline on percutaneous coronary intervention (updating the 2005 guideline and 2007 focused update). Cath cardiovcasc Intervent 2009; 74: E25‐68.

57) Qaseem A, Humphrey LL, Chou R, Snow V, Shekelle P. Use of intensive insulin therapy for the management of glycemic control in hospitalized patients: A clinical practice guideline from the American College of Physicians. Ann Intern Med 2011; 154: 260‐67.

58) Capes SE, Hunt D, Malmberg, K, Pathak P, Gerstein HC. Stress hyperglycaemia and prognosis of stroke in nondiabetic and diabetic patients: a systematic overview. Stroke 2001; 32: 2426‐2432.

59) Baird TA, Parsons MW, Phanh T et al. Persistent poststroke hyperglycemia is independently associated with infarct expansion and worse clinical outcomes. Stroke 2003; 34: 2208‐14.

60) Allport LE, Burcher KS, Baird TA et al. Insular cortex ischemia is independently associated with acute stress hyperglycemia. Stroke 2004; 35: 1886‐91.

61) Alvarez‐Sabin J, Molina CA, Ribo M et al. Impact of admission hyperglycemia on stroke outcome after thrombolysis: Risk stratification in relation to time to reperfusion. Stroke 2004; 35: 2493‐98.

62) Farrokhnia N, Bjork E, Lindback J, Terent A. Blood glucose in acute stroke, different therapeutic targets for diabetic and non‐diabetic patients? Acta Neuro Scand 2005; 112: 81‐87.

63) Stollberger C, Exner I, Finsterer J, Slany J, Steger C. Stroke in diabetic and non‐diabetic patients: course and prognostic value of serum glucose. Annals Med 2005; 37: 357‐64.

64) Gentile NT, Seftchick MW, Huynh T, Kruus LK, Gaughan J. Decreased mortality by normalizing blood glucose after ischemic stroke. Acad Emerg Med 2006; 13: 174‐80.

65) Yong M, Kaste M. Dynamic of hyperglycemia as a predictor of stroke outcome in the ECASS‐II Trial. Stroke 2008; 39: 2749‐55.

66) Fuentes B, Castillo J, San Jose B et al. The prognostic value of capillary glucose levels in acute stroke: The Glycemia in Acute Stroke (GLIAS) Study. Stroke 2009; 40: 562‐68.

67) Stead L, Gilmore R, Bellolio M et al. Hyperglycemia as an independent predictor of worse outcome in non‐diabetic patients presenting with acute ischemic stroke. Neurocrit Care 2009; 10: 181‐868.

68) Poppe AY, Majumdar SR, Jeerakathil T, Ghali W, Buchan AM, Hill MD. Admission hyperglycaemia predicts a worse outcome in stroke patients treated with intravenous thrombolysis. Diabetes Care 2009; 32: 617‐22.

69) Ntaios G, Egli M, Faouzi M, Michel P. J‐shaped association between serum glucose and functional outcome in acute ischemic stroke. Stroke 2010; 41: 2366‐70.

70) Ahmed N, Davalos A, Eriksson E et al. Association of admission blood glucose and outcome in patients treated with intravenous thrombolysis. Arch Neurol 2010; 67: 1123‐30.

71) Dziedzic T, Pera J, Trabka‐Janik E, Szczudlik A, Slowik A. Impact of postadmission glycemia on stroke outcome: glucose normalization is associated with better survival. Atherosclerosis 2010; 211: 584‐88.

72) Saposnik G, Kapral MK, Liu Y et al. IScore: A risk score to predict death early after hospitalization for an acute ischemic stroke. Circulation 2011; 123: 739‐49.

73) Kimura K, Sakamoto Y, Iguchi Y et al. Admission hyperglycemia and serial infarct volume after t‐PA therapy in patients with and without early recanalization. J Neurol Sci 2011; 307: 55‐59.

74) Hu GC, Hsieh SF, Chen YM, Hu YN, Kang YN, Chien KL. The prognostic roles of initial glucose level and functional outcomes in patients with ischemic stroke: difference between diabetic and nondiabetic patients. Disabil Rehab 2012; 34: 34‐39.

75) Kruyt ND, Biessels GJ, DeVries JH, Roos YB. Hyperglycemia in acute ischemic stroke: pathophysiology and clinical management. Nat Rev Neurol 2010; 6: 145‐55.

76) Bellolio MF, Gilmore RM, Stead LG. Insulin for glycaemic control in acute ischaemic stroke. Cochrane Database of SystematicReviews 2011, Issue 9.

77) Walters MR, Weir CJ, Lees KR. A randomized controlled pilot study to investigate the potential benefit of intervention with insulin in hyperglycaemic acute ischemic stroke patients. Cerebrovasc Dis 2006; 22: 116‐22.

78) Gray CS, Hildreth AJ, Sandercock PA et al. Glucose‐potassium‐insulin infusions in the management of post‐stroke hyperglycaemia: the UK Glucose Insulin in Stroke Trial (GIST‐UK). Lancet 2007; 6: 397‐406.

79) Azevedo JR, Lima ER, Cossetti RJ, Azevedo RP. Intensive insulin therapy versus conventional glycemic control in patients with acute neurological injury: a prospective controlled trial. Arq Neuropsiquiatr 2007; 65: 733‐8.

80) Bruno A, Kent TA, Coull BM et al. Treatment of hyperglycemia in ischemic stroke (THIS): A randomized pilot trial. Stroke 2008; 39: 384‐89.

81) Yang M, Guo Q, Zhang X et al. Intensive insulin therapy on infection rate, days in NICU, in‐hospital mortality and neurological outcome in severe traumatic brain injury patients: a randomized controlled trial. Int J Nurs Stud 2009; 46: 753‐8.

82) Vinychuk SM, Malnyk VS, Margitich VM. Hyperglycemia after acute ischemic stroke: prediction, significance and immediate control with insulin‐potassium‐saline‐magnesium infusions. Heart Drug 2005; 5: 197‐204.

83) Kreisel SH, Berschin UM, Hammes HP et al.Pragmatic management of hyperglycaemia in acute ischaemic stroke: safety and feasibility of intensive intravenous insulin treatment. Cerebrovasc Diseases 2009; 27:167–75.

84) Johnston KC, Hall CE, Kissela BM, Bleck TP, Conaway MR, GRASP Investigators. Glucose Regulation in Acute Stroke Patients (GRASP) trial: a randomized pilot trial. Stroke 2009; 40:3804–9.

85) Middleton S, McElduff P, Ward J et al. Implementation of evidence‐based treatment protocols to manage fever, hyperglycaemia and swallowing dysfunction in acute stroke improves 90‐day outcomes: QASC, a cluster randomised controlled trial. Lancet 2011; 378 (9804): 1699‐1706.

86) AHA/ASA Guideline. Guidelines for the early management of adults with ischemic stroke. Circulation 2007; 115: e478‐e534.

87) European Stroke Organisation (ESO) Executive Committee and the ESO Writing Committee. Guidelines for management if ischaemic stroke and transient ischaemic attack 2008. Cerebrovasc Dis 2008; 25: 457‐507.

88) Pomposelli J, Baxter J, Babineau T et al. Early postoperative glucose control predicts nosocomial infection rate in diabetic patients: J Parenter Enter Nutr 22:77–81, 1998.

89) Golden SH, Kao WHL, Peart‐Vigilance C, Brancati FL. Perioperative glycemic control and the risk of infectious complications in a cohort of adults with diabetes. Diabetes Care 1999; 22: 1408‐1414.

90) Umpierrez GE, Isaacs SD, Bazargan N, You X, Thaler LM, Kitabchi AE. Hyperglycemia: An independent marker of in‐hospital mortality in patients with undiagnosed diabetes. J Clin Endocrinol Metab. 2002; 87: 978‐982.

91) Cheung NW, Napier B, Zaccaria C, Fletcher JP. Hyperglycemia is associated with adverse outcomes in patients receiving total parenteral nutrition. Diabetes Care 2005; 28: 2367‐71.

92) Baker ST, Chian CY, Zajac JD, Bach A, Jerums G, MacIsaac RJ. Outcomes for general medical patients with diabetes mellitus and new hyperglycaemia. Med J Aust 2008; 188: 340‐43.

93) Cheung NW, Ma G, Li S, Crampton R. The relationship between admission blood glucose levels and hospital mortality. Diabetologia 2008; 51: 952‐5.

94) Haga KK, McClymont KL, Clarke S et al. The effect of tight glycaemic control, during and after cardiac surgery, on patient mortality and morbidity: A systematic review and meta‐analysis. J Cardiothor Surg 2011; 6:3.

95) Murad MH, Coburn JA, Coto‐Yglesias F et al. Glycemic control in non‐critically ill hospitalized patients: A systematic review and mea‐analysis. J Clin Endocrinol Metab 2012; 97: 49‐58.

96) Lazar HL, Chipkin SR, Fitzgerald CA, Bao Y, Cabral H, Apstein CS. Tight glycemic control in diabetic coronary artery bypass graft patients improves perioperative outcomes and decreases recurrent ischemic events. Circulation 2004; 109: 1497‐502.

97) Ingels C, Debaveye Y, Milants I et al. Strict blood glucose control with insulin during intensive care after cardiac surgery: impact on 4‐years survival, dependency on medical care, and quality of life. Eur Heart J 2006, 27:2716‐2724.

98) Butterworth J, Wagenknecht LE, Legault C et al. Attempted control of hyperglycemia during cardiopulmonary bypass fails to improve neurologic or neurobehavioral outcomes in patients without diabetes mellitus undergoing coronary artery bypass grafting. J Thorac Cardiovasc Surg 2005; 130: 1319.

99) Li JY, Sun S, Wu SJ. Continuous insulin infusion improves postoperative glucose control in patients with diabetes mellitus undergoing coronary artery bypass surgery. Tex Heart Inst J 2006; 33: 445‐51.

100) Barcellos Cda S, Wender OC, Azambuja PC. Clinical and hemodynamic outcome following coronary artery bypass surgery in diabetic patients using glucose‐insulin‐potassium (GIK) solution: a randomized clinical trial. Rev Bras Cir Cardiovasc 2007; 22: 275‐84.

101) Dickerson LM, Ye X, Sack JL, Hueston WJ. Glycemic control in medical inpatients with type 2 diabetes mellitus receiving sliding scale insulin regimens versus routine diabetes medications: A multicenter randomized controlled trial. Ann Fam Med 2003; 1: 29‐35.

102) Umpierrez GE, Smiley D, Zisman A et al. Randomized study of basal‐bolus insulin therapy in the inpatient management of patients with type 2 diabetes (RABBIT 2 trial). Diabetes Care 2007; 30: 2181– 2186.

103) Umpierrez GE, Hor T, Smiley D et al. Comparison of inpatient insulin regimens with detemir plus aspart versus neutral protamine Hagedorn plus regular in medical patients with type 2 diabetes. J Clin Endocrinol Metab 2009; 94:564–569

104) Umpierrez GE, Smiley D, Jacobs S et al. Randomized study of basal bolus insulin therapy in the inpatient management of patients with type 2 diabetes undergoing general surgery (RABBIT2 Surgery). Diabetes Care 2011; 34: 256–261.

105) Joshi GP, Chung F, Vann MA et al. Society for Ambulatory Anesthesia consensus statement on perioperative blood glucose management in diabetic patients undergoing ambulatory surgery. Anesth Analg 2010; 111: 1378‐87.

106) Qaseem A, Humphrey LL, Chou R, Snow V, Shekelle P. Use of intensive insulin therapy for the management of glycemic control in hospitalized patients: A clinical practice guideline from the American College of Physicians. Ann Intern Med 2011; 154: 260‐67.

107) Kreyman KG, Berger MM, Deutz NEP et al. ESPEN Guidelines on Enteral Nutrition: intensive care. Clin Nutr 2006; 25: 210‐23.

108) Korytkowski MT, Salata RJ, Koerbel GL et al. Insulin therapy and glycemic control in hospitalized patients with diabetes during enteral nutrition therapy: a randomized controlled clinical trial. Diabetes Care 2009;32: 594‐6.

109) GraingerA, Eiden K, Kemper J, Reeds D. A pilot study to evaluate the effectiveness of glargine and multiple injections of lispro in patients with type 2 diabetes receiving tube feedings in a cardiovascular intensive care unit. Nutr Clin Pract 2007; 22: 545‐52.

110) Donihi AC, Raval D, Saul M, Korytkowski MT, DeVita MA. Prevalence and predictors of corticosteroid‐ related hyperglycemia in hospitalized patients. Endocr Pract. 2006; 12: 358‐362.

111) Fong AC, Cheung NW. The high incidence of steroid induced hyperglycaemia in hospital. Abstract, Australian Diabetes Society Annual Scientific Meeting 2011.

112) Burt MG, Roberts GW, Aguilar‐Loza NR, Frith P, Stranks SN. Continuous monitoring of circadian glycemic patterns in patients receiving prednisolone for COPD. J Clin Endocrin Metab 2011; 96: 1789‐ 1996.

113) Feldman‐Billard S, Lissak B, Kassaei R, Héron E. Short‐term tolerance of pulse methylprednisolone therapy in patients with diabetes mellitus. Ophthalmology 2005; 112: 511‐515.

114) Gurwitz JH, Bohn RL, Glynn RJ, Monane M, Mogun H, Avorn J. Glucocorticoids and the risk for initiation of hypoglycemic therapy. Arch Intern Med 1994; 154:97‐101.

115) Fajans SS, Conn JW. An approach to prediction of diabetes mellitus by modification of the glucose tolerance test with cortisone. Diabetes 1954; 3: 296‐304.

116) Henriksen JE, Alford F, Ward GM, Beck‐Nielsen H. Risk and mechanism of dexamethasone‐induced deterioration of glucose tolerance in non‐diabetic first‐degree relatives of NIDDM patients. Diabetologia, 1997;40:1439–1448.

117) Feldman‐Billard S, Lissak B, Benrabah R, Kassaei R, Héron E. Intravenous pulse methylprednisolone therapy in eye disease effect on glucose tolerance. Ophthalmology 2003; 110: 2369‐2371

118) Schneiter P, Tappy L. Kinetics of dexamethasone‐induced alterations of glucose metabolism in healthy humans. Am J Physiol 1998; 275(5 Pt 1): E806‐13.

119) Pellacani A, Fornengo P, Bruno A et al. Acute methylprednisolone administration induces a transient alteration of glucose tolerance and pyruvate dehydrogenase in humans. Eur J Clin Invest 1999; 29: 861‐ 7.

120) Greenstone MA, Shaw AB. Alternate day corticosteroid causes alternate day hyperglycaemia. Postgrad Med J 1987; 63: 761‐764.

121) Braithwaite SS, Kaufman C, Wittrock JR. Repaglinide monotherapy for diabetes mellitus during glucocorticoid therapy. The Endocrinol 2003; 13:163‐168.

122) Kasayama S, Tanaka T, Hashimoto K, Koga M, Kawase I. Efficacy of glimepiride for the treatment of diabetes occurring during glucocorticoids therapy. Diabetes Care 2002; 25: 2359‐2360.

123) Henriksen JE, Alford F, Ward GM, Beck‐Nielsen H. Risk and mechanism of dexamethasone‐induced deterioration of glucose tolerance in non‐diabetic first‐degree relatives of NIDDM patients. Diabetologia 1997; 40: 1439‐1448.

124) Clore JN, Thurby‐Hay L. Glucocorticoid‐induced hyperglycemia. Endocr Pract 2009; 15: 469‐474.

125) Oyer DS, Shah A, Bettenhausen S. How to manage steroid diabetes in the patient with cancer. J Support Oncol 2006; 4: 479‐483.

126) Hoogwerf B, Danese RD. Drug Selection and the management of corticosteroid‐related diabetes mellitus. Rheum Diseas Cl N Amer. 1999; 25: 489‐505.

127) Luther P, Baldwin D. Pioglitazone in the management of diabetes mellitus after transplantation. Am J Transplant 2004; 4: 2135‐2138.

128) Fujibayashi K, Nagasaka S, Itabashi N et al. Troglitazone efficacy in a subject with glucocorticoid‐ induced diabetes. Diabetes Care 1999; 22: 2088‐2089.

129) Willi SM, Kennedy A, Brant BP, Wallace P, Rogers NL, Garvey WT. effective use of thiazolidinediones for the treatment of glucocorticoid‐induced diabetes. Diabetes Res Clin Pract 2002; 58: 87‐96.

130) Villanueva G, Baldwin D. Rosiglitazone therapy of posttransplant diabetes mellitus. Transplantation. 2005; 80: 1402‐1405.

131) Haidinger M, Werzowa J, Voigt HC et al. A randomized, placebo‐controlled, double‐blind, prospective trial to evaluate the effect of vildagliptin in new‐onset diabetes mellitus after kidney transplantation. Trials 2010; 11: 91‐97.

132) Van Raalte DH, Genugten RE, Linssen MML, Ouwens DM, Diamant M. Glucagon‐like peptide‐1 receptor agonist treatment prevents glucocorticoid‐induced glucose intolerance and islet‐cell dysfunction in humans. Diab Care 2011; 34: 412‐417.

133) COIITSS Study Investgators. Corticosteroid treatment and intensive insulin therapy for septic shock in adults a randomised controlled trial. JAMA 2010; 303: 341‐348.

134) Scalzo PL, Banks IM, Harris NA, Basu A, Shah P. Glucocorticoid induced hyperglycemia in the inpatient. Abstract, American Diabetes Association 2010; 678‐P.

135) Hecking M, Werzowa J, Haidinger M, Döller D, Pacini G, Säemann M. Early basal insulin therapy prevents new‐onset diabetes after transplantation by improving endogenous insulin secretion. Abstract, American Diabetes Association 2011; 72‐OR.

136) Bevier WC, Zisser HC, Jovanovic L et al. Use of continuous glucose monitoring to estimate insulin requirements in patients with type 1 diabetes mellitus during a short course of prednisone. J Diabetes Sci Technol 2008; 2: 578‐583.

137) Hirsch IB, Paauw DS. Diabetes management in special situations. Endocrin Met Cl 1997; 26: 631‐645.

138) Mooradian AD, Berbaum M, Albert SG. Narrative review: a rational approach to starting insulin therapy. Ann Intern Med 2006; 145: 125‐134.

139) Leahy J. Insulin management of diabetic patients on general medical surgical floors. Endocr Pract 2006; 12(Suppl 3): 86‐90.

140) Wilkinson A, Davidson J, Dotta F et al. Guidelines for the treatment and management of new onset diabetes after transplantation. Clin Transplant 2005; 19: 291‐298.

141) Australian Diabetes Society. Perioperative diabetes management guidelines. ADS 2012. Available at http://www.diabetessociety.com.au/downloads/ADS%20Diabetes%20and%20Surgery%20Guidelines%20DRAFT%20May%202011.pdf.

142) American Diabetes Association. Standards of care in diabetes 2011. Diabetes Care 2011; 34(suppl 1): S11‐S61.

143) Yeldandi RR, Lurie A, Baldwin D. Comparison of once‐daily glargine insulin with twice‐daily NPH/regular insulin for control of hyperglycemia in inpatients after cardiovascular surgery. Diabetes Technol Ther. 2006; 8: 609‐16.

144) Cook CB, Boyle ME, Cisar NS et al Use of Subcutaneous Insulin Infusion (Insulin Pump) Therapy in the Hospital Setting; Proposed Guidelines and Outcome Measures. Diabetes Educator 2005; 31: 849‐857.

145) Inpatient Guidelines: Insulin Infusion Pump Management. Centre for Healthcare Improvement. Queensland Government. July 2010.

146) Noschese M, Di Nardo MM, Donihi AC et al. Patient outcomes after implementation of a protocol for inpatient insulin pump therapy. Endocrin Pract 2009. 15: 415‐424.

147) Bailon RM, Partlow BJ, Miller‐Cage VM et al. Continuous subcutaneous insulin infusion (insulin pump) therapy can be safely used in the hospital in select patients. Endocrin Pract 2009; 15: 24‐29.

148) Lee SW, Im R, Magbual R. Current perspectives on the use of continuous subcutaneous insulin infusion in the acute care setting and overview of therapy. Crit Care Nursing Q 2004. 172‐184.

149) Nassar AA, Partlow BJ, Boyle ME et al. Outpatient‐to‐inpatient transition of insulin pump therapy: successes and continuing challenges. J Diabetes Sci and Technol 2010; 4: 863‐872.

150) Cook CB. Are Two Insulin Pumps Better Than One? Web M&M.http://webmm.ahrq.gov. Case ID 192.

151) Currow DC, Eagar K, Aoun S, Fildes D, Yates P and Kristjanson LJ. Is it feasible to collect voluntarily quality and outcome data nationally in palliative oncology care? J Clin Oncol 2008 26: 3853‐59.

152) Dunning T. Savage S, Duggan N. Martin P. Guidelines for managing diabetes at the end of life. Centre of Nursing and Allied Health Research, 2010 Geelong, ISBN 978‐0‐646‐53610‐1.

153) Kilvert A, Sinclair A, Rowles S. Association of British clinical diabetologists position statement: Diabetes and end of life care. Version 6 30.11.10. Practical Diabetes Internat 2011; 28: 26‐27.

154) Angelo M, Ruchalski C, Sproge JB. An approach to diabetes mellitus in hospice and palliative medicine. J Pall Med 2011; 14: 83‐87.

155) Wong KYC, Wong VW, Ho JT, Torpy DJ, McLean M, Cheung NW. High cortisol levels in hyperglycemic myocardial infarct patients signify stress hyperglycemia and predict subsequent normalization of glucose tolerance. Clin Endocrinol 2010; 72: 189‐95.

156) Krebs JD, Robinson GM, Smith RB, Toomath RJ. Follow up testing of hyperglycaemia during hospital admission: combined use of fasting plasma glucose and HbA1c NZ Med J 2000; 113: 379‐81.

157) Greci LS Kailasam M, Malkani S et al. Utility of HbA1c levels for diabetes case finding in hospitalized patients with hyperglycaemia. Diabetes Care 2003; 26: 1064‐8.

158) Gray CS, Scott JF, French JM, Alberti KG, O'Connell JE. Prevalence and prediction of unrecognized diabetes mellitus and impaired glucose tolerance following acute stroke. Age Ageing 2004; 33: 71.

159) George PM, Valabhjit J, Dawood M, Henry JA. Screening for type 2 diabetes in the accident and emergency department. Diabetic Med 2005; 22: 1766‐69.

160) Valentine NA, Alhawassi TM, Roberts GW, Vora PP, Stranks SN, Doogue MP. Detecting undiagnosed diabetes using glycated haemoglobin: an automated screening test in hospitalized patients. Med J Aust 2011; 194: 160‐164.

161) De Mulder M, Oemrawsingh RM, Stam F, Boersma E, Umans VA. Comparison of diagnostic criteria to detect undiagnosed diabetes in hyperglycaemic patients with acute coronary syndrome. Heart 2012; 98: 37‐41.

162) Olson L, Muchmore J, Lawrence CB. The benefits of inpatient diabetes care: improving quality of care and the bottom line. Endocrin Pract 2006; 12(Suppl 3): 35‐42.

163) Koprosky J, Pretto Z, Poretsky L. Effects of an intervention by a diabetes team in hospitalized patients with diabetes. Diabetes Care 1997; 20: 1553‐55.

164) Davies M, Dixon S, Currie CJ, Davis RE, Peters JR. Evaluation of a hospital diabetes specialist nursing service: a randomized controlled trial. Diabet Med 2001; 18: 301‐7.

165) Cavan DA, Hamilton P, Everett J, Kerr D. Reducing hospital stay for patients with diabetes. Diabet Med 2001; 18: 162‐4.

166) Sampson MJ, Crowle T, Dhatariya K et al. Trends in bed occupancy for inpatients with diabetes before and after the introduction of a diabetes inpatient specialist nurse service. Diabet Med 2006; 23: 1008‐ 15.

167) Newton CA, Young S. Financial implications of glycemic control: results of an inpatient diabetes management program. Endocrin Pract 2006; 12(Suppl 3): 43‐48.

168) Courtney M, Carey N, James J, Hills M, Roland J. An evaluation of a specialist nurse prescriber on diabetes in‐patient service delivery. Practical Diabetes Int 2007; 24: 69‐74.

169) Flanagan D, Moore E, Baker S, Wright D, Lynch P. Diabetes care in hospital – the impact of a dedicated inpatient care team. Diabet Med 2008; 25: 147‐51.

170) Brooks AP, Voon L, Chong JSW et al. The work of a dedicated inpatient diabetes care team in a district general hospital. Practical Diabetes Int 2011; 28: 70‐2.

171) NHS. National Diabetes Inpatient Audit 2010. Prontaprint 2010. Available from http://www.diabetes.nhs.uk/information_and_data/diabetes_audits/national_diabetes_inpatient_audit/.

172) Diabetes UK Task and Finish Group. Commissioning specialist diabetes services for adults with diabetes. Diabetes UK 2010. Available from http://www.diabetes.org.uk/Professionals/Publications‐reports‐and‐resources/Reports‐statistics‐and‐case‐studies/Reports/Commissioning‐Specialist‐Diabetes‐Services‐for‐ Adults‐with‐Diabetes‐‐‐Defining‐A‐Specialist‐Diabetes‐UK‐Task‐and‐Finish‐Group‐Report‐October‐ 2010/

173) Editorial. ‘High alert’ medications and patient safety. Int J Qual Health Care 2001; 13: 339‐40.

174) Clinical Excellence Commission (CEC) and NSW Department of Health 2010. Clinical incident management in the NSW public health system 2009. January‐June. Sydney.

175) Cheung NW, Cinnadaio N, O’Neill A et al. Implementation of a dedicated subcutaneous insulin prescription chart: Effect on glycaemic control. Diabetes Res Clin Pract 2011; 92: 337‐41.

176) The Credentialled Diabetes Educator in Australia: Role and scope of practice. Australian Diabetes Educators Association, 2007. Holder, ACT. ISBN 978 0 9750 790 4 1.