QJM Advance Access originally published online on January 7, 2008
QJM 2008 101(2):87-90; doi:10.1093/qjmed/hcm128
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Immediate in-patient management of hyperglycaemia—confusion rather than consensus?
From the 1Bournemouth Diabetes and Endocrine Centre, Royal Bournemouth Hospital NHS Foundation Trust and 2Centre of Postgraduate Medical Research and Education, Bournemouth University, Bournemouth, Bh7 7DW, UK
Address correspondence to Dr D. Kerr, Bournemouth Diabetes and Endocrine Centre, Royal Bournemouth Hospital NHS Foundation Trust, Castle Lane East, Bournemouth, Bh7 7DW, UK. email: david.kerr{at}rbch.nhs.uk
Received 13 July 2007 and in revised form 18 September 2007
 |
Summary |
|---|
 Top Summary IntroductionMethodsResultsDiscussionReferences |
|---|
Background: In-patients with high blood glucose levels have much greater mortality
and morbidity rates compared to normoglycaemic individuals hospitalized with the same condition.
Aim: To examine prospectively the glucose-lowering treatments used for patients admitted as acute medical emergencies with admission hyperglycaemia (11–17 mmol/l) under the care of non-diabetes specialist teams. Individuals with acute diabetes emergencies (e.g. diabetic ketoacidosis or HONK or glucose levels >17 mmol/l) were excluded.
Methods: Patients’ notes were examined as they were admitted without any interventions from the diabetes team. Choice of treatment for their hyperglycaemia was noted and the average blood glucose level was calculated each day of admission for the first 5 days based on bedside fingerstick glucose measurements.
Results: Seventy-three in-patients [37 men, average (SD) age 74.1(12) years] with hyperglycaemia [average 13.7(1.6) mmol/l] on admission were included. Fourteen were not known to have diabetes, three had type 1 and 56 type 2 diabetes. Glycaemic control was suboptimal and achieved values were unrelated to the mode of delivery of glucose-lowering therapies. Length of stay and death rates in hospital were greatest in the group of patients who were not previously known to have diabetes.
Conclusion: Untreated or under-treated hyperglycaemia was a common occurrence in patients admitted to hospital with an acute medical emergency. There may be a role for hospital-based specialist diabetes teams to take a lead in facilitating more acceptable glucose control to achieve standard 8 of the National Service Framework for Diabetes.
|
Introduction |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Hyperglycaemia is common in patients admitted to hospital.1 Although there is increasing evidence that the level of glycaemia in acutely unwell individuals has important implications for morbidity and mortality,2 there remains a controversy about the benefits of lowering glucose levels outside the critical care setting.3 Nevertheless in 2006, the American Diabetes Association published guidelines on the ideal levels of glycaemia for hospitalized patients suggesting a ‘target’ value of 6.1 mmol/l in critically ill patients and 5.0–7.2 mmol/l before meals in non-critically ill individuals.4
Standard 8 of the UK National Service Framework for Diabetes has emphasized the importance of improving care for diabetic individuals who are hospitalized5 and many hospitals now employ specialist nurses specifically to manage in-patient diabetes care based on published evidence.6 Unfortunately, despite these efforts health professionals dealing with acute medical emergencies, continue to be reluctant to intervene in non-diabetic individuals with admission hyperglycaemia.7
The aim of this study was to examine the glucose-lowering treatments used for patients admitted as acute medical emergencies with hyperglycaemia on admission and their level of glycaemic control. Individuals with acute admissions related to severe hyperglycaemia (e.g. diabetic ketoacidosis or HONK or glucose levels >17 mmol/l) were excluded.
|
Methods |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Patients’ notes were examined as they were admitted via the Clinical Decisions Unit (CDU) without any interventions from the diabetes team. Patients with blood glucose levels between 11 and 17 mmol/l on admission were included and baseline characteristics (age, sex, reason for admission, diabetes history) and their usual medication were recorded. The purpose of the CDU is to admit and assess all acutely unwell patients referred in by GP or brought in by ambulance. Once assessed and any initial tests carried out a decision is made about whether the patient should be transferred to a ward for continued monitoring and care or to be discharged with or without follow-up. The length of stay in the CDU varies between patients but the aim is to move the patient within 24 h. Each patient was traced for the duration of their stay in hospital from their point of admission to CDU to discharge. Choice of treatment for their hyperglycaemia was noted and the average blood glucose level was calculated each day for the first 5 days. These were based on bedside fingerstick glucose measurements taken before meals and before bed unless they were treated with intravenous (IV) insulin, when they would have had their blood glucose levels measured one to two hourly. All patients were managed by their own specialist team (non-diabetic), but the team maintained the option of requesting assistance from a ward-based diabetes specialist nurse.6
|
Results |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Over 3 months, 73 patients were admitted to the CDU [37 men, average(SD) age 74.1(120 years and glucose level on admission 13.7(1.6) mmol/l]. The most common reason for admission was for exclusion of acute coronary syndrome (20%) with a smaller number admitted with acute exacerbations of chronic obstructive pulmonary disease or broncho-pulmonary infections. Of the 73 patients, 14 (19%) were not known previously to have diabetes, three patients had type 1 and 56 patients had type 2 diabetes. None were nil by mouth or unable to swallow.
Characteristics of the diabetic population were obtained on 54 patients, showing that they had had diabetes for a mean (SD) duration of 11.0 years (8.4). It was possible to find results on 50 diabetic patients concerning their recent glycaemic control. Mean (SD) HbA1c for the patients with type 1 diabetes was 9.9% (3.3) compared with 9.1% (1.8) for patients with type 2 diabetes. While 34% of patients were classed as obese with a BMI > 30, another 34% of the sample were classed as overweight with a BMI between 25 and 30. Comorbidities are shown below (Table 1).
|
In the CDU, 14 patients not previously known to have diabetes received no glucose-lowering treatment or subsequent monitoring of their blood glucose levels. For the three individuals with type 1 diabetes, one patient was immediately commenced on continuous intravenous insulin, one received their usual subcutaneous insulin and one had no treatment documented.
Of the 56 patients with type 2 diabetes, 26 (47%) received no additional treatment aimed at acutely lowering blood glucose levels. Six were immediately put onto a continuous intravenous insulin sliding scale, others receiving tablets, insulin or diet alone.
After assessment in the CDU, patients were transferred to the speciality ward dependent on their initial presentation. On these wards, 1 of the 14 non-diabetic patients with hyperglycaemia was started immediately on a continuous intravenous insulin sliding scale and the other 13 were left untreated without further measurements of fingerstick glucose levels. All type 1 patients were treated subsequently with subcutaneous (SC) insulin using multiple daily injections. Thirteen (23%) of the patients with type 2 diabetes had no further additional treatment for hyperglycaemia and only three of these had further checks of fingerstick blood glucose levels. Of the other patients with type 2 diabetes, nine were commenced on continuous intravenous insulin sliding scale infusions. One patient was treated with diet, 14 (25%) with tablets, 15 (27%) with SC insulin and 4 (7%) with tablets and SC insulin.
Figure 1 shows a comparison of the achieved blood glucose levels for each treatment. Table 2 shows the number of patients who received each treatment on days 1–5 of their hospital stay, thereby allowing us to see that the numbers receiving each treatment stay fairly consistent. This is not the case for intravenous insulin as the patient numbers decrease over the 5-day period. This is explained in more detail later.
|
|
The trends of the mean blood glucose levels are shown graphically in Figure 1.
Average length of stay was 11 days for patients with known diabetes and 13 days for individuals not previously known to have diabetes. Of the four patients that died three had new onset hyperglycaemia. The rates of discharge back to own home were also compared, 50 (85%) of known diabetics were discharged home compared with 9 (64%) from the group with new onset hyperglycaemia.
|
Discussion |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Hyperglycaemia in the hospital is an independent prognostic marker for medical admissions irrespective of diabetic status.1,8 In the critical care setting effective treatment of elevated blood glucose levels reduces morbidity and mortality.9 Outside of the critical care unit, there is minimal literature available about treating patients. Our study highlights possible lack of sufficient attention and discrepancies by healthcare professionals in approaching the problem of hyperglycaemia. Treatment options were erratic, not always appropriate, for example patients were treated with IV insulin when eating and drinking, and patients were often under-treated on the wards. As noted by others, it was striking that junior medical staff appeared to be reluctant to initiate insulin in patients in whom diabetes had not previously been diagnosed.7
Control of hyperglycaemia outside the critical care environment can be difficult to achieve due to a number of factors e.g. stress, concurrent medication, reduced activity, disease processes, poor nutrition and the timing of insulin administration.10 Nevertheless, we found a lack of awareness about the importance of treating in-patient hyperglycaemia and a lack of consensus to the management options by junior medical staff. This piece of work was carried out as an audit of treatment and therefore does not contain a control group with which to compare the results but this may be something to consider for the future.
We found that even when treatment was initiated for hyperglycaemia achieved control was sub-optimal. Blood tests for the IV insulin show consistent blood glucose levels over the 5-day period and the number of patients receiving the treatment decreases which is expected as it is a treatment used mostly during acute illness. The blood glucose levels achieved for patients receiving multiple daily injections are higher than expected. Looking at the individual drug charts it was clear that the insulin doses were not prescribed or adjusted as necessary, allowing the blood glucose levels to remain higher than they should be. In our experience we have found that inpatient hyperglycaemia can usually be effectively managed without the use of intravenous insulin when patients are eating and drinking. Here, intravenous insulin was prescribed even though patients were able to eat and drink despite the fact that insulin infusion protocols do not make adjustments for carbohydrate ingestion with meals.
The length of stay and mortality were greatest in non-diabetic patients with new onset hyperglycaemia supporting previous evidence.2,11 Others have also found that patients with hyperglycaemia, who do not have a history of diabetes, receive less attention to glucose control and have higher mortality than previously diagnosed patients.1,12 Our work shows this to be the case for this sample population.
In conclusion, despite growing evidence that glucose levels in acutely ill patients have clinical importance and that patients may benefit from glucose lowering, the care of such individuals by non-specialist teams appears to be suboptimal and random. The reasons for this are unclear but may reflect a lack of knowledge and also a fear of inducing hypoglycaemia. However, there is also a lack of good quality trial data on the benefits or otherwise of acutely lowering blood glucose levels outside of the critical care situation. Despite the growing use of ward-based specialist diabetes nurses there still remains room for improvement. One option may be for hospital-based specialist diabetes teams to take a more active role in the day-to-day management of acutely ill patients admitted with hyperglycaemia under the care of other specialities.
Conflict of interest: None declared.
|
References |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
1. Umpierrez GE, Isaacs SD, Bazargan N, You X, Thaler LM, Kitabachi AE. Hyperglycaemia: an independent marker of in-hospital Mortality in patients with undiagnosed diabetes. J Clin Endocrinol Metabol (2002) 87:978–82.
2. Van Den Berghe G, Wouters P, Weekers F, Verwaest C, Bruyninckx F, Schetz M, et al. Intensive insulin therapy in critically ill patients. New Engl J Med (2001) 345:1359–67.
3. Watkinson P, Barber VS, Young JD. Strict glucose control in the critically ill. Br Med J (2006) 332:865–6.
4. American Diabetes Association. Standards of Medical Care in Diabetes. In: Diabetes Care 2006. (29(Suppl)):S4–S42.
5. National service framework for diabetes, Standard 8. Care of people with diabetes admitted to hospital. In: Department of Health (2001).
6. Cavan DA, Hamilton P, Everett J, Kerr D. Reducing hospital inpatient length of stay for patients with diabetes. Diab Med (2001) 18:162–4.[CrossRef][Web of Science][Medline]
7. Al-Bermani A, Desha YH, Morgan J, Soobrah R, Symonds CS, Taylor R. Management of incidental hyperglycaemia in acute medical emergencies. Diab Med (2004) 22:937–41.[CrossRef][Web of Science]
8. Weir CJ, Murray GD, Dyker AG, Lees KR. Is hyperglycaemia an independent predictor of poor outcome after acute stroke? Results of a long term follow up study. Br Med J (1997) 314:1303–6.
9. Langouche L, Vanhorebeek I, Van den Berghe G. Therapy insight; the effect of tight glycaemic control in acute illness. Endocrinol Metab (2007) 3:270–7.
10. Penfold S, Hamilton P, Richardson T, Cavan D, Kerr D. Patient mealtime medications. Diab Med (2006) 23(Suppl. 2):72.[CrossRef][Web of Science][Medline]
11. Timmer JR, van der Horst IC, Ottervanger JP, Henriques JP, Hoorntje JC, de Boer MJ, et al. Prognostic value of admission glucose in non-diabetic patients with myocardial infarction. Am Heart J (2004) 148:399–404.[CrossRef][Web of Science][Medline]
12. Levetan CS, Passaro M, Jablonski K, Kass M, Ratner RE. Unrecognised diabetes among hospitalised patients. Diabetes Care (1998) 21:246–9.[Abstract]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||