Using routine HbA1c measurements in stroke and the associations of dysglycaemia with stroke outcomes
Introduction
Diabetes mellitus is increasingly prevalent worldwide.1 In Australia, it has been estimated that up to 3 million Australians will have diabetes by 2025.2 Diabetes is a major risk factor for stroke. In fact, a study estimated that one-third of all acute stroke inpatients may have diabetes3 with another study reporting that as high as 20% of patients with acute ischaemic stroke had previously unrecognised diabetes mellitus.4 Given the high rates of diabetes in stroke inpatients, routine HbA1c testing in all inpatients with a stroke admission may have a role to detect hyperglycaemia.
Currently, fasting blood glucose remains the most commonly measured glycaemic parameter to diagnose diabetes in stroke inpatients.5 However, an advantage of measuring HbA1c is that it provides an average measure of glycaemia for the previous 90 days, thus reducing the potential for misdiagnosis based on acute or stress hyperglycaemia within the acute hospital setting.6 It also provides added ease of measurement as patient fasting is not required and it can be used in those on glucocorticoid therapy.7 Both the International Expert Committee and the American Diabetes Association (ADA) have recommended HbA1c as the diagnostic test of choice for diabetes mellitus using a threshold of 6.5% (48 mmol/mol) as it represents an inflection point for retinopathy prevalence, similar to the thresholds used for fasting blood glucose and 2-hour blood glucose.8,9 The ADA also recommend its use for pre-diabetes identification using the range of 5.7–6.4% (39–46 mmol/mol) as prior studies have shown that it is related to increased risk of developing diabetes. However, studies examining the utility of routine HbA1c measurements to guide therapy within the stroke setting are limited.
There is evidence that having co-morbid diabetes mellitus is associated with poorer stroke outcomes with one study concluding that HbA1c >7.2% (55 mmol/mol) is an independent risk factor for one year all-cause mortality after stroke,10 with other studies also reporting that diabetes was associated with increased post-stroke length of stay and readmission rates.11,12 In contrast, some studies have reported no significant association of diabetes with these outcomes.13,14 The discrepancies may relate to differences in diabetes diagnosis and study methodology. For example, studies have used different methods, such as past history,11 ICD-coding,12 fasting plasma glucose,15 and HbA1c16 to diagnose diabetes. Furthermore, the effects of pre-diabetes on stroke outcomes have not been widely investigated in the literature, with one study reporting that diabetes but not pre-diabetes was associated with post-stroke disability.16
The relationship between HbA1c and stroke outcomes is less investigated. Given that HbA1c is a reflection of glycaemia in the preceding three months, investigating it as a continuous variable may provide information that trichotomising into diabetes categories do not. Some previous studies indicate a positive association between HbA1c and ischaemic stroke recurrence, stroke mortality and severity.17., 18., 19. Our study explored this relationship further in both ischaemic and haemorrhagic stroke patients.
Routine HbA1c measurements, especially facilitated through the use of clinical information systems, to aid clinical practice in stroke inpatients is not current practice in Australia. The current study aims to use routine HbA1c measurements, as part of the Austin Health Diabetes Discovery Initiative, to estimate the prevalence of diabetes and pre-diabetes within the inpatient stroke cohort. We also aim to explore the association of glycaemic status, as reflected by diabetes status or inpatient HbA1c level, with stroke outcomes. We hypothesise that poor glycaemic control (as reflected by increasing HbA1c or the presence of pre-diabetes or diabetes) will be associated with poorer stroke outcomes.
Section snippets
Participants
In this prospective observational cohort study, all patients aged 54 years and over and admitted to Austin Health with a primary diagnosis of ischaemic or haemorrhagic stroke during the study period (17 July 2013 to 1 December 2015) that were captured by the Australian Stroke Clinical Registry20 were studied. Patients with multiple admissions during the study period were assessed on the basis of the first admission.
The age cut-off of 54 years or older was implemented because a previous study21
Prevalence of diabetes and pre-diabetes
610 patients aged 54 or above were admitted to Austin Health during the 30-month study period with a primary diagnosis of haemorrhagic or ischaemic stroke and were included in this prospective observational study. Of these, 31% (95% CI: 27.3%–34.8%) (n = 189) had diabetes with 4.9% (95% CI: 3.3%–6.9%) (n = 30) of the cohort having previously unrecognised diabetes. 40% (95% CI: 36.2%–44.2%) (n = 245) had pre-diabetes and 29% (95% CI: 25.3%–32.6%) (n = 176) had no diabetes (Fig. 1).
Baseline characteristics
The baseline
Key findings
The current study for the first time demonstrates the feasibility of using clinical information systems to generate automated HbA1c measurements for all stroke inpatients aged 54 or over to investigate the burden of diabetes and pre-diabetes within the stroke cohort. Among the 610 patients included in the study, 31% (n = 189) had diabetes while 40% (n = 245) had pre-diabetes. After adjusting for age, stroke type, atrial fibrillation status and admission ambulation status (as a measure of stroke
Acknowledgements
The authors would like to acknowledge the Australian Stroke Clinical Registry and Clinical Informatics department of Austin Health for their data contribution to the study.
We are also grateful to Jennifer White for assistance with decision support programs in the Electronic Health Records projects.
Funding
EIE was supported by a National Stroke Foundation Small Project Grant, a Viertel Clinical Investigatorship, Sir Edward Weary Dunlop Medical Research Foundation grant and RACP fellowship.
References (29)
- et al.
Diabetes atlas: global estimates of the prevalence of diabetes for 2010 and 2030
Diabetes Res Clin Pract
(2010) - et al.
Projecting the burden of diabetes in Australia – what is the size of the matter?
Aust N Z J Pub Health
(2009) - et al.
Predictors for readmission of acute ischemic stroke in Taiwan
J Formos Med Assoc
(2011) - et al.
Diabetic and non-diabetic subjects with ischemic stroke: differences, subtype distribution and outcome
Nutr Metab Cardiovasc Dis
(2008) - et al.
Impact of diabetes mellitus (DM) on the health-care utilization and clinical outcomes of patients with stroke in Singapore
Value Health
(2009) - et al.
Association between hemoglobin A1C levels and clinical outcome in ischemic stroke patients with or without diabetes
J Clin Neurosci
(2015) - et al.
New ICD-10 version of the Charlson comorbidity index predicted in-hospital mortality
J Clin Epidemiol
(2004) - et al.
Glycosylated hemoglobin for diagnosis of prediabetes in acute ischemic stroke patients
J Stroke Cerebrovasc Dis
(2013) - et al.
Prevalence and prediction of unrecognised diabetes mellitus and impaired glucose tolerance following acute stroke
Age Ageing
(2004) - et al.
Frequency of newly diagnosed diabetes mellitus in acute ischaemic stroke patients
J Coll Physicians Surg Pak
(2012)
Guidelines for the prevention of stroke in patients with stroke or transient ischemic attack: a guideline for healthcare professionals from the American Heart Association/American Stroke Association
Stroke
The role of HbA1c in the diagnosis of diabetes mellitus in Australia
Med J Aust
Inpatient HbA1c testing: a prospective observational study
BMJ Open Diabetes Res Care
International Expert Committee report on the role of the A1C assay in the diagnosis of diabetes
Diabetes Care
Cited by (3)
Feasibility trial of metformin XR in people with pre-diabetes and stroke (MIPPS)-randomised open blinded endpoint controlled trial
2021, Journal of Clinical NeuroscienceCitation Excerpt :Moreover, there is an increasing prevalence of pre-diabetes, which is a precursor for developing type 2 diabetes [5]. We previously showed that 40% of people admitted to a stroke unit above the age of 54 had pre-diabetes [4]. Treating pre-diabetes in people with stroke could prevent the development of type 2 diabetes and its complications, including further strokes.
Diabetes in ischaemic stroke in a regional Australian hospital: uncharted territory
2022, Internal Medicine JournalHbA1c and clinical outcomes after endovascular treatment in patients with posterior circulation large vessel occlusion: a subgroup analysis of a nationwide registry (BASILAR)
2020, Therapeutic Advances in Neurological Disorders
Disclosures/Declarations of interest: None.