Treatment and Detection of Diabetes Mellitus in Patients with Stroke

Abstract

Introduction The top ten leading causes of death in 2016 were stroke and diabetes according to a World Health Organization report. Diabetes is a well-established independent but modifiable risk factor for both ischemic and haemorrhagic stroke. Stroke outcomes tend to be worse among patients with diabetes, which contributes to increased mortality and disability. Most of the major trials of antidiabetic drugs have focused on establishing glucose-lowering properties, whereas assessment of the effect of these drugs on stroke outcomes has been limited. It could be of great value to determine if there exists an antidiabetic agent has superior protective effect against recurrent stroke. HbA1c level is a necessary test in order to confirm the diagnosis of diabetes. Recent research found that it has been associated with stroke and its further prognosis. Guidelines do not specify target HbA1c levels in people with diabetes and stroke. The incidence of hypoglycaemia is high in general medical in-patients and variably in blood glucose may link with outcome. It is unknown if this is true after a patient suffers a stroke. This research aims to discover the effect of antidiabetic interventions after stroke, to explore whether HbA1c levels are related to the recurrence of stroke and mortality rate, to explore whether testing for HbA1c levels is the best method to confirm diabetes in stroke patients, and to explore if hypoglycaemia, hyperglycaemia and variability in blood glucose levels are associated with a poor outcome during admission for patients who have suffered a stroke. Methods The first chapter covers the background of diabetes and stroke, followed by discussion of limited number of studies that have observed the link between antidiabetic treatments and clinical outcome after stroke. The second chapter contains an analysis of retrospective clinical data for people admitted to hospitals in Glasgow with stroke who had diabetes mellitus using data from the Scottish Stroke Care Audit (SSCA) and the SCI-Diabetes dataset. Data was used to examine the relationship episodes of hypoglycaemia, hyperglycaemia and variability in blood glucose levels and outcome in hospitalised patients with diabetes and acute stroke. In the third chapter, a III web-based survey is discussed, which was distributed to UK stroke physicians through the British Association of Stroke Physicians, the Scottish Stroke Research Network, the UK Clinical Research Network and the Welsh Stroke Research Group to assess current UK practice for detection and treatment of diabetes in patients with stroke. In the fourth chapter, a pilot study is discussed, which entails using HbA1c testing within 14 days on patients with acute stroke and with no history of Diabetes Mellitus (DM). HbA1c test was then repeated and an Oral Glucose Tolerance Test (OGTT) was performed three months after the stroke to assess the incidence of diabetes in stroke patients and compare the findings of HbA1c result to an OGTT and standard care. The fifth chapter includes a retrospective study using the National Health Service Scotland Glasgow and Clyde SAFE HAVEN database. Demographic details, HbA1c values, medication, and diagnostic data using ICD-10 codes were extracted and analysed to define the HbA1c profile level and the antidiabetic drug treatments used in Scotland for stroke survivors with diabetes. Furthermore, this data was used to determine if there is a relationship between attained HbA1c or baseline HbA1c levels and antidiabetic drug treatments for recurrent stroke and occurrence of acute coronary syndrome. Finally, the data was used to determine if there is an association between antidiabetic drug treatments and HbA1c levels. Cox and linear regression models were performed. Results Chapter 2. Hypoglycaemia, hyperglycaemia and Capillary Blood Glucose va