Understanding optimal methods for monitoring glycaemia in people with type 2 diabetes in low resource settings.

Monitoring glycaemic control in type 2 diabetes is essential to allow appropriate titration of medication and prevent diabetes complications. In developed countries glucose control is monitored mainly by glycated haemoglobin (HbA1c) testing or intensive home capillary glucose measurements. HbA1c is not used routinely in low resource settings because of cost, and a number of conditions that are relatively common in this population may result in HbA1c results that poorly reflect blood glucose levels; for example sickle cell and other haemo-globinopathies; anaemia, malaria, renal disease. The alternative measures to HbA1c recommended for monitoring (i.e., fructosamine and glycated albumin, or single glucose measures) in situations where HbA1c may be unreliable have not been well studied in African populations. Current clinical practice in such settings varies, with a single fasting glucose measure used by many clinicians to inform treatment titration, but others routinely use non fasting ‘random’ measurements. A key question for use of fasting glucose in monitoring glycaemic burden is whether it is significantly affected by exercise (prolonged walking to the clinic). This is because majority of the patients in Uganda and other low resource settings (e.g., SSA countries) walk long distances to the diabetes clinics and fasting/non-fasting blood glucose will often be measured after an abnormally prolonged fast (of more than the recommended 8 hours) and/or very long walk to the clinic. On the other hand, a key barrier to therapy intensification in the management of type 2 diabetes is fear of hypoglycaemia. As intensive glucose monitoring is not possible in low resource settings, often sulphonylurea and insulin glucose lowering therapy (two of the 3 therapy classes widely available) are only started and maintained at glycaemic thresholds far higher than recommended elsewhere (therapeutic inertia) because of the fear of hypoglycaemia. It is not clear whether this fear of hypoglycaemia is justified. Little is known about hypoglycaemia in the patients receiving these treatments in sub-Saharan Africa, with only a small number of retrospective studies that have not used objective measurements. The aim of the thesis is to determine the optimal method for monitoring glycaemic burden and impact of exercise on fasting glucose and to understand the rates and determinants of hypoglycaemia with sulphonylurea and insulin treatment. In Chapter 1, we review the current literature for monitoring glycaemic control in the clinical management of diabetes In Chapter 2 we compare the performance of three tests (HbA1c, fasting and non-fasting/random glucose) commonly used for monitoring glycaemic burden in type 2 diabetes patients. We show that HbA1c is the overall best measure of glycaemic burden, despite high prevalence of other medical conditions that may affect its accuracy (e.g. anaemia, haemoglobinopathies). We also demonstrate that fasting plasma glucose and random plasma glucose strongly correlate with CGM glucose and HbA1c, have reasonable sensitivity and specificity to detect poor glycaemic control and the difference in performance between these tests is modest. In Chapter 3 we assess the performance of glycated albumin and fructosamine against continuous glucose monitoring in comparison to other measures as an assessment of glucose burden in participants with type 2 diabetes and determine whether a recently developed automated glycated albumin assay can improve performance over and above fructosamine. In chapter 4 we assess the impact of prolonged walking on fasting glucose in type 2 diabetes patients. We demonstrate that fasting plasma glucose is not significantly affected by walking to the clinic. In Chapter 5 we assess the rates and determinants of continuous glucose monitoring measured hypoglycaemia in patients receiving insulin or sulfonylurea treatment in the Ugandan population, in comparison to those receiving metformin or diet treatment. We show that in a low-resource sub-Saharan African setting, hypoglycaemia is infrequent among people with type 2 diabetes receiving sulphonylurea treatment, and the modest excess occurs predominantly in those with tight glycaemic control