#3459 GLUCOSE-LOWERING THERAPIES IN END STAGE RENAL DISEASE: HOW CAN WE PREDICT THE RISK OF HYPOGLYCAEMIA?

Abstract

Abstract Background and Aims Diabetes is a common co-morbidity in patients with end stage renal disease (ESRD) receiving renal replacement therapy (RRT). Glucose-lowering therapies (GLT) such as sulphonylureas and insulin are commonly used treatments in diabetes but increase the risk of hypoglycaemia. Dialysis is also a risk factor for hypoglycaemia due to reduced drug clearance, decreased gluconeogenesis and glucose loss into the dialysate. Guidelines suggest an HbA1c target of 58 mmol/mol for patients with ESRD receiving glucose lowering therapies, given that tight HbA1c control has not been shown to improve cardiovascular risk, and HbA1c under 58mmol/mol will incur risk of hypoglycaemia in a potentially vulnerable population. We evaluated the clinical and demographic features associated with HbA1c <58 mmol/mol in an ESRD cohort aiming to define characteristics that predict the need for closer glycaemic monitoring. Method We performed a retrospective observational study of patients receiving RRT and treated with GLT attending a single centre nephrology unit using data collected from a clinically used electronic database. Patients with HbA1c <58 mmol/mol in the last 12 months were identified. Clinical, biochemical and demographic factors were compared between groups. Deprivation was measured using the Scottish Index of Multiple Deprivation (SIMD), whereby 1 represented the most deprived areas within Glasgow, while 5 represented the most affluent. Between group characteristics were analysed using Pearson's Chi-squared test or Wilcoxon test, and logistic regression was used to evaluate factors contributing to hypoglycaemia. Results We identified 607 patients, 51.6% of whom had an HbA1c below 58 mmol/mol. 298 patients were on haemodialysis, 14 on peritoneal dialysis and 291 had a kidney transplant. Patients with HbA1c<58 mmol/mol were older, (58±16.8 vs 60±15 years, p = 0.004) but there was no difference in duration of diabetes (9.1±10.2 vs 8.3±12.3 years, p = 0.42) or ESRD vintage (4.6±6.9 vs 4.1±7.5 years, p = 0.44). There was a higher proportion of type 2 diabetes in the group with HbA1c <58 mmol/mol(26 vs 36%, p = 0.04). There was no association between HbA1c and SIMD (p = 0.46) nor gender (p = 0.55). In logistic regression, increasing age (OR 1.19, 95% CI 1.03, 1.38) and ESRD vintage (OR 1.03, 95% CI 1.01, 1.06) increase the likelihood of tight glycaemic control. Transplantation reduced the risk (OR 0.64, 95% CI 0.44, 0.93). There was no association between type of GLT and risk of tight glycaemic control. Subgroup analyses of the haemodialysis and transplant populations was also performed with broadly similar findings. In patients on haemodialysis, increasing age and ESRD increased the risk of HbA1c <58 mmol/mol. In patients with a kidney transplant, there was no association between current estimated glomerular filtration rate and HbA1c <58 mmol/mol. Conclusion Tight glycaemic control is very common in patients with ESRD on GLT. We found signals that patients with type 2 diabetes, receiving haemodialysis, increasing age, and ESRD vintage, increased risk of tight glycaemic control and hypoglycaemia. Transplantation was a substantial protective factor. In patients with a kidney transplant, no trend was found in eGFR and HbA1c. This is a huge concern and suggests HbA1c checks need to be included as part of regular renal profile monitoring every 3 months. More regular follow-up with the Diabetes team may also be useful in this patient group. Continuous glucose monitoring (CGM) devices have revolutionised diabetes care by allowing easy access to data such as time in range and trends in blood sugar throughout the day, thus making it easier to optimise treatment. These devices also provide the patient with a warning when they are approaching hypoglycaemia. This study provides an argument that existing indications for CGM do not match risk factors for tight glycaemic control in ESRD. Hypoglycaemia is difficult to predict in ESRD patients and access to CGM should be widened.