Glycaemic variation in insulin treated diabetic patients with end stage renal disease on maintenance haemodialysis and its effect on cardiac electrical activity

Abstract

Diabetic kidney disease remains the single most common cause of renal failure in UK, accounting for 26.9% of patients needing renal replacement therapy (UK renal registry, 2016). Mortality rates on RRT are worse for the diabetes population compared to the non-diabetic population. Diabetic patients on maintenance haemodialysis experience huge variation in their glycaemia, which is not well understood to guide appropriate therapy. ESRD patients are at higher risk of sudden cardiac death and arrhythmia is suspected to be a major cause. However there is no established guideline in detecting at risk patients for preventative therapy. We aimed to study the glycaemic variation in patients with ESRD on maintenance HD using continuous glucose monitoring for longer periods in order to help understand the variation in relation to dialysis and associated change in cardiac electrical conductivity simultaneously to explore any relation with glycaemia. In a pilot study we studied glucose variation and cardiac electrical activity using CGM and Holter monitor respectively during 37 weeks in 15 diabetic patients and 5 weeks in 5 nondiabetic subjects. Diabetic subjects had a significant variation in their glycaemia through the week. There was a significant drop in the interstitial glucose level during HD, followed by a rise in the post-HD period (preHD vs HD vs postHD: 11.4±5.1 vs 8.4±3.6 vs 11.5 ± 4.6mmol/l). There was a significant change in QTc interval from start to end of HD in this population (468 ± 42 vs 481 ± 36 vs 495 ± 49). Short but frequent episodes of arrhythmia were noted throughout the week. All diabetic patients who were prone for arrhythmias had abnormal QTc. Non-diabetic patients also experienced significant variation in IG levels and were noted to have IG in both the hypo and hyperglycaemic range. CGM helps in understanding the glycaemic variation in this population and real time recording would help in reducing the episodes of hypoglycaemia and hyperglycaemia. There is no relation between glycaemic variation or hypoglycaemia and change in QTc interval or cardiac dysrhythmias, which remain common in this population. Asymptomatic dysrhythmic episodes put these patients at risk of sudden cardiac death. The data suggest that baseline ECG and/or periodic Holter monitoring should be used in clinical care.