The relationship between glycaemic variability and cardiovascular autonomic dysfunction in patients with type 1 diabetes : a systematic review

Rigorous glycaemic control-reflected by low HbA1c goals-is of the utmost importance in the prevention and management of complications in patients with type 1 diabetes mellitus (T1DM). However, previous studies suggested that short-term glycaemic variability (GV) is also important to consider as excessive glucose fluctuations may have an additional impact on the development of diabetic complications. The potential relationship between GV and the risk of cardiovascular autonomic neuropathy (CAN), a clinical expression of cardiovascular autonomic dysfunction, is of increasing interest. This systematic review aimed to summarize existing evidence concerning the relationship between GV and cardiovascular autonomic dysfunction in T1DM. An electronic database search of Medline (PubMed), Web of Science and Embase was performed up to October 2019. There were no limits concerning year of publication. Methodological quality was evaluated using the Newcastle Ottawa Scale for observational studies. Six studies (four cross-sectional and two prospective cohorts) were included. Methodological quality of the studies varied from level C to A2. Two studies examined the association between GV and heart rate variability (HRV), and both found significant negative correlations. Regarding cardiovascular autonomic reflex tests (CARTs), two studies did not, while two other studies did find significant associations between GV parameters and CART scores. However, associations were attenuated after adjusting for covariates such as HbA1c, age and disease duration. In conclusion, this systematic review found some preliminary evidence supporting an association between GV and cardiovascular autonomic dysfunction in T1DM. Hence, uncertainty remains whether high GV can independently contribute to the onset or progression of CAN. The heterogeneity in the methodological approach made it difficult to compare different studies. Future studies should therefore use uniformly evaluated continuous glucose monitoring-derived parameters of GV, while standardized assessment of HRV, CARTs and other potential cardiac autonomic function parameters is needed for an unambiguous definition of CAN