Assessing glycemic variation: why, when and how?

In the post-Diabetes Control and Complications Trial (DCCT) and Epidemiology of Diabetes Interventions and Complications (EDIC) era of type 1 diabetes mellitus (T1DM) care, glycosylated hemoglobin (A1C) has enjoyed primacy as the clinical outcome variable (1). Metabolic control as defined by A1C, however, only defines approximately 25% of the risk of subsequent microvascular pathology (2) and, hence, other glycemic outcome variables are also being canvassed as being of potential significance. Transcription-regulating actions of glucose and the phenomenon of "metabolic memory" have recently become recognized (3,4). Simultaneously, ambulant continuous glucose monitoring (CGM) technologies have become available. The convergence of these factors has increased the interest in the impacts of fluctuations in glycemia, otherwise known as glycemic variation (GV). Initially, this interest was focused upon the effects of post-prandial glycemic excursions (5), but more recently, associations of GV and oxidative stress, microvascular pathology (6), and GV prediction associated with closedloop insulin delivery (7) have evolved. Notwithstanding this emerging interest in GV, there still remains a lack of consensus as to the importance of GV, in what circumstances it can be measured, and what GV metrics are best suited for various purposes. The aim of this review is to discuss these 3 key areas: Why measure GV? When can GV be meaningfully assessed?; How to measure to GV?.