The future of self-monitored blood glucose: Mean blood glucose versus glycosylated hemoglobin

There is a growing interest in re-evaluating the biological and clinical meaning of glycosylated hemoglobin (HbA(1c)) in light of recent research that suggests an "accurate conversion algorithm" that would result in an "A1c-derived" average glucose may be possible, resulting in laboratories reporting HbA(1c) accompanied by its mean blood glucose range. This review addresses the association between HbA(1c) and mean blood glucose and their relationship to diurnal glucose patterns. While researchers agree that there is a close correlation. between mean blood glucose levels measured in laboratory, by self-monitored blood glucose (SMBG) or by continuous glucose monitoring (CGM), it remains ambiguous as to how these summary measures could be used to detect glucose variation. A re-examination of research data collected over the past 4 years in which subjects simultaneously employed SMBG and CGM and for whom the corresponding HbA(1c) was measured shows that HbA(1c) and both CGM and SMBG means are closely correlated: R-2 = 0.71 and 0.69, respectively. However, glucose profiles constructed from either SMBG or CGM showed that the same HbA(1c), at virtually every percentile, produced altogether different and unpredictable diurnal patterns. Thus, relying solely on, HbA(1c) or a "derived" mean glucose maybe misleading, providing "false impressions" of therapeutic efficacy. Consequently, because self-monitoring provides a unique vantage point from which to view diurnal glucose patterns it remains a potent component of the armamentarium of diabetes management.