We were interested in the recent article in Diabetes Care from Tang et al. (1), in which they reported that lower HbA1c in subjects with highly carbamylated albumin (C-Alb) was not clearly related to glucose levels, precluding a relationship of HbA1c with the later progression of chronic kidney disease.
Their findings accord well with the in vitro description of glycation and carbamylation competing for the modification of proteins (2). Despite carbamylated hemoglobin not being directly measured, we can hypothesize that C-Alb reflected carbamylated hemoglobin in the participants of the Chronic Renal Insufficiency Cohort (CRIC), which reduced their HbA1c and probably influenced the anemia of some subjects (3).
The carbamylation of protein and the anemia that develops with the progression of diabetic kidney disease (DKD) can therefore lead to HbA1c measurements underestimating the true glucose exposure, contributing to its poor reliability as an indicator of glucose control in end-stage renal disease (4). The underestimation may also explain why glucose control fails to slow the decline of renal function despite its favorable effect on albuminuria (5).
As discussed by Tang et al. (1), there is no simple alternative to the measurement of HbA1c in DKD. Other circulating proteins are also altered by carbamylation. However, we can expect that carbamylation occurs preferentially at late stages of DKD due to the retention of urea, and DKD stages 2, 3A, 3B, and 4 were represented in the CRIC cohort. The authors can probably help clinicians by providing a simple table, similar to Table 1 in their article, in which they replace the quartiles of HbA1c by stages of DKD and report C-Alb, HbA1c, and their relation to random glucose and later progression of DKD. Then we would know whether HbA1c is biased by carbamylation only at stage 4 or sooner than that.
See accompanying article, p. e116.
Article Information
Duality of Interest. No potential conflicts of interest relevant to this article were reported.
