Effect of chronic kidney disease on A1C in individuals being screened for diabetes

The agreement between fasting glucose and markers of chronic glycaemic exposure in individuals with and without chronic kidney disease: a cross-sectional study

Background

To assess whether the agreement between fasting glucose and glycated proteins is affected by chronic kidney disease (CKD) in a community-based sample of 1621 mixed-ancestry South Africans.

Methods

CKD was defined as an estimated glomerular filtration rate < 60 ml/min/1.73 m². Fasting plasma glucose and haemoglobin A1c (HbA1c) concentrations were measured by enzymatic hexokinase method and high-performance liquid chromatography, respectively, with fructosamine and glycated albumin measured by immunoturbidimetry and enzymatic method, respectively.

Results

Of those with CKD (n = 96), 79, 16 and 5% where in stages 3, 4 and 5, respectively. Those with CKD had higher levels of HbA1c (6.2 vs. 5.7%; p < 0.0001), glycated albumin (15.0 vs. 13.0%; p < 0.0001) and fructosamine levels (269.7 vs. 236.4 μmol/l; p < 0.0001), compared to those without CKD. Higher fasting glucose levels were associated with higher HbA1c, glycated albumin and fructosamine, independent of age, gender, and CKD. However, the association with HbA1c and glycated albumin differed by CKD status, at the upper concentrations of the respective markers (interaction term for both: p ≤ 0.095).

Conclusion

Our results suggest that although HbA1c and glycated albumin perform acceptably under conditions of normoglycaemia, these markers correlate less well with blood glucose levels in people with CKD who are not on dialysis.

HbA1c: a review of non-glycaemic variables

Identification of the correlation between HbA1c and diabetic complications has yielded one of the most clinically useful biomarkers. HbA1c has revolutionised the diagnosis and monitoring of diabetes mellitus. However, with widespread adoption of HbA1c has come increasing recognition that non-glycaemic variables can also affect HbA1c, with varying clinical significance. Furthermore, the identification of a discrepancy between predicted and measured HbA1c in some individuals, the so-called 'glycation gap', may be clinically significant. We aimed to review the current body of evidence relating to non-glycaemic variables to quantify any significance and provide subsequent suggestions. A PubMed-based literature search was performed, using a variety of search terms, to retrieve articles detailing the non-glycaemic variables suggested to affect HbA1c. Articles were reviewed to assess the relevance of any findings in clinical practice and where possible guidance is given. A range of non-glycaemic variables have statistically significant effects on HbA1c. While the clinical implications are generally irrelevant, a small number of non-glycaemic variables do have clinically significant effects and alternative biomarkers should be considered instead of, or in addition to, HbA1c. There are a small number of non-glycaemic variables which have a clinically significant effect on HbA1c, However, the vast majority of non-glycaemic variables have no clinical relevance. While clinicians should have an awareness of those non-glycaemic variables with clinical significance, in the vast majority of clinical scenarios HbA1c should continue to be used with confidence.

Mechanism of insulin resistance in a rat model of kidney disease and the risk of developing type 2 diabetes

Chronic kidney disease is associated with homeostatic imbalances such as insulin resistance. However, the underlying mechanisms leading to these imbalances and whether they promote the development of type 2 diabetes is unknown. The effect of chronic kidney disease on insulin resistance was studied on two different rat strains. First, in a 5/6^th nephrectomised Sprague-Dawley rat model of chronic kidney disease, we observed a correlation between the severity of chronic kidney disease and hyperglycemia as evaluated by serum fructosamine levels (p<0.0001). Further, glucose tolerance tests indicated an increase of 25% in glycemia in chronic kidney disease rats (p<0.0001) as compared to controls whereas insulin levels remained unchanged. We also observed modulation of glucose transporters expression in several tissues such as the liver (decrease of ≈40%, p≤0.01) and muscles (decrease of ≈29%, p≤0.05). Despite a significant reduction of ≈37% in insulin-dependent glucose uptake in the muscles of chronic kidney disease rats (p<0.0001), the development of type 2 diabetes was never observed. Second, in a rat model of metabolic syndrome (Zucker Lepr^fa/fa), chronic kidney disease caused a 50% increased fasting hyperglycemia (p<0.0001) and an exacerbated glycemic response (p<0.0001) during glucose challenge. Similar modulations of glucose transporters expression and glucose uptake were observed in the two models. However, 30% (p<0.05) of chronic kidney disease Zucker rats developed characteristics of type 2 diabetes. Thus, our results suggest that downregulation of GLUT4 in skeletal muscle may be associated with insulin resistance in chronic kidney disease and could lead to type 2 diabetes in predisposed animals.

Association of glycated hemoglobin with hemoglobin levels in elderly nondiabetic subjects

BACKGROUND

Glycated hemoglobin (HgbA1C) is being increasingly used for the diagnosis of diabetes mellitus due to its high availability and reproducibility. Several studies have shown that HgbA1C levels may be affected by levels of hemoglobin and nutritional factors associated with anemia, such as vitamin B12 and iron deficiency. However, none included elderly subjects. The aim of the present study was to investigate these effects in the older nondiabetic population.

METHODS

A retrospective cohort study design was used. The computerized database of a large health management organization was reviewed for all subjects without diabetes mellitus who underwent at least one measurement of HgbA1C and other hemoglobin parameters in 2002 at age≥65years. HgbA1C levels were correlated with hemoglobin, hematocrit, ferritin, iron, transferrin, vitamin B12, and folic acid levels.

RESULTS

A total of 11,352 subjects met the study criteria. Those with HgbA1C levels in the highest quintile (6.21-6.49%, 44.4-47.7mmol/mol) had significantly lower levels of hemoglobin, hematocrit, and iron than patients with HgbA1C levels in the lowest quintile (<5.4%, 36mmol/mol), but no linear correlation was found. There was no correlation of HgbA1C level with levels of ferritin, vitamin B12, and folic acid.

CONCLUSIONS

In elderly nondiabetic subjects, HgbA1C levels are not correlated with hemoglobin level or nutritional factors associated with anemia and may be interpreted without consideration of these factors.