Diagnostic Application of the A_1c Assay in Renal Disease

A recent International Expert Committee, jointly appointed by the American Diabetes Association, the European Association for the Study of Diabetes, and the International Diabetes Federation, now advocates the use of glycated hemoglobin, measured by A_1c assay, for diagnosing diabetes in nonpregnant individuals.¹ Their recommendations suggest the diagnosis of diabetes when the A_1c level is >6.5%, confirmed on repeat testing when patients are asymptomatic. In addition, patients with an A_1c between 6.0 and 6.5% should be considered high risk, analogous to the state of prediabetes defined by either impaired fasting glucose or glucose tolerance. The rationale for this new approach emphasizes the advantages of the A_1c assay over current diagnostic measures for accuracy, precise measurement of chronic glycemia, correlation with diabetic complications, and logistical ease of use. This proposal is under consideration by diabetic associations for possible implementation but lacks analysis of the validity of the A_1c assay for diagnosing diabetes in the setting of renal impairment and renal replacement therapy (RRT).

Hemoglobin A_1c is the product of a nonenzymatic reaction between circulating hemoglobin and glucose. In the normal 120-day life cycle of the average red blood cell, the magnitude of glycated hemoglobin is reflective of longitudinal exposure to glucose. Glycation rate can be influenced by temperature, pH, red cell turnover, hemoglobin concentration, glucose concentration, and length of exposure to glucose.² The Diabetes Control and Complications Trial (DCCT) and United Kingdom Prospective Diabetes Study (UKPDS), conducted of patients with type 1 and type 2 diabetes respectively, were the pivotal studies demonstrating a link between A_1c and diabetes-related complications.^3,4 High performance liquid chromatography (HPLC) using Diamat calibration is the current technique, and all laboratories are standardized against the DCCT assay run under the auspices of the National Glycohemoglobin Standardization Program. In an additional ongoing development, the A_1c assay will be standardized globally and reported solely in International Federation of Clinical Chemistry units.

It is clear from our understanding of the A_1c assay that an imbalance in red blood cell survival distorts the clinical measurement of glycated hemoglobin. In the setting of renal impairment or RRT, in conjunction with relevant concomitant pharmacotherapy or relevant comorbidities, numerous confounders will affect the legitimacy of the A_1c assay.

The uremic state indirectly affects the accuracy of the A_1c assay through altered erythropoiesis but also through direct interactions with glycated hemoglobin analyses. Uremia-induced hemoglobin modification, forming carbamylated hemoglobin, also interferes with the laboratory analysis of the A_1c assay, although current HPLC, standardized and aligned to the DCCT assay, should minimize or eliminate this interference.⁵

In addition, the onset of renal anemia with advancing renal dysfunction is associated with deficiencies in iron, folate, and erythropoietin that often require replenishment. Iron deficiency anemia associates with an elevated A_1c in patients without diabetes, pharmacologic replacement reversing the effect,⁶ although no similar data exist in the context of renal disease. Folate metabolism is altered in renal failure as a result of impaired intestinal absorption and contributes to altered red blood cell survival and, hence, is likely to affect the A_1c level. The effect of erythropoietin therapy on lowering glycated hemoglobin can be putatively linked to either shortening in red blood cell lifespan or changing proportion of old to young erythrocytes. Finally, declining renal function is associated with the development of systemic acidosis, and this environment contributes to enhanced glycated hemoglobin levels in chronic renal failure.⁷ More recently, the role of oxidative stress in the glycation of hemoglobin in chronic renal failure opens new uncertainty.⁸ These observations translate to patients on RRT, with alterations in glycated hemoglobin affected by erythropoietin therapy and arterial pH.⁹

Other factors that artificially decrease the A_1c assay in hemodialysis patients include blood loss (during treatment or frequent blood sampling), shortened red blood cell survival, red blood cell transfusions, and erythropoietin treatment.¹⁰ In the context of peritoneal dialysis, initiation of continuous peritoneal ambulatory dialysis coincides with an increase in levels of glycated hemoglobin perhaps as a result of confounding by dialysate absorption.¹¹

The diagnosis of new-onset diabetes after transplantation (NODAT) has been aligned with American Diabetes Association guidelines since 2003.¹² The NODAT guidelines do not recommend the use of the A_1c assay for diagnosis of NODAT because of lack of sensitivity in line with contemporary statements regarding the general population; the new A_1c assay recommendations, however, provide evidence to dismiss these concerns and therefore abrogate previous statements. The original NODAT guidelines do recommend using the A_1c assay for monitoring established NODAT but urge caution in interpreting the assay in the context of anemia and renal impairment.

The prevalence of late posttransplantation anemia (occurring during 12 mo after surgery) is high, reported between 20 and 57%, and has important implications for the adoption of glycated hemoglobin for diagnostic use after transplantation.¹³ The role of immunosuppression, notably the antiproliferative agents, on posttransplantation anemia is also well documented. Vanrenterghem et al.¹⁴ found significant differences in the hemoglobin levels of transplant recipients who were taking mycophenolate mofetil (MMF). No similar finding is associated with azathioprine, apart from a subgroup of patients with serum creatinine ≤2 mg/dl. MMF selectively blocks de novo purine synthesis and would not be expected to have a significant effect on the proliferation of bone marrow cells, although reproducible erythroid aplasia associated with MMF was observed in a pediatric case.¹⁵ Azathioprine is widely known to cause bone marrow suppression and is also implicated in macroscopic anemia.¹⁶ Compared with MMF, treatment with sirolimus has a higher prevalence of posttransplantation anemia independent of variables such as graft function.¹⁷ Sirolimus presumably inhibits erythropoiesis at the receptor level, with mammalian target of rapamycin inhibition blocking S6 kinase activity, and impairs replication in an erythroid cell line.¹⁸ Other pharmacologic agents that are used in a posttransplantation setting and associated with anemia include inhibitors of the renin-angiotensin-aldosterone system, trimethoprim-sulfamethoxazole, OKT3, anti-thymocyte globulin, and ganciclovir.¹³ Non–immunosuppression-related factors attributable to posttransplantation anemia include allograft dysfunction, rejection, nonpolycystic kidney disease, infection, antihypertensive use, iron/B₁₂/folate deficiency, and donor age.¹³

It is easily appreciated that the adoption of a new A_1c-based diagnostic model for diabetes in patients who have renal failure or are on RRT is fraught with analytical questions that will affect the validity of the assay. These concerns seem more legitimate in the context of hemodialysis compared with other renal patient groups as a result of a multitude of potential confounders that interfere with the A_1c assay. This would be consistent with articles arguing for a weak correlation between glycated hemoglobin and blood glucose levels in dialysis patients. There is a paucity of data in patients across a wide spectrum of chronic kidney disease, and it is logical to assume the validity of the A_1c assay will diminish in the context of advancing renal failure. There is a stronger argument for the adoption of the A_1c assay in renal transplant recipients because of their high-risk status, and the logistical advantages of such an approach over current glucose-based strategies may outweigh associated caveats.

At this time, the current glucose-based approach is the sole method for diagnosing diabetes or NODAT. Should global diabetes associations advocate adoption of the International Expert Committee proposals for a transition to an A_1c-based diagnosis of diabetes, it is unlikely that any specific recommendations will be made for patients who have renal disease or have received a renal transplant.¹ In this setting, nephrologists will need to interpret such recommendations in the context of the unique circumstances across a range of renal patients rather than simply translating the guidelines directly. We urge further work in this area to determine the optimum diagnostic criteria for diabetes in patients with renal disease. In light of current deliberations among diabetologists on this change in diagnostic strategy, we encourage preemptive debate on this matter to ensure the renal community can reach consensus in a timely manner.

Disclosures

None.