Impact of glucose control

Blood glucose control lowers the incidence of new-onset microalbuminuria and retards the progression of established diabetic nephropathy in patients with relatively preserved GFR, according to the DCCT and UKPDS. Still unclear, though, is what benefit blood glucose control has on CKD progression in patients with DM who are at stage 3 CKD or higher.

The impact of blood glucose control on mortality in patients with type 1 DM, normal serum creatinine, and variable degrees of albuminuria was examined in a single study (BMJ. 1996;313:779-784). Researchers followed 939 adults with type 1 DM for more than 10 years and found that a higher HbA1c was associated with higher mortality.

Continue Reading

In our study of veterans with type 2 diabetes and CKD stages 3 and 4, however, lower HbA1c levels were associated with higher mortality (J Am Soc Nephrol. 2007;18:573A). More studies are needed in patients with non-dialysis-dependent CKD to corroborate these seemingly provocative findings, although they do dovetail with recent findings from randomized controlled trials in patients with normal kidney function and long-standing diabetes and cardiovascular disease.

Outcomes associated with glycemic control are somewhat better studied in dialysis patients, even though the number of studies in this field is also limited and they, too, are mostly observational in nature. In a large study of 24,875 U.S. diabetic dialysis patients, Williams et al found no association between HbA1c and survival at 12 months (Kidney Int. 2006;70:1503-1509).

However, in the previously cited study by Kalantar-Zadeh et al., 23,618 diabetic patients on maintenance HD had a paradoxically higher unadjusted mortality associated with lower HbA1c levels, but adjustment for markers of malnutrition and inflammation resulted in lower HbA1c levels’ becoming associated with lower mortality.

The discrepancies between observational studies can be resolved by randomized controlled trials; unfortunately, only a single interventional study examined clinical outcomes in 83 dialysis patients undergoing intensive diabetes-related intervention compared with standard care (Am J Kidney Dis. 2002;40:566-575). Patients in the intensive intervention arm of that study had better glycemic control and experienced improved quality of life and a reduction in the need for amputations and hospitalizations.

The scarcity of clinical trials and the equivocal results of observational studies examining the impact of glycemic control on outcomes in CKD and ESRD should prompt the cautious implementation of guidelines established in patients with normal kidney function. Careful consideration of the unique characteristics of glycemic dysregulation in CKD and the dangers of hypoglycemia are warranted in these patients.

Other caveats

Measures of glycemic control. The HbA1c level has been used as an indicator of integrated glucose control in all patients, including those with various stages of CKD. HbA1c levels can, however, be affected by several factors that are specific to the uremic state.

Elevated blood urea nitrogen (BUN) levels can result in the formation of carbamylated hemoglobin, which cannot be distinguished from HbA1c if using electrical charge-based laboratory assays (cation exchange chromatography). This issue is of lesser concern with the use of different laboratory assays. In addition, factors associated with potential lowering of HbA1c levels in uremia include a shortened RBC life span and blood transfusions.

The life span of RBCs is, however, close to normal in well-dialyzed patients, and routine blood transfusions have been rarely needed with the advent of erythropoesis-stimulating agents. Other potential measures of long-term glucose control in uremia, such as fructosamine or glycated albumin, have been suggested (Diabetes Res. 1990;13:127-131), but they depend on normal serum albumin levels and therefore could be affected by the frequent presence of hypoalbuminemia in this patient population.

Antidiabetic agents. Antidiabetic medications depend in varying degrees on renal excretion; hence they may or may not require adjustments with advancing stages of CKD. Biguanides represent a special category in that they should be withheld in patients with even a modest decline of kidney function (the exact degree of which is still ill-defined) because of the rare but potentially fatal lactic acidosis that is caused by the accumulation of these drugs, particularly metformin.

The other hypoglycemic agents may or may not require dose adjustments, but dosing guidelines are hampered by variable quantification of kidney function in the dosing recommendations of individual medications.


DM represents an important comorbidity in CKD. Control of hyperglycemia could be beneficial in improving outcomes in this patient population, as glycemic control retards progression of diabetic nephropathy in its early stages and is probably beneficial in later stages as well.

Glycemic control might also reduce mortality in diabetic patients with CKD, but avoidance of hypoglycemia is important, especially since uremic dysregulation of glycemic control can result in spontaneous improvement of hyperglycemia and sometimes even normoglycemia. Further characterizations of the pathophysiology and of the clinical impact of burnt-out diabetes represent a major challenge for future research.

Any efforts to treat hyperglycemia in CKD must take into account the complex changes in glucose and insulin metabolism, the effect of decreased kidney function on the pharmacokinetics of many hypoglycemic medications, and the specific aspects of the type of renal replacement therapy administered in patients with ESRD. Careful monitoring and individualized therapy is recommended to achieve good glycemic control and to minimize the occurrence of hypoglycemic episodes.

Dr. Kovesdy is Chief of Nephrology at Salem VA Medical Center in Salem, Va., and assistant professor of clinical internal medicine, at the University of Virginia in Charlottesville.