Association between glycemic control and adverse outcomes in people with diabetes mellitus and chronic kidney disease: a population-based cohort study.

Better glycemic control as reflected by lower hemoglobin A(1c) (HbA(1c)) level may prevent or slow progression of nephropathy in people with diabetes mellitus (DM). Whether a lower HbA(1c) level improves outcomes in people with DM and chronic kidney disease (CKD) is unknown. From all people with serum creatinine measured as part of routine care in a single Canadian province from 2005 through 2006, we identified those with CKD based on laboratory data (estimated glomerular filtration rate [eGFR], <60.0 mL/min/1.73 m(2)]) and DM using a validated algorithm applied to hospitalization and claims data. Patients were classified based on their first HbA(1c) measurement; Cox regression models were used to assess independent associations between HbA(1c) level and 5 study outcomes (death, progression of kidney disease based on a doubling of serum creatinine level, or new end-stage renal disease [ESRD], cardiovascular events, all-cause hospitalization). We identified 23,296 people with DM and an eGFR lower than 60.0 mL/min/1.73 m(2). The median HbA(1c) level was 6.9% (range, 2.8%-20.0%), and 11% had an HbA(1c) value higher than 9%. Over the median follow-up period of 46 months, 3665 people died, and 401 developed ESRD. Regardless of baseline eGFR, a higher HbA(1c) level was strongly and independently associated with excess risk of all 5 outcomes studied (P < .001 for all comparisons). However, the association with mortality was U-shaped, with increases in the risk of mortality apparent at HbA(1c) levels lower than 6.5% and higher than 8.0%. The increased risk of ESRD associated with a higher HbA(1c) level was attenuated at a lower baseline eGFR (P value for interaction, <.001). Specifically, among those with an eGFR of 30.0 to 59.9 mL/min/1.73 m(2), the risk of ESRD was increased by 22% and 152% in patients with HbA(1c) levels of 7% to 9% and higher than 9%, respectively, compared with patients with an HbA(1c) level lower than 7% (P < .001), whereas corresponding increases were 3% and 13%, respectively, in those with an eGFR of 15.0 to 29.9 mL/min/1.73 m(2). A hemoglobin A(1c) level higher than 9% is common in people with non-hemodialysis-dependent CKD and is associated with markedly worse clinical outcomes; lower levels of HbA(1c) (<6.5%) also seemed to be associated with excess mortality. The excess risk of kidney failure associated with a higher HbA(1c) level was most pronounced among people with better kidney function. These findings suggest that appropriate and timely control of HbA(1c) level in people with DM and CKD may be more important than previously realized, but suggest also that intensive glycemic control (HbA(1c) level <6.5%) may be associated with increased mortality.