Abstract
Although poor glycemic control is strongly associated with adverse clinical outcomes, few studies have evaluated their risk across the full glycemic spectrum in patients with moderate to advanced chronic kidney disease (CKD). Guidelines for HbA1c targets are mainly derived from extrapolation of trials conducted in people with normal or mildly reduced kidney function. In non-dialysis CKD individualized targets are recommended, often aiming for less strict glycemic control, due to lack of evidence for the efficacy of lower HbA1c, and potentially greater harm. In this study, we analyzed long-term outcomes across the full spectrum of HbA1c in CKD stage 4-5.
Using data from the Swedish Renal Registry, we included patients with diabetes and CKD stages 4-5. Adjusted cause-specific Cox models were used to analyze associations between the one-year baseline mean HbA1c level (<43, 43-52 [reference], 53-64, 65-74, 75-89, ≥90 mmol/mol) and the occurrence of major adverse cardiovascular events (MACE, a composite of cardiovascular death, hospitalization for myocardial infarction or stroke), kidney replacement therapy (KRT), all-cause mortality, and hospitalization/death due to hypoglycemia.
We included 2571 patients (71 years, 64% men, eGFR 20 mL/min per 1.73m², HbA1c 59.7 mmol/mol [7.6%], 75% insulin users). At baseline, patients with a higher HbA1c level were younger, had a higher prevalence of cardiovascular disease and were more likely prescribed insulin. Kidney function and malnutrition-inflammation markers were, however, not different between groups.
Over a median follow-up of 4.9 [4.6;5.2] years, 939 (37%) KRT, 1078 (42%) MACE, and 1523 (59%) deaths occurred. Increasing HbA1c was gradually associated with the risk of MACE (adjusted HR from 0.76 [95% CI: 0.60-0.96] for HbA1c <43 mmol/mol to 1.39 [1.01-1.92] for HbA1c ≥ 90 mmol/mol) (Fig.). In contrast, a J-shaped relationship was observed for the risk of all-cause mortality. Compared to the reference group (43-52 mmol/mol; 5-year absolute risk 52% [48-56]), a HbA1c <43 mmol/mol was not associated with a higher mortality (adjusted HR 1.03 [0.86-1.23], 5-year absolute risk 57% [51-63]), while any HbA1c above 75 mmol/mol was associated with a higher risk of death. The risks of KRT and severe hypoglycemia were not different across HbA1c levels.
In CKD stage 4-5, HbA1c above the general recommendation for non-CKD patients (43-52 mmol/mol) is associated with a higher risk of MACE and all-cause mortality, but not KRT, without increased risk of severe hypoglycemia. This suggests that targeting a lower HbA1c, together with other cardio-reno-metabolic prevention strategies, might contribute to improve health outcomes in this very high-risk population.
Adjusted cumulative incidence curves for adverse clinical outcomes, by baseline HbA1c level.
Comments