Intensive Glycemic Therapy in Type 2 Diabetes Patients With Cardiac Autonomic Dysfunction: The ACCORD Trial

Abstract

ObjectiveTo assess whether the presence of cardiac autonomic dysfunction denoted by low heart rate variability (HRV) modifies the effect of intensive glycemic therapy on outcomes in patients with type 2 diabetes. Patients and MethodsThis study included 7946 participants in the ACCORD (Action to Control Cardiovascular Risk in Diabetes) trial from January 2001 through June 2009. Heart rate variability measures included standard deviation of all normal-to-normal intervals (SDNN) and root mean square of successive differences between normal-to-normal intervals (rMSSD). Abnormal values were defined based on less than the 10th percentile for SDNN and rMSSD. ResultsCompared with standard therapy, intensive therapy was associated with improved primary outcome (composite of cardiovascular events) in the low-HRV group (SDNN: HR, 0.57; 95% CI, 0.39 to 0.84; rMSSD: HR, 0.57; 95% CI, 0.38 to 0.84), but not in the normal-HRV group (SDNN: HR, 0.90; 95% CI, 0.77 to 1.05; rMSSD: HR, 0.90; 95% CI, 0.77 to 1.05). A similar pattern was found for coronary heart disease. Conversely, intensive therapy had a neutral effect on all cause death in the low-HRV group (SDNN: HR, 0.88; 95% CI, 0.54 to 1.41; rMSSD: HR, 0.71; 95% CI, 0.43 to 1.17;), but increase risk of all-cause death in the normal-HRV group (SDNN: HR, 1.21; 95% CI, 1.00 to 1.46; rMSSD: HR, 1.25; 95% CI, 1.03 to 1.51). Intensive therapy induced a greater risk of hypoglycemia in the normal-HRV group than that in the low-HRV group. ConclusionCardiac autonomic dysfunction expressed as low HRV identified subpopulations in ACCORD with more benefits and less harms from intensive therapy.