Autonomic function in type I diabetes mellitus complicated by nephropathy a cross-sectional analysis in the presymptomatic phase

The purpose of this study was to determine the prevalence of parasympathetic and sympathetic autonomic dysfunction in long-standing type I diabetics with established nephropathy and to correlate autonomic function with cardiac risk factors. We used prospective analysis of heart rate variations to standardized testing and 24-hour blood pressure control prior to enrollment in a study utilizing various methods of intense diabetic control to prevent deterioration of kidney function. The settings were outpatient clinical research units. The patients were 42 type I diabetics with proteinuria (total urinary protein ≥300 mg/day or urinary albumin ≥100 mg/day) and creatinine clearance ≥30 mL/min. Heart rate variation during respiratory cycles with change in posture from supine to upright, and during the Valsalva maneuver was recorded by a computerized method. Mean arterial blood pressure was recorded for 24 h by a computerized method. Heart rate variations in this group were abnormal during timed respiratory cycles in 26 of 40 patients (56%), during changes in posture in 15 of 40 patients (38%), and during Valsalva maneuver in 13 of 34 patients (38%) whose retinal disease permitted Valsalva testing. Blunted day/night mean arterial pressure ratios occurred in 18 of 41 (44%) patients and were more severe in men than in women (1.00 v 1.06, P ≤.05). Absence of deep tendon reflexes was associated with an increased incidence of both parasympathetic (respiratory rate variation) and sympathetic (postural rate variation) abnormalities (both P ≤.05). Loss of vibration sensation was not associated with autonomic functional abnormalities. Patients in whom heart rate variations during respiratory cycles were decreased had significantly lower mean serum albumin and higher incidence of the need for laser treatment for retinal microvascular disease. Individuals with an HDL cholesterol ≤40 mg/dL had a significantly higher incidence of blunted day/night mean arterial pressure ratios. Gender differences occurred in mean levels of HDL cholesterol (women, 54 v men, 45 mg/dL, P< .05). The results of autonomic nerve testing were not predictable from history, physical examination, or blood testing. The levels of blood pressure, serum, total cholesterol, triglyceride, creatinine, urinary protein excretion, and creatinine clearance did not correlate with either prevalence or severity of autonomic nerve testing. Follow-up to determine the relevance of autonomic nerve testing as an independent cardiac risk factor in this high-risk population is ongoing.