Autonomic dysregulation in patients awaiting kidney transplantation.

Overview

abstract

Sudden cardiac death occurring in patients with end-stage renal disease (ESRD) may be related to poor autonomic function (AF). It is not known whether patients having a sudden death can be identified by commonly used AF evoked tests or if a newer test evaluating heart rate variability (HRV) with power spectral analysis can better distinguish at-risk patients. This study sought to characterize AF in patients awaiting kidney transplantation, to identify factors associated with poor AF and sudden death, and to compare evoked versus 24-hour measures of cardiac AF. All patients underwent evoked cardiac AF tests, which included changes in heart rate with deep breathing (deltaBPM) and valsalva (VR). In addition, 24-hour HRV was assessed with time domain measurements of interbeat variability (pNN50, SDANN, and SDNN), which are associated with vagal function, circadian function, and sudden cardiac death (SDNN < 50), respectively. Frequency domain measures obtained by power spectral analysis (total, low, and high hertz) quantify total neural, sympathetic, and parasympathetic activity of the heart, respectively. Data were collected on 184 nondiabetic patients, 60 type 1 diabetic patients, and 34 type 2 diabetic patients with ESRD referred for transplantation. Five patients, all receiving peritoneal dialysis, experienced nontraumatic sudden cardiac death during the study. Evoked and 24-hour HRV control data were obtained from 67 and 48 healthy adults, respectively. Data show that regardless of subgroup, there was significant AF dysregulation in the 278 patients with ESRD, particularly for those with diabetes and those receiving peritoneal dialysis. Frequency domain measurements (three in each group: nondiabetic patients, type 1 diabetic patients, type 2 diabetic patients, deceased patients, hemodialysis patients, peritoneal dialysis, and nondialysis patients [n = 21]) were most sensitive to dysregulation, with 16 of 21 (76%) measurements more than 2 SD from the mean of the control group. This is in contrast to the time domain measurements (one of 21 [0.04%] > 2 SD from the mean of the control group) and evoked measurements (eight of 14 [57%] outside of the established norms). Of the five deceased patients, only one displayed normal values for all eight AF measurements reported; three (60%) had SDNNs less than 50. Of the 248 surviving patients, 42 (17%) had an SDNN less than 50. When analyzed, the ability of the SDNN to identify an at-risk group was found to have a sensitivity of 60%, a specificity of 83%, a positive predictive value of 7%, a negative predictive value of 99%, and an accuracy of 83%. While AF and time on dialysis were not found to be correlated, the length of diabetes was inversely related to all AF measures (r = -0.27 to -0.48; P < 0.0001), except pNN50. These data suggest that all groups of ESRD patients have severely compromised AF and, regardless of the type of diabetes, those with diabetes have the greatest degree of dysregulation. In addition, individuals receiving peritoneal dialysis were more prone to dysregulation, and the SDNN, a time domain measurement of 24-hour HRV, holds the promise of identifying patients at increased risk for early death in this population.