Reproducibility of methods for assessing baroreflex sensitivity and temporal QT variability in end-stage renal disease and healthy subjects

Abstract

Spontaneous baroreflex sensitivity (BRS), the reflex heart rate modulation in response to blood pressure changes (predominantly an index of cardiac vagal activity) and temporal QT variability (an index of myocardial repolarization) have been demonstrated to convey important prognostic information. The information about reproducibility of BRS and temporal QT variability is limited and there is lack of information regarding patients with cardiovascular diseases. We investigated reproducibility of spontaneous BRS using the sequence technique and temporal QT variability index (QTVI) in terms of intra-, interexaminer and within-subject variability in end-stage renal disease patients (ESRD, n=17, age 55±14 years) and healthy subjects (HS, n=29, age 32±12 years, P<0.01). ECG and blood pressure (Portapres) were recorded on two separate days and BRS and QTVI were evaluated by two independent examiners. The mean heart rate was similar in ESRD patients in comparison to healthy controls, whereas the mean arterial pressure was 13 % higher in ESRD patients (P<0.01). Spontaneous BRS was 62% lower (P<0.01) and QTVI was 41% higher in ESRD patients (P<0.01) compared to healthy subjects, respectively. Coefficient of variation (CV) of within-subject reproducibility of BRS and QTVI measurements was moderate (BRS: 33 % for ESRD, 27% for HS; QTVI: 40% for ESRD, 18% for HS). The 95% limit of within-subject reproducibility of BRS measurements was 3.8ms/mm Hg for ESRD patients and 8.1ms/mm Hg for healthy subjects; whereas the 95% limit of reproducibility of within-subject reproducibility of QTVI measurements was 0.73 for ESRD patients and 0.55 for healthy subjects. Concordance correlation coefficients of within-subject variability of BRS and QTVI were between 0.74 and 0.83 in both groups. CV of intra- and inter-examiner reproducibility of BRS and QTVI measurements in both groups ranged between 1 and 11%. In conclusion, the intra- and inter-examiner reproducibility/agreement of BRS and QTVI were high, whereas the within-subject reproducibility of these two methods was moderate, in both ESRD patients and healthy subjects. Thus, small differences in BRS and QTVI in longitudinal/interventional studies should be interpreted with caution.