Intraindividual Reproducibility of Heart Rate Variability Before and During Postural Tilt in Patients with Syncope of Unknown Origin

Fast Fourier analysis (FFT) of heart rate may be used to characterize overall heart rate variability (HRV) as well as low (LF: 0.06–0.10 Hz) and high (HF: 0.15–0.40 Hz) frequency components, which are attributable to sympathetic and vagal influences. This study examined the reproducibility of the spectral characteristics of HRV before and during postural tilt in patients being evaluated for unexplained syncope. Twenty patients (14 men, age 55 ± 12 years) with a history of syncopal episodes underwent 2 tilt table tests (T1 and T2) 1 to 6 weeks apart. For purposes of analyses the patients were divided into 2 groups according to the results of tilt testing. In 11 patients (Group A: 7 men, mean age 53.5 ± 12 years) syncope was induced in 1 (5 patients) or both tilt tests. In the remaining 9 patients (Group R: 7 men, age 56.7 ± 14 years) syncope did not develop during either test. Spectral indexes (SI) of HRV were computed by FFT for 2‐minute segments derived from Holter recordings before and during 60° head‐up tilt. We compared the SI measured in the two tilt, tests during the following 2‐minute periods: P1, just before tilt at rest, in a supine position; P2, immediately after tilt; P3, just before the end of the tilt. Results: P1 and P2: There were no significant differences in SI between T1 and T2 for low frequencies, high frequencies, or total power in either Group. In Group A, LF at time of maximum tilt exposure (P3) tended to be slower during T2 than in T1 (5.58 ± 1.44 vs. 6.56 ± 1.11, P < 0.05). However, this difference primarily reflects the findings in the five patients in whom syncope did not occur during T2. When these 5 patients were excluded the LF were essentially identical (6.43 ± 0.97 vs 6.80 ± 1.30; P = NS). Conclusions: Spectral indexes of HRV before and during the early part of tilt testing are reproducible. The SI during the later part of the test are related to the patient's response. However, this response is not predictable on the basis of sympathetic or parasympathetic tone before or during the early part of the test.