Endocrine Abstracts

Background: Clinical manifestations of hyperthyroidism resemble those of hyperadrenergic state. Previous studies of linear analysis of heart rate variability (HRV) revealed cardiac autonomic dysfunctions in hyperthyroidism. In recent years, increasing attention is being directed to the nonlinear analysis of HRV to gain insight into neural modulation of heart rate dynamics. Based on the chaos and fractal theory, an attractor in phase space can characterize the dynamics of a chaotic system. Besides, the complexity of the system can be quantified by the properties of the attractor, one of which is the correlation dimension (CD).

Objective: We investigated whether it is possible to distinguish between hyperthyroid patients and normal control subjects based on CD analysis of HRV.

Subjects and Methods: The study subjects comprised 15 newly diagnosed hyperthyroid Graves’ disease patients and 15 sex-, age-, and body mass index- matched normal control subjects. All subjects received one-channel electrocardiogram recording for 30 minutes. We computed CD analysis of HRV from the sequence of normal R-R intervals by the Grassberger and Procaccia algorithm.

Results: The slope of the straight line of best fit in the linear scaling region of the plot of [ln C(r) vs ln r] is determined as the value of CD. We observed that the values of CD reached a constant value of saturation (which is the value of CD) with the increasing embedding dimension. The CD was significantly reduced in hyperthyroid patients compared with the normal control subjects (4.88±0.12 vs 6.35±0.11, mean ± S.E., P<0.001).

Conclusions: These results suggest reduced complexity in hyperthyroidism. In analyzing the heart rate dynamics in hyperthyroid patients, the CD could be used to distinguish hyperthyroid patients from normal subjects. Thus, the CD analysis of HRV gives additional information about neural modulation of the cardiovascular system in hyperthyroidism.