Beta slope is a nonlinear index derived from the power-law analysis of 24-hour heart rate variability (HRV); in healthy subjects, the beta index is typically near -1, suggesting a fractal behavior of HRV. There is scarce data on HRV in Chagas\' disease. This transversal study intends to describe power-law and linear HRV patterns in different forms of Chagas\' disease. Patients and healthy controls (n = 26) without other diseases were submitted to a standardized protocol, including electrocardiography, echocardiography, and 24-hour Holter monitoring. Patients with Chagas\' disease were divided into groups according to their left ventricular (LV) systolic function: normal (group 1, n = 85), segmental abnormalities (group 2, n = 49), and reduced ejection fraction (group 3, n = 26). The temporal series were carefully processed to obtain the HRV indexes and beta slope. Despite the differences in the LV systolic function, the 3 groups were comparable in terms of long-term HRV index values. After adjustment for covariates, short-term HRV index values were consistently reduced in the Chagas\' disease groups. The beta-index values were also diminished in Chagas\' disease groups (group 1: -1.09 +/- 0.03, group 2: -1.11 +/- 0.06; group 3: -1.14 +/- 0.03; and controls: 0.95 +/- 0.03, p <0.001). This breakdown of fractal long-range correlation of RR interval dynamics, a strong predictor of mortality in other cardiomyopathies, may reflect cardiac dysautonomia that may have gone undetected in long-term time-domain analysis. This abnormality may explain the increased risk for arrhythmic sudden death found in Chagas\' disease even in the absence of signs of LV dysfunction.