OBJECTIVE: To study the effects of postmenopausal estrogen therapy (ET) on nocturnal nonlinear heart rate variability (HRV). DESIGN: In this prospective, randomized, double-blind, placebo-controlled study, 71 healthy hysterectomized postmenopausal women received either transdermal estradiol or placebo for 3 months. After a washout period of 1 month, the treatments were reversed. Sleep studies were performed after both treatment periods. One steady-state epoch per night of the awake state, stage 2 (light) non-rapid eye movement (REM) sleep, stage 3-4 (deep) non-REM sleep, also known as slow-wave sleep, and REM sleep was extracted. From the electrocardiogram, nonlinear HRV was analyzed as the fractal scaling exponents alpha1 and alpha2, approximate entropy (ApEn), and the Poincare plot variability coefficients SD1 and SD2. These were correlated to ET use in both different sleep stages and averaged across all sleep stages. RESULTS: During ET, the nocturnal ApEn decreased from 0.80 +/- 0.01 to 0.74 +/- 0.02 (P < 0.05), the most marked reduction occurring during slow-wave sleep (from 0.77 +/- 0.05 to 0.63 +/- 0.06, P < 0.05). In addition, SD2 decreased in slow-wave sleep and REM sleep during ET (P < 0.05 for both). In light non-REM sleep, alpha1 slightly increased during ET (P < 0.05). CONCLUSIONS: ET has a slightly but distinctively attenuating effect on some nocturnal nonlinear measures of HRV, especially on complexity of heart rate dynamics. This implies that ET may have potentially deleterious effects on cardiovascular health during sleep.