Beat-to-beat assessment of spontaneously occurring fluctuations in heart rate and arterial blood pressure allows noninvasive determination of cardiovagal function, but little is known regarding the effects of general anesthesia on spontaneous baroreflex (SBR) indices. We examined (a) concentration-dependent effects of sevoflurane on SBR indices, heart rate variability (HRV), and blood pressure variability and (b) correlation and agreement between pharmacological baroreflex gains and SBR indices during sevoflurane anesthesia. Continuous electrocardiogram and invasive arterial blood pressure were monitored in nine healthy volunteers before, during, and for 3 h after sevoflurane anesthesia, during which end-tidal sevoflurane was maintained at 0.7%, 1.4%, and 2.0% in random sequences. We derived three SBR indices (sequence method, alpha-index, and low-frequency transfer function) and compared them with pressor and depressor test gains by the pharmacological method. HRV and blood pressure variability were analyzed at a fixed respiratory rate (12 breaths/min) in awake and anesthetized conditions. Except for low-frequency transfer function, SBR indices were depressed by sevoflurane and remained depressed for 30 min after emergence from anesthesia, compared with the conscious baseline value. Spontaneous sequence indices and high- and low-frequency powers of HRV demonstrated concentration-dependent depression. Pharmacological gains and SBR indices during anesthesia generally correlated well, but Bland-Altman analysis revealed that SBR indices had limits of agreement as large as the baroreflex gain itself. These data suggest that spontaneous indices are inadequate estimates of, and are inconsistent with, the pharmacological baroreflex gain during sevoflurane anesthesia.