Reduced heart rate variability (HRV) and attenuated baroreflex sensitivity (BS) after myocardial infarction and in patients with chronic congestive heart failure (CHF) are associated with poor prognosis. Recent studies have shown that a large proportion of the prognostic power from HRV measurements is localized in heart rate turbulence immediately after ventricular premature complexes. The mechanism of heart rate turbulence remains unknown. In the present study, we explore its relation to BS. In 45 patients with CHF and ≥3 ectopic beats in a 30-minute period, measurements of RR interval and continuous, noninvasive blood pressure (BP) were studied at rest. In response to an ectopic beat, average heart rate turbulence was 9.4 ms/beat (SD 6.1). Mean BP turbulence was 0.72 mm Hg/beat (SD 0.56). Using the ratio of heart rate and BP turbulence slopes to estimate BS showed good agreement (r = 0.67, p <0.0001) with the α-index method (BSα). This relation was attributable to a marked correlation between heart rate turbulence and BSα (r = 0.70, p <0.0001); there was no correlation between BP turbulence and the BSα (r = 0.1, p = NS). Twenty-nine percent of patients had postectopic pulsus alternans, with a mean decay time of 1.4 beats (SD 0.5). The presence of pulsus alternans was associated with a significantly lower heart rate turbulence slope (6.3 [SEM 1.0] vs 10.7 [SEM 1.2] ms/beat, p = 0.03). Thus, heart rate turbulence is an effective measure of the baroreflex, correlating strongly with a standard measure. This is because it is the heart rate, rather than the BP, response to an ectopic beat that conveys the information relevant to BS measurement.