Background —Altered cardiovascular variability is a prognostic indicator for cardiovascular events. Patients with obstructive sleep apnea (OSA) are at an increased risk for cardiovascular disease. We tested the hypothesis that OSA is accompanied by alterations in cardiovascular variability, even in the absence of overt cardiovascular disease. Methods and Results —Spectral analysis of variability of muscle sympathetic nerve activity, RR interval, and blood pressure were obtained during undisturbed supine rest in 15 patients with moderate-to-severe OSA, 18 patients with mild OSA, and 16 healthy control subjects in whom sleep disordered breathing was excluded by complete overnight polysomnography. Patients with OSA were newly diagnosed, never treated for OSA, and free of any other known diseases. Patients with moderate-to-severe OSA had shorter RR intervals (793±27 ms) and increased sympathetic burst frequency (49±4 bursts/min) compared with control subjects (947±42 ms; 24±3 bursts/min; P =0.008 and P <0.001, respectively). In these patients, total variance of RR was reduced ( P =0.01) and spectral analysis of RR variability showed an increase in low frequency normalized units, a decrease in high frequency normalized units, and an increase in the ratio of low to high frequency (all P <0.05). Even though blood pressure was similar to that of the control subjects, blood pressure variance in patients with moderate-to-severe OSA was more than double the variance in control subjects ( P =0.01). Patients with mild OSA also had a reduction in RR variance ( P =0.02) in the absence of any significant difference in absolute RR interval. For all patients with OSA, linear regression showed a positive correlation ( r =0.40; P =0.02) between sleep apnea severity and blood pressure variance. Conclusions —Cardiovascular variability is altered in patients with OSA. This alteration is evident even in the absence of hypertension, heart failure, or other disease states and may be linked to the severity of OSA. Abnormalities in cardiovascular variability may be implicated in the subsequent development of overt cardiovascular disease in patients with OSA.

Background— Sympathetic and parasympathetic systems are considered the principal rapidly reacting systems that control heart rate. Methods and Results— We propose a symbolic analysis series to quantify the prevalence of sympathetic or parasympathetic cardiac modulation. This analysis decomposes the heart rate variability series in patterns lasting 3 beats and classifies them into 3 categories: nonvariable, variable, and very variable patterns referred to as 0V, 1V, and 2V patterns. First, we applied this method to experimental and pharmacological conditions characterized by sympathetic activation (tilt test, handgrip, nitroprusside, and high-dose atropine administration) or parasympathetic activation (phenylephrine and low-dose atropine administration) in 60 healthy subjects. An increase in sympathetic modulation and a vagal withdrawal elicited a significant increase in 0V patterns and a decrease in 2V patterns, whereas parasympathetic dominance induced the opposite, reflecting a reciprocal sympathovagal balance. The second part of the study considered a series of 300 beats before the onset of major arrhythmic events in patients with an implantable cardioverter-defibrillator. Symbolic analysis detected an increase in the percentage of 0V patterns before the onset of major arrhythmias compared with baseline (41.6±3.9% and 24.4±2.9%, respectively; P <0.01), indicating a sympathetic prevalence. On the other hand, the 2V patterns did not decrease before major arrhythmias, suggesting the presence of nonreciprocal autonomic modulations. Conclusions— Symbolic analysis of 3 beat sequences takes into account the different time course of sympathetic and parasympathetic cardiac modulations and seems appropriate for elucidating the neural pathophysiological mechanisms occurring during the short periods that precede acute cardiac events.

Background —Short-term variability of RR interval and blood pressure occurs predominantly at low frequency (LF; ≈0.1 Hz) and high frequency (≈0.25 Hz). The arterial baroreflex is thought to be the predominant determinant of the LF component of RR variability. Patients with severe congestive heart failure (CHF) have an attenuated or absent LF oscillation in RR variability. The left ventricular assist device (LVAD) offers a unique possibility for analysis of spectral oscillations in RR interval independent of any effects of blood pressure that influence these oscillations via the baroreflex. Methods and Results —We performed spectral analysis of RR, blood pressure, and respiration in 2 patients with CHF before and after LVAD implantation. LF components of the RR-interval and blood pressure variability were absent in both CHF patients before LVAD implantation. After LVAD implantation, spectral analysis of the RR interval showed restoration of a clear and predominant LF oscillation in the native hearts of both patients, with no such oscillation evident in the blood pressure profile. Conclusions —During total circulatory support with the LVAD, the LF oscillation in RR interval of the native heart, absent in CHF, is restored. This LF oscillation in RR interval occurs in the absence of LF oscillations in blood pressure and thus is unlikely to be explained by baroreflex mechanisms. Hence, the absence of LF oscillation in the RR interval in CHF is functional and is reversible by LVAD circulation. The presence of a predominant LF oscillation in RR interval independent of any oscillation in blood pressure suggests that the LF oscillation is a fundamental property of central autonomic outflow.