Prognostic role of baroreflex control of heart rate: further insights from dynamic assessment of baroreceptor-cardiac reflex?

The arterial baroreflex represents a fundamental mechanism of cardiovascular homeostasis [1] and it is evident that its impairment plays an adverse role in several cardiovascular diseases. An example is myocardial infarction, in which an impairment of the baroreflex mechanisms controlling heart period, as expressed by a reduced baroreflex sensitivity (BRS), has a negative prognostic value for cardiac mortality and cardiac events [2]. Another example is heart failure, in which a reduced BRS occurs that is linked to the severity of disease and negatively affects the prognosis [3]. Again, an impaired baroreflex modulation of heart rate is also a key feature of arterial hypertension, contributing to the maintenance of high blood pressure [4], and has also been described in the course of diabetic disease, even in the early stages [5,6]. The above findings have stimulated a growing interest in the investigation of BRS in man. An important advancement in the assessment of the arterial baroreflex control of heart rate on clinical grounds has been represented by the development of baroreflex testing methodologies that are based on the computer analysis of the continuous linkage between beat-by-beat spontaneous fluctuations in blood pressure and heart period (calculated from R–R or the pulse interval) [7,8]. The main advantage of these techniques is that they permit a dynamic assessment of the cardiac component of the integrated baroreflex response, relying on a natural stimulus of physiological magnitude (i.e. spontaneous blood pressure increases and decreases at the current, prevailing levels of blood pressure, with which the body must cope in any physiological or pathological condition). This is possible with a complete non-invasive approach, without having to induce any pharmacological or mechanical disturbance from outside the cardiovascular system. Among the spontaneous techniques, the so-called ‘sequences method’ [9] has been widely employed in a variety of physiological and clinical studies [8,10]. This technique is based on computer scanning of systolic blood pressure (SBP) and R–R interval (or pulse interval) time series to identify sequences of spontaneously occurring consecutive beats in which SBP ramps (i.e. progressive SBP increases and decreases) are accompanied by directionally similar changes in the R–R interval (i.e. hypertensive/bradycardic and hypotensive/tachycardic sequences). The slope of these sequences is taken as a index of the sensitivity of the cardiac baroreceptor reflex, similar to the Oxford technique of vasoactive drug injection [11], and mainly reflects vagally mediated baroreceptor-cardiac responses [8]. In this issue of the journal, Johansson et al. [12] report, for the first time, that BRS, as estimated by the sequences technique, was an independent predictor of sudden death in patients suffering from chronic renal failure (CRF). This is a novel finding of clinical importance in as much as patients with CRF have an increased risk for ventricular dysrhythmias and a large fraction of cardiovascular mortality in these patients is attribuited to sudden death [13]. Several studies have shown that CRF is characterized by an impairment in baroreflex control of heart rate [14–17], which is reversible, in part, by therapy [14,17]. The mechanism(s) responsible for autonomic alterations in CRF have not yet completely clarified, and may involve central as well as peripheral neuropathy. Patients with end-stage CRF feature abnormal responses in either blood pressure or heart rate during dialysis, which are mediated by increased sympathetic and decreased parasympathetic activity [18–20]. Moreover, vagal nerve damage has been described in patients with end-stage renal failure [21]. The finding that a reduced BRS is predictive of sudden death in CRF patients fits well with the findings of several animal studies indicating that the presence of depressed vagal reflexes, as estimated by BRS, with concomitant sympathetic activation is associated with a greater risk for ventricular fibrillation during transient ischemia [22], and with clinical studies showing that a depressed BRS may play a role in the occurrence of malignant dysrhythmias long after myocardial infarction [23]. Hence, the finding by Johansson et al. [12] that a depressed BRS is predictive of sudden death in patients with CRF appears to be sound, based on both basic and clinical studies, although the small percentage of patients who died suddenly in their study (22 as opposed to the reported 60% in literature) prevents any definitive conclusions being made. In their study Johansson et al. [12] also report that the extent to which the arterial baroreflex is engaged in the control of heart rate is an independent predictor of all-cause mortality in CFR patients. In other words, the less time the arterial baroreflex is engaged in modulating heart period on a recurring basis by spontaneous increases and decreases in blood pressure, the greater is the risk of overall mortality in this class of patients. The magnitude of baroreflex engagement in the study of Johansson et al. [12] has been assessed by the so-called baroreflex effectiveness index (BEI). BEI is defined as the ratio between the number of SBP ramps followed by concomitant reflex changes in the R–R interval (i.e. the number of baroreflex sequences) and the overall number of SBP ramps detected in a given time period, also including the ramps that are not accompanied by directionally similar, reflex changes in R–R interval [24]. This index of baroreflex function has been proposed based on the observation that not every SBP ramp recorded in a given time period is invariably followed by a reflex R–R interval ramp, and has been interpreted as indicating that the baroreflex would not always be effective in driving the sinus node in response to spontaneous fluctuations in SBP [24]. BEI would furnish complementary information to the BRS on baroreflex control of heart rate [24]. BEI is surely related to arterial baroreflex functioning, as indicated by its striking reduction after surgical baroreceptor denervation in the experimental animal [24]. However, at variance with a BRS derived from the sequences method, which has been well validated on both experimental and clinical grounds, the extent to which BEI could reflect the actual effectiveness or, vice versa, an impaired effectiveness of the baroreflex control of heart rate has received little attention and still needs to be fully validated. Arterial baroreflex function is very complex and is strongly influenced by inputs from other reflexogenic areas and by nonbaroreflex mechanisms [25–27] in controlling the sinus node. The model underlying calculation of the BRS by the sequences and other spontaneous methods, as well as by the Oxford methods, hypothesizes that R–R interval changes are driven only by SBP variations through a simple baroreflex loop, and the influence of other factors, such as respiration, stimulation of cardiac receptors with vagal and sympathetic afferents and also nonbaroreflex mechanisms [28,29], although acknowledged, are usually not taken into account [28]. In spite of this simplification, evaluation of the BRS by different baroreflex testing methodologies provides information of paramount importance, both physiological and clinical. However, the arterial baroreflex does not modulate the sinus node only in a sequence-like fashion, but also by causing single beat changes in the R–R interval in response to single beat changes in blood pressure. Thus, the sequences method accounts for only a fraction of the baroreceptor reflex engagement that occurs over time [30]. Hence BRS, as estimated by the sequences method, may provide a reliable quantification of the ability of the baroreflex to modulate the sinus node rate in response to a specific stimulus, represented by blood pressure changes, in a given experimental condition, whereas BEI might not completely reflect the amount of time the baroreceptors are actually engaged in reflexly modulating the sinus node. The baroreceptor-like sequences reflect a specific stimulus–response relationship and are only partly the way by which the arterial baroreceptors modulate the sinoatrial node. This might explain why BEI was not predictive of sudden death in the study of Johansson et al. [12], whereas the BRS, which is a more specific estimate of baroreflex control of heart rate, was predictive. By contrast, because BEI is related in some way to baroreflex functioning [24], this might partly explain the predictive role of BEI on overall mortality, in as much as CRF patients suffered from concomitant diseases or metabolic derangements that may impair baroreflex functioning. However, the mechanism(s) responsible for the link between a reduced BEI and all-cause mortality in patients with CRF remain to be determined. At present, BEI could be very useful in assessing the robustness of BRS because BEI is related to the number of baroreflex sequences [12] and this may critically affect the reliability of BRS measurements by the sequences method in a given recording period [7]. To what extent the degree of baroreflex engagement, as estimated by BEI, could reflect the actual effectiveness of the arterial baroreflex in controlling heart rate, as as well its possible clinical utility as an independent baroreflex function index, needs to be tested in future experimental and clinical studies.