| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
| ||||||||||||||
| ||||||||||||||
|
|
|
|
||||||||||||
|
||||||||||||||
From the Neurovascular Research Laboratory, School of Kinesiology, The University of Western Ontario, London, Ontario.
Address correspondence and reprint requests to J. Kevin Shoemaker, PhD, Neurovascular Research Laboratory, School of Kinesiology, Room 3110 Thames Hall, The University of Western Ontario, London, ON N6A 3K7, Canada. Address e-mail to kshoemak{at}uwo.ca.
Reflex adjustments to physical and orthostatic stress include a complex series of neurogenic, cardiac, and vascular reactions that defend cardiac output and systemic vascular resistance, so that arterial blood pressure and cerebral perfusion are sustained and near normal levels. In addition, an array of metabolic, circulating, endothelial, and local pressure-dependent regulatory mechanisms contribute to changes in cerebrovascular resistance. The ability of cerebral blood vessels to adjust their caliber and sustain blood flow over a wide range of perfusion pressure is termed autoregulation and has been the focus of research for many decades (1–3). Without such systemic and cerebrovascular responses, hypotension and cerebral hypoperfusion would occur.
During surgery in anesthetized patients where normal physiology may be compromised, it is the role of anesthesiologists to sustain arterial blood pressure and cerebral perfusion at levels that are optimal for the circumstances. This objective often requires the enhancement of circulating blood volume with methods that dilute oxygen-carrying capacity. Yet the very practice of maintaining blood volume and cardiac output by hemodilution may be detrimental to cerebral function through reductions in oxygen delivery. The dangers of excessive hemodilution include poor neurologic (4,5) and other life-threatening outcomes after surgery (6). Several large observational (7–9) and experimental (10) studies have identified an association between very low intraoperative hematocrit levels (<30%), altered cerebral metabolism, risk of in-hospital mortality, or adverse neurological outcomes. Hemodilution challenges cerebrovascular perfusion.
Complex physiology is not regulated completely by simplistic strategies. The mechanism by which hemodilution affects cerebral blood flow (CBF) is perhaps as complex as the means by which CBF itself is sustained. One important question is the impact of hemodilution on cerebral autoregulation (CA). Further, if hemodilution impairs autoregulation, does it really matter under conditions where cardiac output is sustained under reflexive control?
In the current issue of Anesthesia & Analgesia, Ogawa et al. (11) test the hypothesis that changes in circulating blood volume with hemodilution affect CA in conscious humans. This was accomplished by (a) reducing central blood volume with two levels of lower body suction (a simulated orthostatic stress), and (b) infusing normal saline at two rates to achieve up to approximately 5% reduction in hematocrit. With the saline infusion, there was a concurrent increase in circulating blood volume, cardiac filling pressure, and cardiac output. Examination of CA was accomplished using the transfer function gain approach (12–14), whereby spontaneous fluctuations in arterial blood pressure are correlated to concurrent variations in CBF velocity at various oscillating frequencies. With this comprehensive approach, analysis of static and dynamic autoregulation can be investigated along with questions regarding the relationship of cardiac output to CBF and oxygen delivery. This study offers a comprehensive approach in studying CA within the context of central and cerebral hemodynamics. Moreover, it offers opportunities for discussion of several points that are relevant to those interested in perioperative care. These include 1) the method of spectral analysis to assess autoregulation, 2) what impaired autoregulation means in this context, and 3) how altered cerebrovascular control mixes with the altered central hemodynamics.
Spectral analysis methods have become popular for assessing cardiovascular oscillations, including cerebrovascular control. The relationship between periodic changes in arterial blood pressure and CBF velocity across a range of relevant oscillation frequencies has been reported in hundreds of papers since the early report by Birch et al. (14) in 1995. The method's popularity is due to 1) ease of use, 2) a wide range of applicability, 3) relatively low cost, 4) lack of need to induce arterial blood pressure changes but, rather, reliance on spontaneous fluctuations, and 5) the apparent diagnostic utility. This transfer function approach relies solely on the relationship between spontaneous fluctuations in arterial blood pressure and mean flow velocity in the middle cerebral artery, this artery being used as a generalized indicator of cerebrovascular control. With intact autoregulation, fluctuations in arterial blood pressure are not reflected in blood flow velocity (i.e., low transfer gain) and with loss of autoregulation there is increased transfer of arterial blood pressure to the CBF velocity. Despite these benefits, there is concern that information about variables such as vasomotor tone (15,16) are not included in the determination of coherence between arterial blood pressure and mean flow velocity. Also, the magnitude of impairment is difficult to determine from transfer function methods. Certainly, these additional features explain some of the challenges in the interpretation of spectral analytical results [see (16) for example].
The current study in volunteers by Ogawa et al. created a relatively small hemodilution with hematocrit changes of <5%. Nonetheless, this moderate level of hemodilution was associated with a higher transfer function gain across a range of frequencies in the arterial blood pressure–blood flow velocity frequency spectrum, i.e., impaired autoregulation. In time-domain and steady-state analysis, saline infusions produced a 20% increase in cardiac output together with a10% increase in CBF with no change in arterial blood pressure. Higher cerebral flow velocity at the same arterial blood pressure suggests that either static autoregulation was impaired or that the pressure-flow autoregulation curve was adjusted upwards. The spectral-based methods confirmed the alteration of autoregulation. But what does "impaired" autoregulation mean in this context? Does it mean that cerebral perfusion is in jeopardy?
A methodological constraint of spectral analysis is that it cannot provide information on the position of the autoregulatory curve. Neither can it describe the range of arterial blood pressures over which autoregulation is affected or effective. In fact, this pressure range has never been assessed adequately in conscious humans and a "gold standard" method for the assessment of autoregulation has not been identified for human investigations. Nonetheless, comparisons across methods indicate that the spectral gain features do reflect accurately the directional change in autoregulation (2). Also, by itself, this index of CA does not provide a mechanistic basis for the change in cerebrovascular control. From this perspective, an attractive feature of the Ogawa et al. study is the comprehensiveness of data acquisition allowing the reader to examine the relationships between central and cerebral vasomotor control.
From Ogawa et al.'s data, cerebral hypoperfusion does not seem to be the problem in this hypervolemic hemodilution condition despite altered autoregulation. Rather, it is the reduction in oxygen content (but not delivery) that appears to be the more potent physiological challenge. Also, the maintenance of arterial blood pressure and cardiac output suggests that altered autoregulation is not due to pressure-dependent mechanisms. Using cerebral vascular resistance as the index, the authors indicate that vasodilation occurred in the current study. This dilation likely was related to altered cerebral metabolism as oxygen content was reduced. In addition, the authors discuss the potential for a baroreflex-medicated reduction in overall sympathetic outflow as a result of hemodilution causing cardiac output to increase. The altered sympathetic reflex outflow may affect cerebrovascular tissue and the autoregulatory response (17). In conscious healthy humans, sympathetic innervation appears to enhance autoregulatory control, at least when arterial blood pressure is high (18). Vasodilation, including that resulting from a reduction in sympathetic outflow, may alter the autoregulatory response. There is an inverse relationship between cerebrovascular contractile state and the extent of the autoregulatory response (2). The reason is likely related to (a) the diminished potential for dilated vessels to dilate further in response to a decrease in pressure, or (b) a greater difficulty to constrict when arterial blood pressure increases due to heightened competition from local metabolic dilators. The impaired autoregulatory response in the current study appears to be related to the indirect effects of hemodilution on the complex factors that affect cerebrovascular tone, rather than on the pressure-dependent nature of these vessels, per se. This study indicates that reduced oxygen transport provides a consistent challenge to multiple cerebrovascular control features, even with moderate degrees of hemodilution. Therefore, hypervolemia with hemodilution may erode CA through multiple mechanisms.
A major challenge in the literature on CA is putting the cerebrovascular response in the context of systemic hemodynamics. For example, to what extent is CBF regulated by cardiac output? Ogawa et al. (19) report a relationship that is weaker than previously reported. Differences between studies regarding the coupling of cerebral to systemic hemodynamics remain to be reconciled. Regardless, the "link" between cardiac output and CBF is arterial blood pressure as it is impacted by total blood flow. Conditions that modify cardiac output with little direct impact on cerebral metabolism or arterial blood pressure, such as postural shifts in healthy individuals, normally have little impact on CBF velocity. Thus, if autoregulation is intact, one would expect a weak relationship between total flow and CBF (assuming local cerebrovascular control factors remain constant). In contrast, events that impair CA or affect a vasodilatory response in the brain should improve the relationship between cardiac output and CBF. Unfortunately, it is not clear from the current results whether this correlation differed between the period of hypovolemia (induced by lower body suction) when autoregulation was unchanged from a control test, and hypervolemia (saline infusion) when it was impaired. Future studies might examine how "impaired" autoregulation directly affects the relationship between central and CBFs. Nonetheless, this study supports the idea that cerebral perfusion is poorly associated with total circulating blood flow even when autoregulation is impaired.
In summary, the inducement of hypervolemia with hemodilution, while a clinically relevant scenario, is a complex challenge for cerebral perfusion, as it alters oxygen delivery and reflex autonomic outflow while enhancing overall blood flow and pressure. As Ogawa et al. acknowledge, it is difficult to isolate independent effects of acute changes in blood volume without altering variables that exert additional influence on the measurements of CA. Furthermore, even alterations in autoregulation are difficult to interpret in humans because of limitations to the methods available. Thus, it is difficult to comprehend the integrated effect of altered central blood volume with hemodilution using the limited tools available for noninvasive studies in conscious humans. Nonetheless, this study raises the important issue that it is the integrated response that may be most informative to improved patient outcomes.
 |
Footnotes |
|---|
|
REFERENCES |
|---|
 Top REFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
