| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2000;69:983-985
© 2000 The Society of Thoracic Surgeons
a Department of Anesthesiology, The University of Texas Medical Branch, Galveston, Texas, USA
Address reprint requests to Dr Prough, Department of Anesthesiology, The University of Texas Medical Branch, 301 University Blvd, Galveston, TX 77555-0591
e-mail: dsprough{at}utmb.edu
Unlike most circulatory beds, the cerebral vasculature responds to increases in perfusion pressure by increasing cerebrovascular resistance and responds to decreases in perfusion pressure by decreasing cerebrovascular resistance. As a consequence, cerebral blood flow (CBF) changes much less than perfusion pressure. Classically, the pressure autoregulatory curve is described as having a lower limit of autoregulation, below which the slope of the relationship between CBF and perfusion pressure is steeper, and an upper limit of autoregulation, above which the slope also is steeper [1]. The relatively flat plateau between the lower and upper limits offers partial protection against cerebral ischemia as perfusion pressure declines and limits cerebral hyperperfusion as perfusion pressure increases.
Both basic scientists and clinical investigators have studied autoregulatory phenomena extensively since Lassen [1] first popularized the concept in 1959. Of particular interest have been impairments in autoregulation secondary to physiologic, pharmacologic, and pathophysiologic factors, including hypercapnia [2], most volatile anesthetics [3], traumatic brain injury [4–6], and cerebral ischemia [7, 8]. Impaired cerebral pressure autoregulation is clinically important if it increases the risk of secondary cerebral ischemia during hypotension or if it increases the risk of adverse consequences of cerebral hyperperfusion. Inappropriately increased CBF could increase intracranial pressure in patients who have decreased intracranial compliance or could increase the proportion of arterial emboli delivered to the brain rather than the systemic circulation.
In this issue of The Annals of Thoracic Surgery, Sungurtekin and colleagues [9] report that cerebral microembolization during normothermic cardiopulmonary bypass impairs cerebral pressure autoregulation. This report builds on previous publications [10, 11] from the same laboratory that describe a similar experimental model of embolic injury associated with cardiac surgical procedures. Careful reading of this article prompts two questions: (1) do the authors establish a convincing case that this model of microembolization impairs autoregulation during cardiopulmonary bypass, and (2) what are the clinical implications of these observations?
Do these data convincingly establish the case that microembolization during cardiopulmonary bypass impairs autoregulation? Certainly the data are provocative, but they fall short of providing compelling evidence. The protocol for autoregulatory testing merits praise because of the randomized sequence of pressure changes, although other aspects of the experimental design and methods prompt caution. Few previous studies of autoregulation have used a randomized sequence of perfusion pressures to investigate the relationship between perfusion pressure and CBF. Typical methods to reduce cerebral perfusion pressure include progressive, incremental hemorrhage [4] and rapid deflation of inflated thigh cuffs [3]. Cerebral perfusion pressure is typically increased using phenylephrine [3]. The technical aspects of autoregulatory testing are further complicated by the fact that most quantitative methods of measuring CBF (eg, the Kety-Schmidt technique, xenon 133 clearance, and labeled microspheres) require 10 or 15 minutes for completion and are suitable for only a limited number of determinations.
In the present study, Sungurtekin and colleagues [9] varied pump flow and phenylephrine infusion to achieve a randomly ordered sequence of mean arterial pressures (MAP) ranging from 40 to 85 mm Hg, and used continuous laser Doppler flowmetry to assess changes in flow. For comparisons between various MAPs, the authors chose as baselines the flow velocities before and after embolization at a MAP of 60 mm Hg. Because laser Doppler flowmetry is not quantitative, flow velocity is expressed in arbitrary units and changes in flow usually are expressed as percents of baseline values. Before embolization, the mean slope of regression lines drawn through the data points for individual dogs was 0.21 ± 0.74 velocity units/mm Hg; after embolization the mean slope was 1.31 ± 0.87 velocity units/mm Hg. Before embolization, flow velocity increased 39% as MAP varied from 40 to 85 mm Hg; after embolization, flow velocity increased 94% over the same range. From these differences, the authors concluded that autoregulation was impaired by embolization during cardiopulmonary bypass.
That conclusion necessitates close inspection of the data, which demonstrate that the insult was not homogeneous among animals. Figures 1 and 2 (preembolization and postembolization) display autoregulatory curves from individual animals. In Figure 1, in 5 animals autoregulation appears to have been intact, (ie, flow velocity changes gradually over the range of pressures tested). However, 1 animal (asterisk) had markedly increased flow velocities at pressures above 60 mm Hg; 2 animals (open and closed squares) had markedly decreased flow velocities at pressures below 60 mm Hg. In conventional terms used to describe autoregulatory curves, the first animal had a left-shifted upper limit of autoregulation and the other 2 had right-shifted lower limits of autoregulation. These responses cannot be accurately represented by a straight regression line over the range of pressures tested.
After embolization, the animals demonstrated a wide variety of responses (Fig 2). One animal (open diamonds) had a markedly left-shifted upper limit with an apparent increase in flow velocity exceeding 250% between 70 and 85 mm Hg. Below 70 mm Hg, flow velocity was relatively stable. One animal (open squares) showed a right shift in the lower limit of autoregulation to 60 mm Hg, with severely reduced velocities below that level and stable velocities above. One animal (open circles) had an abrupt decrease in velocity below 50 mm Hg but stable velocities above 50 mm Hg. Five of 8 animals appear to have had autoregulatory responses that are little different from the 5 animals in which preembolization responses were intact. The text does not specify if the symbols in the two figures designate the same animals. If they do, autoregulation actually improved in 2 animals between the pre- and postembolization testing.
Regardless, these data suggest that pressure autoregulation is variable both before and after embolization during normothermic bypass, with some animals showing impaired autoregulation over at least part of the range under both conditions. Moreover, because the entire range of pressure responses is not impaired in any of the animals before or after bypass, it may be misleading to compare the slopes of straight regression lines drawn through the data. For example, a straight regression line drawn through the data points of the animal represented by the open triangles in Figure 2 would have a substantial slope, but the entire slope would be attributable to the 85 mm Hg data point.
An additional methodologic issue is the lack of a concurrent control group with autoregulation determined both shortly after initiating cardiopulmonary bypass and approximately 30 minutes later without the intervening embolic insult. This protocol required a substantial duration of cardiopulmonary bypass. After an undefined interval during which stable conditions were established after initiating bypass, each of five MAPs was maintained for 15 minutes or more before flow velocity measurements were recorded. After embolization, another 30 minutes was permitted for stabilization before the second autoregulatory testing sequence. Therefore, the second set of measurements began a minimum of 105 minutes after the first set and the final measurement was not complete for a minimum of 180 minutes after initiating bypass. The authors argue, based on previous animal and clinical studies, that the duration of bypass per se should not have impaired autoregulation, but the heterogeneity of the individual animal’s responses, both before and after embolization, argues strongly that control responses should have been measured, not assumed.
Appropriate controls are particularly important because previous studies have used different measurements of global and regional CBF. Extrapolation from Kety-Schmidt or microsphere measurements to laser Doppler measurements may not be appropriate. For example, in previous studies [10, 11], the authors of the present report characterized the cerebral embolization model using fluorescent microspheres to measure regional CBF after embolization of 67 µm spheres (versus 97 µm spheres in the present study).
What inferences from this animal study might prompt appropriate clinical investigation and perhaps ultimately a change in the management of cardiopulmonary bypass? Impaired autoregulation could prompt either (or both) of two clinical strategies. First, if CBF declines excessively at lower perfusion pressures, one reasonable response would be to maintain higher pressures to limit the risk of secondary cerebral ischemia. Such an approach is commonly used in caring for patients with severe head injuries [12, 13]. Second, if excessive CBF poses a risk because of a left-shifted upper limit of autoregulation, avoidance of higher pressures could be advantageous. Unfortunately, neither strategy can be proposed on the basis of these data. A fraction of the animals, some before and some after bypass, appeared vulnerable to decreases in MAP. A smaller fraction (1 of 8 before or after bypass) appeared vulnerable to increases in MAP. If these results were extrapolated to humans undergoing cardiac operation, those at risk from high perfusion pressures would have suffered an embolic insult early in the course of the surgical procedure—perhaps during cannulation of an atherosclerotic aorta. They would have a left-shifted upper limit of autoregulation, and if perfusion pressures were excessive, would be at risk for greater transmission of embolic debris to the cerebral circulation at the conclusion of bypass. Unfortunately, at present we do not know if autoregulation is impaired in as heterogeneous a fashion in humans undergoing cardiopulmonary bypass, nor can we routinely distinguish patients with right-shifted lower limits from those with left-shifted upper limits at critical intervals during the surgical procedure. Perhaps the answer lies in the development of quantitative, noninvasive cerebral circulatory monitors that can be applied routinely during cardiac operations.
In summary, Sungurtekin and colleagues [9] performed a provocative study of considerable heuristic value. Additional research is necessary to ensure that prolonged cardiopulmonary bypass has no direct effects on cerebral pressure autoregulation and to determine the mechanisms of the highly variable autoregulatory responses both before and after bypass. Ideally, that research will lead to clinical research that permits recognition of those cardiac surgical patients at risk for cerebral hypoperfusion with reduced arterial pressure or disproportionate cerebral embolization with increased arterial pressure.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
