JOURNAL HOME CME HOME THIS MONTH PAST ISSUES ETOC COLLECTIONS AUTHORS REVIEWERS EDITORIAL BOARD FEEDBACK RSS HELP
		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Web of Science (23) Citing Articles via Google Scholar Google Scholar Articles by Piechnik, S. K. Articles by Pickard, J. D. Search for Related Content PubMed PubMed Citation Articles by Piechnik, S. K. Articles by Pickard, J. D.

---

NEUROSURGICAL ANESTHESIA

The Continuous Assessment of Cerebrovascular Reactivity: A Validation of the Method in Healthy Volunteers

Wolfson Brain Imaging Centre, Cambridge Medical Research Council’s Centre for Brain Repair and Academic Neurosurgery Unit, Addenbrooke’s Hospital, Cambridge, United Kingdom

Address correspondence and reprint requests to Stefan Piechnik, MScEE, Academic Neurosurgery Unit, PO Box 167, Addenbrooke’s Hospital, Hills Rd., Cambridge CB2 2QQ, UK. Address e-mail to S.K.Piechnik{at}iname.com

	Abstract

Top Abstract Introduction Methods Results Discussion References

 
Using transcranial Doppler ultrasonography, we investigated the moving correlation between slow waves in arterial blood pressure (ABP) and blood flow velocity (FV) at different levels of cerebrovascular vasodilation provoked by changing PETCO₂. Fourteen healthy volunteers were examined. The FV in middle cerebral arteries, PETCO₂, and ABP were recorded during normocapnia, hypercapnia, and hypocapnia. The moving correlation coefficients between ABP and mean FV (FVm) or systolic FV (FVs) during spontaneous fluctuations in ABP were calculated for 3-min epochs and averaged for each investigation, thus yielding the mean index (Mx) and systolic index (Sx). As a reference method, Aaslid’s cuff tests were performed to obtain the rate of regulation (RoR). RoR, Mx, and Sx significantly depended on PETCO₂ (analysis of variance, P < 0.00001). At high PETCO₂, cerebrovascular reactivity was disturbed as reflected in RoR values of <0.17/s for all volunteers and increased values of Mx (>0.4 in 86% of volunteers) and Sx (>0.2 in 79% of volunteers). Overall, there was a reasonably good correlation of both Mx and Sx with RoR (R² = 0.65 and 0.58, respectively).

Implications: Indices derived from the correlation between spontaneous fluctuations of blood flow velocity wave form and arterial blood pressure may be used for the noninvasive continuous monitoring of cerebrovascular reactivity.

	Introduction

Top Abstract Introduction Methods Results Discussion References

 
An ideal brain monitoring system should not only detect cerebral ischemia but also identify defective cerebrovascular reactivity, which renders the brain more susceptible to such insults. Most tests of cerebrovascular reactivity are based on the detection of changes in cerebral blood flow (CBF) to an evoked change in arterial blood pressure (ABP) (1–3). Such tests require the attention of many medical staff members and can be repeated only with limited frequency. More recently, methods suitable for the continuous assessment of vasodilatory responses, which use spontaneous variations in cerebral perfusion pressure (CPP) have been described (4–7). They do not need any additional clinical maneuvers; hence, they are less laborious and may be used in a wider range of patients, including those particularly unstable or, because of other medical reasons, unavailable for challenges required by standard tests. These methods have already proved to be good predictors of outcome after head injury and sensitive indices of an optimal CPP in sedated, paralyzed, and ventilated patients (6,7). They are potentially useful in situations in which, rather than a snapshot image, a change in cerebral autoregulation in response to change of ventilation pattern, osmotherapy, or hypertension-hypervolemia therapy is to be monitored.

Our aim was to test noninvasively in healthy volunteers, whether indices of the correlation between slow changes in systolic (systolic index [Sx]) and mean (mean index [Mx]) flow velocities (FV) and slow waves seen in ABP, depend on the cerebrovascular dilation caused by graded changes in the PETCO₂. We also investigated the level of agreement between these indices and the so-called rate of regulation (RoR) calculated from the hemodynamic response observed with transcranial Doppler (TCD) ultrasonography in response to a rapid, short-term decrease in ABP after sudden deflation of leg cuffs (1).

	Methods

Top Abstract Introduction Methods Results Discussion References

 
Volunteers and Study Set-Up
The experiment was approved by our local ethics committee. Informed consent was obtained from each of the 14 healthy volunteers, 11 male and 3 female. The age range was 20–44 yr (mean 23).

Each subject sat in a reclining chair and was positioned with his or her upper body at approximately 45° to the horizontal. TCD probes (Neuroguard, 2 MHz; Medasonics, Fremont, CA) were placed over the temporal window, just above the zygomatic arch, on both sides. In all subjects, the middle cerebral artery was insonated at a depth of 50 mm bilaterally, with only a small variation between subjects. The probe position was fixed with the aid of a headband that could be fastened tightly around the subject’s head.

ABP was measured noninvasively (Finapres, Ohmeda 2300; BOC Group Inc, Englewood, CO) via a miniature cuff placed around the middle finger of the left hand. Subjects were asked to keep this hand immobile at the heart level to prevent artifacts from being recorded.

Large blood pressure cuffs were fastened around the upper thighs of the subject. These were conveniently inflated from a hospital main oxygen supply via a standard anesthetic valve and rapidly deflated by removing a rubber plug from a large-diameter tube connecting the two cuffs. Pressure in the cuff was checked using a sphygmomanometer connected to the inlet pipe. The experiments were performed in a large ventilated laboratory; hence, the limited amount of oxygen released during cuff deflation was unlikely to have affected the results.

PETCO₂ was measured using an anesthetic face mask that was attached via a small diameter tube to a CO₂ monitor (Normocap CO2 Monitor; Datex Instrumentarium Corp., Helsinki, Finland). PETCO₂ was raised by attaching a plastic tube to the face mask worn by the subject. The tube was 3.12-m long with a 25-mm diameter. Thus, the subjects respiratory dead space was increased by 1.5L.

Data Collection
The wave forms of ABP, TCD, and PETCO₂ were sampled (50 Hz) and digitized (12 bits) using an analog-to-digital converter (DT 2814; Data Translation, Marlborough, CA) fitted in a personal computer, running our own software (8). The digital signals were then analyzed off-line using software devised for our cerebrovascular laboratory (8) and for intensive care data analysis (9).

Study Protocol
At normocapnia, the leg cuffs were inflated to approximately 50 mm Hg above systolic ABP for 1 min, 30 s with the legs raised. For the last 30 s of inflation, the legs were slowly lowered and rested until the readings were stable. The cuffs were then rapidly deflated. At hypercapnia, the tube was attached to the face mask until a steady PETCO₂ level was reached, and then the cuff test was performed. At hypocapnia, the subject was asked to start hyperventilating 30 s after the cuffs were inflated, because it was found that low PETCO₂ was reached quite rapidly. The cuff test was repeated twice for each PETCO₂ level, and the results were averaged.

Calculations
Average time values of ABP were calculated by averaging the ABP samples for each subsequent 5-s interval. Mean FV (FVm) and systolic FV (FVs) were averaged over the same 5-s periods. To reduce the influence of noise, FVm was calculated after careful spectral filtration (finite response filter, Gaussian shape). Pulse and respiratory waves were filtered out (>99.5% and >90%, respectively), leaving only slow oscillations (20-s to 2-min period) for further analysis (Fig. 1A). For simplicity, the filtering may be omitted without significantly influencing the results.

View larger version (38K): [in this window] [in a new window]   Figure 1. A, The steps of signal processing used to obtain subsequent samples of Mx. B, Two examples of linear regression plots with positive and negative values of Mx corresponding to nonautoregulating and autoregulating system. C, Example of time trends of stable recordings of physiological parameters during normo-, hyper-, and hypocapnia. Sample-to-sample Mx and Sx fluctuate significantly. They become meaningful only after averaging. ABPm = mean arterial blood pressure, FVm = middle cerebral artery blood flow velocity, Mx = mean index, Sx = systolic index.

The calculation of RoR has been described in full elsewhere (1,3). In short, cerebrovascular resistance (CVR) was calculated as the momentary value of ABP divided by FV. After cuff deflation, after an initial sharp decrease in ABP, autoregulating vessels start to dilate, progressively reducing CVR. To obtain the RoR value, the speed of the decrease of CVR is divided by an amount of initial drop in ABP produced by cuff deflation, both standardized by the baseline values of CVR and ABP, respectively.

All statistical analysis was performed using STATGRAPHICS Plus for Windows (ver. 2.0, Statistical Graphics Corporation, Rockville, MD). The normality of the distribution of calculated indices was confirmed, and one-way analysis of variance (ANOVA) was subsequently used.

	Results

Top Abstract Introduction Methods Results Discussion References

 
An example of the time trends of the recorded variables and samples of Mx and Sx is presented in Figure 1C.

Hypotheses of nonnormal distribution of RoR, Mx, and Sx at different levels of PETCO₂ were tested and found to be nonsignificant (P > 0.1). Variances of RoR, Mx, and Sx were not significantly different within separate PETCO₂ groups. Therefore, ANOVA could be applied. ANOVA plots for RoR, Mx, and Sx at each of the three PETCO₂ levels demonstrated that these three indices of cerebrovascular reactivity change with PETCO₂ (Fig. 2). The average values of all the recorded variables for the three PETCO₂ levels are presented in Table 1. RoR, Sx, Mx, and FVm were significantly related to the degree of cerebral vasodilation, whereas ABP did not change significantly. RoR decreased with cerebral vasodilation, indicating a decrease in the speed of the autoregulatory response. Both Sx and Mx increased with vasodilation, indicating a stronger passive association between slow waves in FV and ABP when vascular reactivity weakened.

View larger version (29K): [in this window] [in a new window]   Figure 2. Analysis of variance (mean value and 95% confidence limits) of A) RoR, B) Sx, and C) Mx versus level of carbon dioxide concentration. Stars denote statistically homogeneous groups (P < 0.05). RoR = rate of regulation, Mx = mean index, Sx = systolic index.

However, the statistical convergence of RoR was better than that of Sx and Mx; the confidence limits for the mean values were clearly wider for Mx and Sx than for RoR (Fig. 2). This is also expressed by a lower ANOVA F value for Sx and Mx than RoR (Table 1).

The overall relationship between RoR and Mx demonstrated a good degree of accordance, with a R2 of 0.65 (Fig. 3). The corresponding R2 value for Sx was slightly lower at 0.58.

View larger version (25K): [in this window] [in a new window]   Figure 3. Scatterplot of the rate of regulation (RoR) versus mean index (Mx). The line indicates best-fit linear model.

The right-to-left differences were not significant for any of the measured variables (P > 0.05).

	Discussion

Top Abstract Introduction Methods Results Discussion References

 
Cerebral autoregulation is a complex phenomenon involving an integrated reaction of different sections of cerebrovascular bed (10). The interaction between the autoregulatory response and CO₂-induced dilation of cerebral vessels, first quantified by Harper and Glass (11), mainly involves the smaller arterial vessels. In this study, the sequence of normo-, hypo-, and hyperventilation were used as the standard vasostimulation to compare noninvasive continuous indices of cerebrovascular reactivity, Mx and Sx, with Aaslid’s cuff test.

Aaslid’s leg cuff test relies on a short-term evoked decrease in ABP and was introduced in the mid 1980s. It is fully noninvasive, has been validated in volunteers (1,3), and subsequently used in clinical practice (2,12). New indices of cerebrovascular reactivity have been proposed by Czosnyka et al. (6) and subsequently used to study cerebral autoregulation in patients with head injuries. They have been proven to correlate with ICP, CPP, and outcome after head injury (6). Although the indices have been shown to correlate with the transient hyperemic response test (3), they have never been compared with Aaslid’s method.

The interpretation of Sx or Mx based on their definition as the correlation coefficients between slow fluctuations in systolic or mean Fv and ABP is straightforward. Positive correlation reflects the passive response of CBF to a change in ABP; negative correlation denotes active regulation. This was confirmed by the positive association of a low rate of regulation with positive values of Mx and Sx recorded in volunteers during hypercapnia. However, such an explanation oversimplifies the complex nature of CBF control, as vascular reactivity is not an all-or-none phenomenon. It may be measured and graded using such indices as RoR. Therefore, the real interpretation of Mx and Sx is more complicated. Both indices depend upon, and hence indirectly describe, a phase shift between slow oscillations of ABP and the Fv wave form (13), a concept similar to that described by Steinmeier et al. (4). A high positive correlation denotes zero phase shift and a fully passive response of the blood flow regulation system. In contrast, a strong and very quick vasodilatory response would dynamically produce 180° phase shift and result in a high negative correlation (Fig. 1B). From this point of view, both Mx and Sx may be understood as indices expressing vascular reactivity not only continuously in time but also in a continuous manner over the whole range of its strength.

Figure 3 indicates that Mx and RoR react similarly to vasodilation (11). Overall, the correlation between Mx, Sx, and RoR is moderate and does not justify the use of these methods as a replacement for each other. The value of R2 is probably protocol-dependent and cannot be applied in different clinical situations. The leg-cuff test demonstrated greater statistical convergence to changing PETCO₂ than Sx and Mx. This is not surprising, because the deflation of leg cuffs produces a clear decrease in ABP of 10–15 mm Hg. The slow oscillations of ABP have a limited magnitude of approximately 5 mm Hg, and therefore their hemodynamic response may not be immediately readable. It may also happen that the patient, aroused from external stimulation, will respond with simultaneous increases in ABP and FV, resulting in false positive correlation. Sx and Mx are time-dependent, fluctuating from positive to negative values (Fig. 1C). To minimize these effects, averaging for a period of at least 15 min is suggested. Normally during this period, there is enough slow wave activity to produce interpretable Mx and Sx values, even during relatively stable monitoring periods. The temporal effectiveness of both methods is therefore comparable, as it is difficult to prepare and perform a leg cuff test in much less than a quarter of an hour. The cuff test demands attention of medical staff and causes significant discomfort to the patient. In contrast, the calculation of the correlation indices may be left unattended to a computer, to be performed continuously without causing any additional stress.

When first defined (6), the Mx and Sx indices were calculated using CPP, not ABP. Obviously, in volunteers or in cases when there is no indication for ICP monitoring, ABP must be used instead of CPP. Although initially undertaken with some reservation, the results of this study prove that this modification does not invalidate sensitivity of Mx and Sx to CO₂-manipulated cerebral vasodilatory capacity.

We demonstrated that noninvasive Mx and Sx are closely correlated with each other (R2 = 0.80). Therefore, there is no obvious reason to use both. However, in clinical practice, particularly when ICP may be elevated, a divergence between the behavior of systolic and mean FV has been documented (14). A similar situation has been demonstrated in experimental models of intracranial hypertension (15) and arterial hypotension (16). The discrepancy between indices when Sx indicates residual reactivity and Mx disturbed autoregulation have been observed (6) and demonstrated with modeling simulations (17). Such divergence signifies that the cerebral autoregulation remains in a "gray zone" when every effort should be made to improve the ability of the cerebrovascular bed to combat further secondary ischemic brain insults.

The absence of any discrepancy in this study may reflect the use of noninvasive ABP rather than CPP. Alternatively, the current experiment examined only healthy volunteers. Values in this population may differ from those with brain injury (6,14–16). Therefore, until this matter is resolved, we would advocate the use of both indices, which potentially allows the detection of moderately impaired cerebrovascular reactivity as indicated by the presence of divergent behavior between Mx and Sx.

In summary, the noninvasive indices, Mx and Sx, adapted from invasive studies, have been shown to respond to graded vascular dilation induced by normo-, hypo- and hyperventilation. Their reliability is not as good as RoR obtained using Asslid’s leg cuff tests. However, the continuous character and more minor disturbance to the patient may render these indices useful when continuous or ambulatory monitoring of cerebrovascular reactivity is needed.

	Footnotes

 
SKP is a recipient of the "Fees Studentship" from the Cambridge Overseas Trust and the "Overseas Research Studentship" from the Committee of Vice-Chancellors and Principals of the Universities of United Kingdom.

	References

Top Abstract Introduction Methods Results Discussion References

 

1. Aaslid R, Lindegaard KF, Sorteberg W, Nornes H. Cerebral autoregulation dynamics in humans. Stroke 1989;20:45–52.[Abstract/Free Full Text]
2. Strebel S, Lam AM, Matta B, et al. Dynamic and static cerebral autoregulation during isoflurane, desflurane and propofol anaesthesia. Anesthesiology 1995;83:66–76.[Web of Science][Medline]
3. Smielewski P, Czosnyka M, Kirkpatrick P, Pickard JD. Evaluation of transient hyperemic response test in head injured patients. J Neurosurg 1997;86:773–8.[Web of Science][Medline]
4. Steinmeier R, Bauhuf C, Hubner U, et al. Slow rhythmic oscillations of blood pressure, intracranial pressure, microcirculation, and cerebral oxygenation: dynamic interrelation and time course in humans. Stroke 1996;27:2236–43.[Abstract/Free Full Text]
5. Zhang R, Zuckerman JH, Giller CA, Levine BD. Transfer function analysis of dynamic cerebral autoregulation in humans. Am J Physiol 1998;274:H233–41.[Abstract/Free Full Text]
6. Czosnyka M, Smielewski P, Kirkpatrick P, et al. Monitoring of cerebral autoregulation in head-injured patients. Stroke 1996;27:829–34.
7. Czosnyka M, Smielewski P, Kirkpatrick P, et al. Continuous assessment of cerebral vasomotor reactivity in head injury. Neurosurgery 1997;41:11–9.[Web of Science][Medline]
8. Smielewski P, Czosnyka M, Zabolotny W, et al. A computing system for the clinical and experimental investigation of cerebrovascular reactivity. Int J Clin Monit Comput 1997;14:185–8.[Web of Science][Medline]
9. Czosnyka M, Whitehouse H, Smielewski P, et al. Computer supported multimodal monitoring in neuro intensive care. Comput 1994;11:223–32.
10. Kontos HA, Wei EP, Navari RM, et al. Responses of cerebral arteries and arterioles to acute hypotension and hypertension. Am J Physiol 1978;234:H371–83.[Abstract/Free Full Text]
11. Harper M, Glass HI. Effect of alterations in the carbon dioxide tension on the blood flow through the cerebral cortex at normal and low blood pressures. Neurosurg Psychiatry 1965;28:449–52.
12. Steiger HJ, Aaslid R, Stooss R, Seiler RW. Transcranial Doppler monitoring in head injury: relations between type of injury, flow velocities, vasoreactivity and outcome. Neurosurgery 1994;34:79–86.[Web of Science][Medline]
13. Czosnyka M, Smielewski P, Richards HK, et al. Continuous monitoring of cerebral autoregulation in head injury. In: Klingelhofer J, ed. New trends in cerebral hemodynamics and neurosonology. Amsterdam:Elsevier Science BV, 1997:632–7.
14. Czosnyka M, Guazzo E, Iyer V, et al. Testing of cerebral autoregulation in head injury by waveform analysis of blood flow velocity and cerebral perfusion pressure. Acta Neurochir Suppl Wien 1994;60:468–71.[Medline]
15. Ungersbock K, Tenckhoff D, Heimann A, et al. Transcranial Doppler and cortical microcirculation at increased intracranial pressure and during the Cushing response: an experimental study on rabbits. Neurosurgery 1995;36:147–57.[Web of Science][Medline]
16. Nelson RJ, Czosnyka M, Pickard JD, et al. Experimental aspects of cerebrospinal haemodynamics: the relationship between blood flow velocity waveform and cerebral autoregulation. Neurosurgery 1992;31:705–10.[Web of Science][Medline]
17. Piechnik S, Czosnyka M, Pickard JD. Indices for decreased cerebral blood flow control: a modelling study. Acta Neurochir Suppl Wien 1998;71:269–71.[Medline]

This article has been cited by other articles:

				J. K. Lee, K. K. Kibler, P. B. Benni, R. B. Easley, M. Czosnyka, P. Smielewski, R. C. Koehler, D. H. Shaffner, and K. M. Brady Cerebrovascular Reactivity Measured by Near-Infrared Spectroscopy Stroke, May 1, 2009; 40(5): 1820 - 1826. [Abstract] [Full Text] [PDF]

				M. Czosnyka, P. Smielewski, A. Lavinio, J. D. Pickard, and R. Panerai An Assessment of Dynamic Autoregulation from Spontaneous Fluctuations of Cerebral Blood Flow Velocity: A Comparison of Two Models, Index of Autoregulation and Mean Flow Index Anesth. Analg., January 1, 2008; 106(1): 234 - 239. [Abstract] [Full Text] [PDF]

				A. Lavinio, E. A. Schmidt, C. Haubrich, P. Smielewski, J. D. Pickard, and M. Czosnyka Noninvasive Evaluation of Dynamic Cerebrovascular Autoregulation Using Finapres Plethysmograph and Transcranial Doppler Stroke, February 1, 2007; 38(2): 402 - 404. [Abstract] [Full Text] [PDF]

				M. Latka, M. Turalska, M. Glaubic-Latka, W. Kolodziej, D. Latka, and B. J. West Phase dynamics in cerebral autoregulation Am J Physiol Heart Circ Physiol, November 1, 2005; 289(5): H2272 - H2279. [Abstract] [Full Text] [PDF]

				M. Reinhard, M. Roth, T. Muller, B. Guschlbauer, J. Timmer, M. Czosnyka, and A. Hetzel Effect of Carotid Endarterectomy or Stenting on Impairment of Dynamic Cerebral Autoregulation Stroke, June 1, 2004; 35(6): 1381 - 1387. [Abstract] [Full Text] [PDF]

				M. Soehle, M. Czosnyka, J. D. Pickard, and P. J. Kirkpatrick Continuous Assessment of Cerebral Autoregulation in Subarachnoid Hemorrhage Anesth. Analg., April 1, 2004; 98(4): 1133 - 1139. [Abstract] [Full Text] [PDF]

				M. Reinhard, M. Roth, T. Muller, M. Czosnyka, J. Timmer, and A. Hetzel Cerebral Autoregulation in Carotid Artery Occlusive Disease Assessed From Spontaneous Blood Pressure Fluctuations by the Correlation Coefficient Index Stroke, September 1, 2003; 34(9): 2138 - 2144. [Abstract] [Full Text] [PDF]

				E W Lang, J Lagopoulos, J Griffith, K Yip, A Yam, Y Mudaliar, H M Mehdorn, and N W C Dorsch Cerebral vasomotor reactivity testing in head injury: the link between pressure and flow J. Neurol. Neurosurg. Psychiatry, August 1, 2003; 74(8): 1053 - 1059. [Abstract] [Full Text] [PDF]

				B. Schmidt, M. Czosnyka, A. Raabe, H. Yahya, J. J. Schwarze, D. Sackerer, D. Sander, and J. Klingelhofer Adaptive Noninvasive Assessment of Intracranial Pressure and Cerebral Autoregulation Stroke, January 1, 2003; 34(1): 84 - 89. [Abstract] [Full Text] [PDF]

				E W Lang, H M Mehdorn, N W C Dorsch, and M Czosnyka Continuous monitoring of cerebrovascular autoregulation: a validation study J. Neurol. Neurosurg. Psychiatry, May 1, 2002; 72(5): 583 - 586. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Web of Science (23) Citing Articles via Google Scholar Google Scholar Articles by Piechnik, S. K. Articles by Pickard, J. D. Search for Related Content PubMed PubMed Citation Articles by Piechnik, S. K. Articles by Pickard, J. D.

Anesthesia & Analgesia® is published for the International Anesthesia Research Society® by Lippincott Williams & Wilkins with the assistance of Stanford University Libraries' HighWire Press®. Copyright 2006 by the International Anesthesia Research Society. Online ISSN: 1526-7598   Print ISSN: 0003-2999