The Impact of Hypercapnia on Systolic Cerebrospinal Fluid Peak Velocity in the Aqueduct of Sylvius : Anesthesia & Analgesia

Secondary Logo

Journal Logo


The Impact of Hypercapnia on Systolic Cerebrospinal Fluid Peak Velocity in the Aqueduct of Sylvius

Kolbitsch, Christian MD, DEAA*,; Lorenz, Ingo H. MD*,; Hörmann, Christoph MD*,; Schocke, Michael F. MD†,; Kremser, Christian PhD†,; Moser, Patrizia L. MD‡,; Pfeiffer, Karl P. PhD§,; Benzer, Arnulf MD, DEAA*

Author Information
Anesthesia & Analgesia 95(4):p 1049-1051, October 2002. | DOI: 10.1213/00000539-200210000-00047
  • Free


It was previously reported that systolic cerebrospinal fluid (CSF) peak velocity (CSFVPeak) decreased when CSF pressure (Pcsf) was experimentally increased by continuous positive airway pressure (CPAP) breathing at 12 cm H2O (1,2). It was concluded that monitoring of systolic CSFVPeak in the aqueduct of Sylvius is a sensitive means of detecting even minor changes in cerebral capacity (e.g., compliance) (1). Clinically relevant changes in cerebral compliance can also be caused by changes in cerebral blood volume (CBV). Inhaled anesthetics produce cerebral vasodilation and increase CBV (3).

Before evaluating the effect of these anesthetics on cerebral compliance by measuring systolic CSFVPeak, however, the effect of nonanesthetic-induced changes in CBV on systolic CSFVPeak needs to be studied. This study therefore focused on the effect of hypercapnia-induced changes in CBV on systolic CSFVPeak in anesthetized patients.


After approval by the local University Ethics Committee and written informed consent, eight anesthetized patients (ASA physical status I) scheduled for lumbar hemilaminectomy underwent two phase-contrast magnetic resonance imaging (MRI) measurements of systolic CSFVPeak in the aqueduct of Sylvius prior to onset of surgery.

Intravenous boli of propofol (3 mg/kg), fentanyl (5 μg/kg), and rocuronium (0.6 mg/kg) were used to induce anesthesia. After orotracheal intubation, IV propofol infusion (5 mg · kg−1 · h−1) was commenced to maintain anesthesia. MRI measurements of systolic CSFVPeak during normocapnia (e.g., end-tidal carbon dioxide concentration [ETco2] = 40 mm Hg; oxygen/air = 0.3; positive end-expiratory pressure [PEEP] = 5 cm H2O) and hypercapnia (e.g., ETco2 = 60 mm Hg; oxygen/air = 0.3; PEEP = 5 cm H2O) were each preceded by a minimum 10-min stabilization period. The sequence of normocapnia and hypercapnia was randomized.

Hemodynamic (heart rate, noninvasive mean arterial blood pressure) and respiratory (ETco2, pulse oximetry hemoglobin saturation, and fraction of inspired oxygen) variables were continuously monitored (S/5 MRI Monitor™; Datex-Ohmeda, Helsinki, Finland).

Measurements of systolic CSFVPeak in the aqueduct of Sylvius were performed on a 1.5 tesla whole-body scanner (Magnetom VISION®; Siemens, Erlangen, Germany) by using a standard circular polarized head coil. A two-dimensional gradient echo sequence (2D-FISP) (repetition time [TR], 100 ms; echo time [TE], 12 ms; α, 10°; acquisition matrix, 256 × 512; field of view, 160 mm) with flow velocity encoding in the slice-select direction was used. For this high-resolution axial technique, which is sensitive to through-plane flow, the maximum detectable flow velocity was 20 cm/s. Electrocardiography triggering was used for prospective gating of the acquisition. The disadvantage of prospective gating, however, is that acquisition is stopped within approximately 200 ms of the next R wave for accurate detection of the next trigger. Thus, the diastolic phase was not evaluated (4). Cardiac gating produced a series of phase-contrast images in various cardiac phases. From these phase-contrast images, a blinded investigator measured the systolic CSFVPeak in the aqueduct of Sylvius by using region of interest measurements. By convention, systolic CSFVPeak values are given as negative values.

Data are presented as mean ± sd. Student’s t-test for paired samples was used for data analysis. P < 0.05 was considered statistically significant. The statistical computer package SPSS® Version 8.0.0 for Windows 95 (SPSS, Inc., Chicago, IL), run on a Compaq® Deskpro EP Series 6350/6.4, was used for statistical analysis.


All patients (n = 8; age, 41 ± 7 yr; weight, 74 ± 11 kg; height, 177 ± 8 cm) completed the study without complication. Hypercapnia (ETco2 = 60 mm Hg) increased systolic CSFVPeak in the aqueduct of Sylvius as compared with normocapnia (ETco2 = 40 mm Hg) (hypercapnia: −5.67 ± 0.74 cm/s versus normocapnia: −3.54 ± 0.98 cm/s) (Table 1).

Table 1:
Individual Systolic CSFVPeak Values During Normocapnia and Hypercapnia

Physiologic values are presented in Table 2. Except for the ETco2 differences between normocapnia and hypercapnia, there were no statistically significant differences.

Table 2:
Hemodynamic and Respiratory Variables (n = 8 patients)


Hypercapnia increased systolic CSFVPeak in anesthetized patients. In a previous volunteer study, systolic CSFVPeak decreased (1) when Pcsf was experimentally increased by CPAP breathing at 12 cm H2O (2). It was concluded that during CPAP breathing, pressure transmission from the thorax to craniospinal space increased Pcsf, thereby increasing outflow resistance for systolic craniocaudal CSF displacement. Consequently, CPAP breathing decreased systolic CSFVPeak in the aqueduct of Sylvius (1).

Changes in arterial carbon dioxide partial pressure, which correlate well with ETco2(5–7), cause changes in CBV (8). Voluntary hypoventilation, however, is not sufficient to permanently increase ETco2 to >45 mm Hg. We therefore tested the sensitivity of systolic CSFVPeak measurements to relevant increases in ETco2 in anesthetized patients.

In this study, hypercapnia increased systolic CSFVPeak as compared with normocapnia. Bearing in mind the systolic CSFVPeak decrease during CPAP breathing, one possibility would be to assume that the increase in systolic CSFVPeak observed in the present study was caused by a decreased outflow resistance for systolic craniocaudal CSF displacement. A more likely explanation for the increase observed in systolic CSFVPeak, however, is as follows: a relevant increase in CBV, e.g., during hypercapnia, increases to a certain extent the pressure on the cerebral ventricles. Assuming the validity of the known monoexponential relationship between CBV and intracranial pressure (ICP) (9), then a new static equilibrium is simultaneously achieved in the rising part of the ICP-versus-CBV curve. At a sustained cardiac output, the beat-by-beat systolic increase in CBV is maintained. In the ascending part of the ICP-versus-CBV curve, this beat-by-beat change in CBV causes a greater change in pressure. Consequently, the pressure gradient in the aqueduct of Sylvius increases, which, in turn, causes the observed increase in systolic CSFVPeak. It seems likely that cardiac output was unchanged in this study, because hypercapnia had no influence on heart rate or mean arterial blood pressure. To further substantiate this hypothesis, static (e.g., independent of cardiac cycle) and dynamic (e.g., dependent on cardiac cycle) measurements of ventricular CSF volume (10) and pressure would be needed which were not feasible in this study. For the sake of completeness, compression of the aqueduct of Sylvius with a consequent increase in the systolic CSFVPeak must also be considered. The regions of interest used to measure systolic CSFVPeak covered the whole cross-sectional area of the aqueduct. Because there was no change in the cross-sectional area of the aqueduct, this hypothesis can be excluded.

Nevertheless, methodical inaccuracies may influence measurements of systolic CSFVPeak in the aqueduct of Sylvius because of the small size of this structure (normally 2–3 mm) (11). When investigating steady and unsteady flow, inaccuracies between 5%(12) and 7.5%(13) have been described, but in our study control systolic CSFVPeak values (−3.54 ± 0.98 cm/s) were in good accordance with previously reported data (−2.0 to −5.2 cm/s) (4,14). Furthermore, evaluation of diastolic CSF flow velocity profiles was not possible because of the prospective instead of continuous retrospective cardiac gating involved in our study.

In all the studied patients, lumbar disk herniation affected the lumbar CSF space to a varying degree. The most likely effect would be an increased outflow resistance for systolic craniocaudal CSF displacement, which—as for CPAP breathing—would have to cause a decrease in systolic CSFVPeak. In this study, however, systolic CSFVPeak was increased, which suggests a negligible effect of that pathology on our results.

In conclusion, hypercapnia increased systolic CSFVPeak in the aqueduct of Sylvius in anesthetized patients. This finding forces us to expand the conclusions made in our previous study (1) in two respects. First, an increase in CBV (e.g., during hypercapnia, as in this study) influences systolic CSFVPeak, probably by means of a mechanism other than an increase in Pcsf (e.g., during CPAP breathing, as in our previous study). Second, depending on the particular mechanism, a change in cerebral compliance can cause an increase in systolic CSFVPeak—as in this study—as well as a decrease in systolic CSFVPeak, as in our previous study. However, further work is needed to fully establish noninvasive assessment of cerebral compliance by measuring systolic CSFVPeak in the aqueduct of Sylvius.

The authors are indebted to those patients at Innsbruck University Hospital whose participation made this study possible.


1. Kolbitsch C, Schocke M, Lorenz IH, et al. Phase-contrast MRI measurement of systolic cerebrospinal fluid peak velocity (CSFV Peak) in the aqueduct of Sylvius: a noninvasive tool for measurement of cerebral capacity. Anesthesiology 1999; 90: 1546–50.
2. Hormann C, Mohsenipour I, Gottardis M, Benzer A. Response of cerebrospinal fluid pressure to continuous positive airway pressure in volunteers. Anesth Analg 1994; 78: 54–7.
3. Sakabe T, Nakakimura K. Effects of anesthetic agents and other drugs on cerebral blood flow, metabolism, and intracranial pressure. In: Cottrell JE, Smith DS, eds. Anesthesia and neurosurgery. St. Louis: Mosby, 2001: 129–43.
4. Nitz WR, Bradley WG Jr, Watanabe AS, et al. Flow dynamics of cerebrospinal fluid: assessment with phase-contrast velocity MR imaging performed with retrospective cardiac gating. Radiology 1992; 183: 395–405.
5. Young WL, Prohovnik I, Ornstein E, et al. Cerebral blood flow reactivity to changes in carbon dioxide calculated using end-tidal versus arterial tensions. J Cereb Blood Flow Metab 1991; 11: 1031–5.
6. Campbell FA, McLeod ME, Bissonnette B, Swartz JS. End-tidal carbon dioxide measurement in infants and children during and after general anaesthesia. Can J Anaesth 1994; 41: 107–10.
7. Bongard F, Wu Y, Lee TS, Klein S. Capnographic monitoring of extubated postoperative patients. J Invest Surg 1994; 7: 259–64.
8. Quint SR, Scremin OU, Sonnenschein RR, Rubinstein EH. Enhancement of cerebrovascular effect of CO2 by hypoxia. Stroke 1980; 11: 286–9.
9. Sullivan HG, Miller JD, Griffith RL III, Becker DP. CSF pressure transients in response to epidural and ventricular volume loading. Am J Physiol 1978; 234: R167–71.
10. Ertl-Wagner BB, Lienemann A, Reith W, Reiser MF. Demonstration of periventricular brain motion during a Valsalva maneuver: description of technique, evaluation in healthy volunteers and first results in hydrocephalic patients. Eur Radiol 2001; 11: 1998–2003.
11. Quencer RM, Post MJ, Hinks RS. Cine MR in the evaluation of normal and abnormal CSF flow: intracranial and intraspinal studies. Neuroradiology 1990; 32: 371–91.
12. Ku DN, Biancheri CL, Pettigrew RI, et al. Evaluation of magnetic resonance velocimetry for steady flow. J Biomech Eng 1990; 112: 464–72.
13. Frayne R, Steinman DA, Ethier CR, Rutt BK. Accuracy of MR phase contrast velocity measurements for unsteady flow. J Magn Reson Imaging 1995; 5: 428–31.
14. Henry-Feugeas MC, Idy-Peretti I, Blanchet B, et al. Temporal and spatial assessment of normal cerebrospinal fluid dynamics with MR imaging. Magn Reson Imaging 1993; 11: 1107–18.
© 2002 International Anesthesia Research Society