| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1
Department of Radiology and Radiological Sciences, Johns Hopkins Hospital, 600
N. Wolfe St., Baltimore, MD 21287.
2
Present address: Boston University School of Medicine, 88 E. Newton St.,
Boston, MA 02118.
Received September 16, 1999;
accepted after revision December 17, 1999.
Address correspondence to E. R. Melhem.
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. Fifteen volunteers were imaged using nontriggered and triggered phase-contrast MR venography of the superior sagittal and transverse sinuses. Triggered scans were obtained during regular breathing; nontriggered scans were obtained during regular breathing, breath-holding, deep inspiratory breath-holding, and deep expiratory breath-holding. Analysis of variance, Bonferroni method, and Dunn post hoc analysis were used to determine any significant differences in the mean flow and velocity between the different breathing maneuvers. A paired t test was used to compare flow between sinuses during regular breathing.
RESULTS. Deep inspiratory breath-holding and deep expiratory breath-holding resulted in a significant decrease in blood flow and velocity in all dural sinuses compared with regular breathing. During deep inspiratory breath-holding, blood flow decreased 30.8% in the superior sagittal sinus, 19.7% in the left transverse sinus, and 19.1% in the right transverse sinus. Similarly, during deep expiratory breath-holding, blood flow decreased 30.2% in the superior sagittal sinus, 20.8% in the left transverse sinus, and 20.3% in the right transverse sinus. The sum of the flow in the transverse sinuses was significantly greater than in the sagittal sinus. Normal pulsatility of dural sinus blood velocity was also characterized for all measured sinuses.
CONCLUSION. Characterization of variations in dural sinus velocity and flow as a function of the cardiac cycle and breathing maneuvers, using phase-contrast MR imaging, may help separate physiologic from pathologic changes of flow resulting from conditions that influence the cerebrovascular circulation.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The effects of physiologic maneuvers (such as Valsalva's maneuver and Müeller's maneuver) on the cardiovascular system have been extensively studied and divided into distinct phases of mechanical stress and autonomic response [11]. The effect of these physiologic maneuvers on the cerebrovascular system has been assessed in the arterial circulation only. The effects on the venous circulation have not been well established [12]. These physiologic stressors may influence flow in the dural sinuses as a result of changes in intrathoracic pressures. Valsalva's maneuver generates a positive intrathoracic pressure, whereas Müeller's maneuver generates negative pressures. These pressure changes, through their primary effects and the physiologic responses they consequently elicit, result in dramatic changes in the cardiovascular system, and by extension may affect the cerebrovascular circulation. Using phase-contrast MR imaging, the present study serves to further clarify the effects of these physiologic stressors on blood flow in the dural sinuses.
Our goals in this study were to characterize dural sinus flow patterns through the cardiac cycle and to assess the impact of intrathoracic pressure variation on these patterns.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Single-slice nontriggered and retrospectively triggered two-dimensional phase-contrast MR venography of the dural sinuses was performed using the following parameters: TR range/TE range, 9-16/4.6-7.1; flip angle, 15°; number of excitations, two; receiver bandwidth, 37.7 Hz per pixel; velocity encoding, 30-50 cm/sec (bipolar gradients were applied in the anteroposterior direction); field of view, 14-23 cm; matrix, 179 x 256; slice thickness, 4 mm. The acquisition time for the nontriggered scan was 11 sec and for the triggered scan varied between 4 and 5 min depending on the subject's heart rate. A total of 18 phases were resolved over the cardiac cycle on the triggered acquisition. Magnitude and velocity images were reconstructed on-line using the complex difference and the phase difference of the flow-sensitive and the flow-compensated scans, respectively.
Guided by sagittal and oblique axial two-dimensional phase-contrast MR venograms (TR/TE, 25/7; velocity encoding, 40 cm/sec) showing the superior sagittal sinus and transverse sinuses, respectively, planes of imaging were placed perpendicular to the anterosuperior sagittal sinus, the mid superior sagittal sinus (Fig. 1A,1B,1C), and the mid transverse sinuses. The triggered scans were obtained in all locations during regular breathing. The nontriggered scanning, on the other hand, was repeated four times: during regular breathing, breath-holding, deep inspiratory breath-holding at end inspiration, and deep expiratory breath-holding.
|
|
|
Subjects and Data Analysis
Fifteen healthy young volunteers (10 men and five women, 21-41 years old)
were imaged using this protocol. Subjects were screened for brain
abnormalities and contraindications to MR imaging before scanning.
Dural sinus flow velocity and flow measurements were obtained using variable-sized regions of interest drawn by two independent observers on the magnitude images (PC-M) and copied to the phase images (PC-P) (Figs. 1B and 1C). On the basis of visual assessment, the regions of interest encompassed the entirety of the vascular lumen without extending into the extraluminal stationary tissues. From the nontriggered phase images obtained during the four breathing maneuvers, blood velocity and flow measurements were averaged over the acquisition time of the scans. From the triggered phase images (PC-P), 18 velocity and flow measurements were obtained during each cardiac cycle and were averaged over multiple cycles.
On triggered scans, the pulsatility of blood velocity was mapped out in the anterosuperior sagittal sinus, the mid superior sagittal sinus, the left transverse sinus, and the right transverse sinus. No cardiovascular abnormalities were detected on triggered scans. Data were normalized for each subject with the maximum velocity in the series. After normalization, data for all subjects were averaged together for each sinus and plotted.
Statistical Analysis
Kappa statistics were used to evaluate interobserver agreement on velocity
and flow measurements.
Repeated measures analysis of variance (ANOVA) was used to determine the existence of significant differences in the mean flow and velocity between the regular breathing, breath-holding, deep inspiratory breath-holding, and deep expiratory breath-holding groups for all the dural sinuses. When differences were found, paired analyses of differences were performed using the Bonferroni adjustment and Dunn post hoc analysis.
A two-tailed paired t test was used to compare flow measurements from the anterosuperior sagittal sinus versus the mid superior sagittal sinus, the left transverse sinus versus the right transverse sinus, and the sum of flow in both transverse sinuses versus the flow in the anterosuperior sagittal sinus and the mid superior sagittal sinus.
For the repeated measures ANOVA and twotailed paired t test, p values of less than 0.05 were considered significant, and for the Bonferroni t method, p values of less than 0.005 were considered significant.
Sources of Systemic Error
Steps taken to reduce systemic errors in the venous flow measurements
included reducing the pixel size (in-plane resolution) by varying the field of
view to ensure that the vessel lumen contained at least 16 pixels; choosing
the shortest allowable TE for a given field of view to minimize the effects of
volume averaging, intravoxel phase dispersion, and phase offsets; and choosing
a relatively small flip angle to provide proton density-weighting, which
decreases non-flow-related differences in signal intensity between stationary
tissues and moving blood
[13].
Sources of Random Error
For mildly pulsatile flow, such as in the dural sinuses, the random error
has been shown to be proportional to the choice of velocity sensitivity and
inversely proportional to the signal-to-noise ratio of the flowing blood on
the magnitude images (PC-M). To reduce random error in our measurements, the
encoded velocity was set just greater than the maximum velocity anticipated in
the dural sinus of interest. Also, two excitations were implemented for both
the triggered and nontriggered experiments to increase the signal-to-noise
ratio of the flowing blood on the magnitude images (PC-M)
[13].
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
= 0.60) for
measurements made of the nontriggered and triggered phase images (PC-P).
Nontriggered Phase-Contrast MR Imaging
For the different breathing maneuvers, average and standard deviation of
the velocity and flow measurements in the dural sinuses are listed in
Table 1.
|
Repeated measures ANOVA showed a significant difference in the velocity and flow measurements between the four breathing maneuver groups for all dural sinuses (Table 2). Paired analysis of the groups showed significant differences between regular breathing and deep inspiratory breath-holding, regular breathing and deep expiratory breath-holding, breath-holding and deep inspiratory breath-holding, and breath-holding and deep expiratory breath-holding (p < 0.0001 by Bonferroni t method).
|
Compared with regular breathing, deep inspiratory breath-holding resulted in both decreased blood flow and decreased velocity in all three measured sinuses (sagittal sinus, left transverse sinus, and right transverse sinus). In the anterosuperior sagittal sinus, velocity decreased 29.8%, from 14.87 cm/sec during regular breathing to 10.44 cm/sec during deep inspiratory breath-holding. Blood flow was directly proportional to the velocity decrease, showing a 30.8% reduction, from 3.76 ml/sec to 2.60 ml/sec. A similar pattern was observed in both transverse sinuses. Velocity decrease in the left transverse sinus was 21.9% and flow decrease was 19.7%. In the right transverse sinus, the velocity and flow decreases were close to those observed on the left, 21.1% and 19.1%, respectively.
Compared with regular breathing, deep expiratory breath-holding resulted in changes similar to those seen with deep inspiratory breath-holding. Anterosuperior sagittal sinus velocity and flow decreased 29.6% (from 14.87 to 10.47 ml/sec) and 30.2% (from 3.76 to 2.63 ml/sec), respectively. The left transverse sinus velocity decreased 23.8% and flow decreased 20.8%. The right transverse sinus velocity decreased 21.0%, and the flow decreased 20.3%.
No difference was seen in blood flow and velocity during regular breathing and breath-holding in any of the sinuses in which the measurements were made.
For measurements performed during regular breathing, flow in the mid sagittal sinus (anterosuperior sagittal sinus) was significantly lower than at the mid superior sagittal sinus (3.76 ml/sec versus 6.46 ml/sec, respectively) (p < 0.001). Also, flow in the right transverse sinus was greater than in the left transverse sinus (7.05 ml/sec versus 5.76 ml/sec, respectively), but this difference did not reach statistical significance (p < 0.3). The sum of the flow in the transverse sinuses was significantly greater than in the mid superior sagittal sinus and the anterosuperior sagittal sinus (12.81 ml/sec versus 6.46 ml/sec versus 3.76 ml/sec, respectively) (p < 0.0001) (Table 3).
|
Triggered Phase-Contrast MR Imaging
Figure 2 shows the averaged
data of all subjects. Within each sinus, the velocities were normalized with
the maximum velocity within the series. The cardiac cycle was evaluated over
18 time points, with the first time point corresponding to the beginning of
the cycle. The length of each cardiac cycle varied among subjects depending on
their heart rates. Maximum velocity occurred during systole, minimum during
diastole.
|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The effects of a Valsalva's maneuver on the cardiovascular system involve four phases. Phase I begins at the initiation of the maneuver and involves an immediate increase in arterial blood pressure, which is thought to be a direct result of the sudden increase in intrathoracic pressure. This pressure increase translates not only to the arterial system but also to the venous system [17]. The intrathoracic venous system experiences a reduction in flow and velocity, backing up the dural venous system. Phase II can be separated into two subgroups, IIa (early) and IIb (late). Phase IIa is a decline in arterial blood pressure resulting from impaired atrial filling. It is followed by phase IIb, which involves an increase in blood pressure after a sympathetic response. Phases III and IV both occur after the strain is released and consist of a decrease in arterial blood pressure followed by another sympathetic restoration, respectively [11].
Correlation of these phases with cerebral blood flow has been performed with transcranial Doppler sonography. Tiecks et al. [18] showed that peak flow velocity in the middle cerebral artery decreased significantly during early phase II (phase IIa) of Valsalva's maneuver. Normal breath-holding at tidal volume does not produce this effect [19] because of insufficient elevation in intrathoracic pressure. With Valsalva's maneuver, both decreased carotid flow and decreased thoracic venous return have been documented [20].
Our data showed both decreased velocity and decreased flow in the superior and transverse sinuses during deep inspiratory breath-holding. The acquisition time for the images was 11 sec, corresponding in time with phase IIa of the response to Valsalva's maneuver. The reduced right atrial filling during this phase would result in both a decreased stroke volume and a backup of venous return [17]. The question arises as to whether the major factors contributing to the decline in dural sinus flow represent a primary arterial decrease with secondary venous decrease, primary venous decrease with no contribution from arterial decrease, or combination of arterial and venous decreases. Most likely, this decline represents not only the decreased cerebral arterial blood flow but also the decreased venous return to the heart.
Decreased Flow and Velocity During Deep Expiratory
Breath-Holding
The inspiratory analog of the Valsalva's maneuver is
Müeller's maneuver. In this maneuver, the
volunteer inhales against a closed glottis, thereby producing a negative
intrathoracic pressure resulting from the elastic recoil of the chest wall. In
our study, we used deep expiratory breath-holding to produce the condition of
sustained negative intrathoracic pressures.
Traditionally, Müeller's maneuver was thought to increase venous return, and hence could be used clinically to exaggerate disorders such as tricuspid regurgitation. However, studies have shown that a sustained Müeller's maneuver results in a decreased flow in the superior vena cava. It has also been shown that even though negative intrathoracic pressures facilitate intrathoracic patency of the superior vena cava, veins at the thoracic inlet collapse, reducing their contribution to venous return [21].
In addition to the decreased venous return, decreases in ventricular ejection fraction and in arterial outflow have been shown. The negative intrathoracic pressure effect of Müeller's maneuver causes an increased left ventricular volume. However, the increase in end diastolic volume is less than the increase in end systolic volume, resulting in a net decreased ejection fraction [12]. Studies have also shown decreased left ventricular filling during Müeller's maneuver resulting from changes in left ventricle geometry and decreased mean arterial pressures [22, 23].
Our results showed significant decreases in velocity and flow during deep expiratory breath-holding, most likely resulting from the combined effect of decreased left ventricular filling and collapse of the veins at the thoracic inlet.
Blood Flow Differences Between the Sinuses
Differences in flow within the different dural sinuses can be explained by
the fact that in addition to drainage from the superior sagittal sinus, the
transverse sinuses receive direct venous drainage from the temporal and
occipital lobes, the cerebellum, and the deep venous system
[24].
Also, in most cases, the right transverse sinus drains a greater volume of blood than the left transverse sinus because in most people the sagittal sinus drains preferentially into the right transverse sinus.
Pulsatility During Cardiac Cycle
The pulsatility of dural sinus velocity was well characterized by the
triggered phase-contrast MR imaging. The maximum velocity begins during
systole, with blood flow aided by ventricular descent creating a greater
effective volume of atrium to fill. As the right atrium progressively fills,
velocity steadily decreases to a minimum toward the end of diastole and
corresponds to the trough seen in Figure
2. A subtle mid cycle inflection point is noted that interrupts
the steady decrease in velocity. This short-lived increase in velocity most
likely corresponds to the opening of the tricuspid valve. Velocity then
declines again and courses to its nadir, representing the combined resistance
of right atrial contraction and progressive ventricular filling. The
fluctuations in velocity during the cardiac cycle as shown on phase-contrast
MR imaging are in agreement with previous studies
[1,
3].
In conclusion, characterization of variations in dural sinus velocity and flow as a function of cardiac cycle and breathing maneuvers, using phase-contrast MR imaging, may help elucidate normal and pathologic cerebrovascular physiology. We characterized both velocity and flow in the dural sinuses over the cardiac cycle and during physiologic stress maneuvers. Both deep inspiratory breath-holding and deep expiratory breath-holding were found to decrease dural sinus flow.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  K. Kudo, S. Terae, A. Ishii, T. Omatsu, T. Asano, K. Tha, and K. Miyasaka Physiologic Change in Flow Velocity and Direction of Dural Venous Sinuses with Respiration: MR Venography and Flow Analysis AJNR Am. J. Neuroradiol., April 1, 2004; 25(4): 551 - 557. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. G. Wetzel, V. S. Lee, A. G. S. Tan, O. Heid, S. Cha, G. Johnson, and N. M. Rofsky Real-Time Interactive Duplex MR Measurements: Application in Neurovascular Imaging Am. J. Roentgenol., September 1, 2001; 177(3): 703 - 707. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
= anterior portion of superior sagittal sinus, 
= left
transverse sinus, 
= right transverse sinus, * = mid portion of
superior sagittal sinus.