The flow of cerebrospinal fluid (CSF) through the aqueduct was studied with an echoplanar imaging technique. Images (1024) of a slice perpendicular to the aqueduct were acquired with a repetition time of 107 msec and a flip angle of 90°. This imaging technique is very sensitive for flow into the selected slice, although a quantitative assessment of flow velocities is not possible. Simultaneously with the image data acquisition, data from a pulse oximeter and a respiration belt were recorded. For each data point, a delay time to the preceding cardiac pulse was determined from the recorded pulse wave. The signal intensities could then be assigned to the cardiac cycle. Each cardiac interval was assigned to one of eight respiratory phases, and an average signal curve during the cardiac interval was calculated for each respiration phase. The evaluation showed to signal maxima within the cardiac interval, which could be identified as a downward flow at 10% and an upward flow at 80% of the cardiac pulse interval by measurements with additional saturation pulses. In examinations of 22 healthy volunteers, an influence of respiration on the flow through the aqueduct was found. In spite of interindividual variability, comparable effects could be observed in all volunteers. In the late expiration phase the caudally directed flow was at its maximum, whereas the cranially directed flow was maximal in the post‐inspiration phase. J. Magn. Reson. Imaging 2000;11:438–444. © 2000 Wiley‐Liss, Inc.

Purpose: To detect oscillations of the cerebrospinal fluid (CSF) flow related to the heartbeat and frequencies lower than 0.6 Hz and to compare these oscillations of CSF and blood flow in cerebral vessels by using echo planar imaging in real time mode. The existence of such waves has been well known but has not yet been shown by MRI. Materials and Methods:In a slice perpendicular to the aqueduct, CSF flow as well as CBF, could be determined in sagittal sinus, basilar artery, and capillary vessels. After Fourier analysis, four frequency bands were assigned. Results:In the very high-frequency (heart rate) range, the integrals under the CSF curves were more closely related to arterial CBF than to changes in the sinus. Also, in the high-frequency (respiration rate), low-frequency (0.05-0.15 Hz), and very-low-frequency (0.008 -0.05 Hz) ranges, the integrals under the CSF curves corresponded with arterial and capillary CBF. Conclusion:Slow and fast oscillations in CSF flow are detectable in healthy persons with a proportional allotment to arterial and capillary CBF. RHYTHMIC OSCILLATIONS WITH LOW frequencies below the heart rate were first described during arterial blood pressure monitoring by Hering, Traube, and Mayer as early as the 19th century (1-3). Subsequently, similar fluctuations were observed in various other physiological and pathological parameters, such as heart rate variability, cerebral blood flow (CBF) velocity, and vessel diameter variation (4 -6).In the cranium, slow oscillations of varying frequencies were first documented in association with intracranial pressure (ICP) measurements. The distinction between B-waves with a frequency of 0.5-2/minute and C-waves with a range of 4 -8/minute was first introduced by Lundberg (7). Subsequently, corresponding cerebral oscillations were found in studies of blood flow (8) and blood flow velocity (9). Prior to magnetic resonance imaging (MRI), intracranial slow rhythmic oscillations were detectable only by invasive intracranial pressure measurement or, in the case of CBF, by noninvasive Doppler measurements.Pulsation and flow of the cerebrospinal fluid (CSF) are closely linked to the changes in the CBF induced by the heartbeat and by respiration. These have been described in a number of MRI studies, but additional slow oscillations within the CSF-detectable with MRIhave not been reported. The majority of the studies of CSF flow are based on cardiac-gated MRI and have used gradient echo sequences and averaging of repetitive CSF signals to identify distinct frequencies and waveforms. With this averaging technique, a pulsatile character of CSF flow within the aqueduct was repeatedly demonstrated with a systolic down and diastolic up movement of the CSF (10,11). However, ECG-gated examinations are largely restricted to CSF oscillations in the range of the heart rate.Additional influences of the respiration can be observed if an echo-planar imaging (EPI) sequence is used (12). To detect slower rhythmic oscillations within the CSF motion, real-time techniques are ne...

No abstract