Physiological impact of continuous positive airway pressure (CPAP) on cerebral blood flow and cerebrospinal fluid dynamics in healthy awake volunteers (MRI)


If you would like to help with this study as a master's or bachelor's thesis project, please send Bryn Martin an email with your resume and research interests.


The problem

The influence of continuous positive airway pressure (CPAP) on cerebral blood flow (CBF) is not yet fully understood and could be important in treatment of people with obstructive sleep apnea syndrome and patients suffering from acute stroke. There are many treatments for obstructive sleep apnea syndrome, but the most widely accepted is continuous positive airway pressure (CPAP) that acts as a pneumatic “splint” by preventing upper airway collapse during sleep. Published data about the effects of CPAP upon cerebral hemodynamics are conflicting (Haring, Hormann et al. 1994; Bowie, O'Connor et al. 2001; Scala, Turkington et al. 2003). Some studies reported an increase in total cerebral blood flow (CBF) (Klingelhofer, Hajak et al. 1992; Haring, Hormann et al. 1994), others a decrease (Netzer, Werner et al. 1998; Scala, Turkington et al. 2003) and some no significant changes (Bowie, O'Connor et al. 2001). It has been postulated that CPAP is the appropriate treatment for OSAS in the first hours after acute stroke when ischemic penumbra is still present (Loube, Gay et al. 1999) and loss of local cerebral autoregulation is likely to occur making CBF preload-dependent. Consequently, CPAP could have detrimental effects on ischemic but viable brain tissue by reducing CBF resulting in unwanted side effects such as decrease of mean arterial pressure and augmentation of intracranial pressure (ICP) due to impaired intracranial venous outflow(Hormann, Mohsenipour et al. 1994).

The impact of CPAP on CSF hydrodynamics has not been studied. However, there is evidence to support that an increase in intrathorasic pressure coinciding with CPAP usage could alter CSF hydrodynamics. Communication between the intrathorasic and ICP due to postural changes (Dabrowski 2007; De Keulenaer, Cheatham et al. 2009), coughing (Sansur, Heiss et al. 2003), valsalva and Queckenstedt’s test, and abdominal pressure (Bloomfield, Ridings et al. 1997) has been well documented (Williams 1981; Williams 1981; Heiss, Patronas et al. 1999; Sansur, Heiss et al. 2003). These pressure changes are transmitted from the abdomen into the CSF system (ICP) through the dural venous sinuses and epidural venous plexus. The change in ICP could then have an impact on CSF hydrodynamics including CSF velocity wave speed in the spinal subarachnoid space (Kalata, Martin et al. 2009). It is unknown if these changes in CSF hydrodynamics brought by CPAP usage would alter cerebral hemodynamics.

Hypothesis and research objectives

We hypothesize that CPAP will have an impact on cerebral blood flow and cerebrospinal fluid dynamics in healthy subjects and that the impact can be measured by non-invasive MRI measurements. The goal of this research is to perform a pilot study to determine what impact, if any, CPAP has on cerebral hemodynamics and CSF hydrodynamics.

Methods and study outline

in vivo MR measurements. Our approach is to obtain MRI measurements on healthy subjects with and without CPAP. Young healthy male subjects without any previous cardiovascular disorders or sleep apnea will be selected for the study to be conducted in the early afternoon at least two hours after any food or caffeine consumption. PtcCO2 will be monitored with a ‘Tosca 500’ system (Radiometer Basel AG, Basel, Switzerland) using a single sensor applied to the foot. Blood pressure (BP) will be recorded before and after each test. A medical doctor will be present throughout the duration of testing. MRI measurements under CPAP will begin after the PtcCO2 level return to baseline (±2mmHg) or at steady state after at least 15 minutes of CPAP use. Following the MRI measurements with CPAP, the mask will be removed and the same procedure repeated without application of the mask.

  1. The subjects will be scanned in the supine position with their necks in a neutral position according to the protocol by Baledent et al. (Baledent, Henry-Feugeas et al. 2001). Cerebrospinal fluid flow acquisition planes perpendicular to the assumed direction of flow will be selected on a midsagittal scan at the C2–C3 subarachnoid space. Axial vascular flow will be determined in the internal carotid (ICAs) and vertebral arteries and the internal jugular veins at C2–C3. Flow images will be acquired with a velocity-encoded (Venc) cine phase-contrast pulse sequence with peripheral or electrocardiographic gating. Magnetic resonance imaging parameters will include a repetition time = 29 to 43 ms, echo time = 11.6 to 17 ms, a 160 x 120-mm (frequency x phase direction) rectangular field of view, a 256 x 128 matrix, a section thickness of 5 mm, a flip angle of 20° to 30°, and two excitations. The minimum repetition time available will be used to optimize temporal resolution, and the minimum echo time available will be used to optimize signal-to-noise ratio and to reduce intravoxel phase dispersion. The 160 x 120 field of view bill be a compromise between scan time, spatial resolution, and signal-to-noise ratio and sufficiently large to avoid aliasing in the center of the image where flow was measured. Retrospective cardiac gating will be used, so that the 16 frames to be analyzed cover the entire CC. Velocity sensitization will be set to 5 cm/s for the cervical subarachnoid spaces and 80 cm/s for the vessels.
  1. Measurement of CSF velocity wave speed will be obtained on the subject using the methods of Kalata et al. (Kalata, Martin et al. 2009). A slice location will be identified which passes through the center of the spinal cord and spinal canal in the sagittal plane. At this slice location, a cine phase contrast velocity scan will be acquired with in-plane velocity encoding in the foot head direction at a value of 20 cm/s. Retrospective ECG or peripheral pulse (PPU) gating will be used to reconstruct each case with >100 frames over the cardiac cycle. The TR and TE for the sequence will be ~4.4 and 2.4 ms, respectively. Slice thickness will be ~8 mm and in-plane reconstructed pixel sizes will be ~1.5 x 1.5 mm. Overall scan time will be two to three minutes, depending on the heart rate.

Single-blind processing of MR data. Measurement of CSF and CBF through the vessels of interest will be performed using Segment software. CSF velocity wave speed in the spinal canal for each subject will be obtained from the MR study data using the technique employed by Kalata et al. (Kalata, Martin et al. 2009) and Filden et al. (Fielden, Fornwalt et al. 2008). These calculations will be performed by an engineer blinded to the test case.

Expected results and potential impact

The proposed work will determine what impact, if any, CPAP has on CBF and CSF flow dynamics. It will give new information about the physiological impact of CPAP and provide a new MRI methodology to measure the influence of CPAP on the cerebrovascular and CSF system.

Preliminary results


Measurement scheduling


Baledent, O., M. C. Henry-Feugeas, et al. (2001). "Cerebrospinal fluid dynamics and relation with blood flow: a magnetic resonance study with semiautomated cerebrospinal fluid segmentation." Invest Radiol 36(7): 368-377. Bloomfield, G. L., P. C. Ridings, et al. (1997). "A proposed relationship between increased intra-abdominal, intrathoracic, and intracranial pressure." Crit Care Med 25(3): 496-503.

Bowie, R. A., P. J. O'Connor, et al. (2001). "The effect of continuous positive airway pressure on cerebral blood flow velocity in awake volunteers." Anesth Analg 92(2): 415-417.

Dabrowski, W. (2007). "Changes in intra-abdominal pressure and central venous and brain venous blood pressure in patients during extracorporeal circulation." Med Sci Monit 13(12): CR548-554.

De Keulenaer, B. L., M. L. Cheatham, et al. (2009). "Intra-abdominal pressure measurements in lateral decubitus versus supine position." Acta Clin Belg 64(3): 210-215.

Fielden, S. W., B. K. Fornwalt, et al. (2008). "A new method for the determination of aortic pulse wave velocity using cross-correlation on 2D PCMR velocity data." J Magn Reson Imaging 27(6): 1382-1387.

Haring, H. P., C. Hormann, et al. (1994). "Continuous positive airway pressure breathing increases cerebral blood flow velocity in humans." Anesth Analg 79(5): 883-885.

Heiss, J. D., N. Patronas, et al. (1999). "Elucidating the pathophysiology of syringomyelia." J Neurosurg 91(4): 553-562.

Hormann, C., I. Mohsenipour, et al. (1994). "Response of cerebrospinal fluid pressure to continuous positive airway pressure in volunteers." Anesth Analg 78(1): 54-57.

Kalata, W., B. A. Martin, et al. (2009). "MR measurement of cerebrospinal fluid velocity wave speed in the spinal canal." IEEE Trans Biomed Eng 56(6): 1765-1768.

Kalata, W., B. A. Martin, et al. (2009). "MR Measurement of Cerebrospinal Fluid Velocity Wave Speed in the Spinal Canal." IEEE Transactions on Biomedical Engineering.

Klingelhofer, J., G. Hajak, et al. (1992). "Assessment of intracranial hemodynamics in sleep apnea syndrome." Stroke 23(10): 1427-1433.

Loube, D. I., P. C. Gay, et al. (1999). "Indications for positive airway pressure treatment of adult obstructive sleep apnea patients: a consensus statement." Chest 115(3): 863-866.

Netzer, N., P. Werner, et al. (1998). "Blood flow of the middle cerebral artery with sleep-disordered breathing: correlation with obstructive hypopneas." Stroke 29(1): 87-93.

Sansur, C. A., J. D. Heiss, et al. (2003). "Pathophysiology of headache associated with cough in patients with Chiari I malformation." J Neurosurg 98(3): 453-458.

Scala, R., P. M. Turkington, et al. (2003). "Effects of incremental levels of continuous positive airway pressure on cerebral blood flow velocity in healthy adult humans." Clin Sci (Lond) 104(6): 633-639.

Williams, B. (1981). "Simultaneous cerebral and spinal fluid pressure recordings. 2. Cerebrospinal dissociation with lesions at the foramen magnum." Acta Neurochir (Wien) 59(1-2): 123-142.

Williams, B. (1981). "Simultaneous cerebral and spinal fluid pressure recordings. I. Technique, physiology, and normal results." Acta Neurochir (Wien) 58(3-4): 167-185.