Physiological and Pathological Changes in Cranial Cerebro-Spinal Fluid Volume in Man, As Defined by Magnetic Resonance Imaging

Grant, Robert (1987) Physiological and Pathological Changes in Cranial Cerebro-Spinal Fluid Volume in Man, As Defined by Magnetic Resonance Imaging. MD thesis, University of Glasgow.

Abstract

An accurate and reproducible method for measuring the volume of the cranial CSF spaces was developed in the MRI unit in Glasgow by Dr. B. Condon in 1986. Using this MRI method the total cranial, cortical sulcal, ventricular and posterior fossa CSF volumes could be accurately measured, whereas only ventricular CSF volume could be estimated by previous techniques. The aim of this thesis was firstly to examine the technique critically and to reduce factors that might affect the accuracy or reproducibility of CSF volume measurement; secondly, to determine the normal range of CSF volume; thirdly, to study physiological factors that might influence the cranial CSF volume; and lastly to assess the research and clinical potential of these measurements in conditions where the CSF volumes might be altered. The original technique was modified. The accuracy of the method was improved by using 0.9% sodium chloride as a calibration reference solution, rather than water, as saline was found to produce a signal intensity per unit volume closest to that of anaerobically obtained CSF. The reference phial was sealed, in order to eliminate errors due to phial filling, and placed in an insulated casing that was designed to fit inside the MRI standard head coil. The insulated casing reduced error due to phial cooling during the examination. As the phial was no longer strapped to the head, errors due to phial movement were minimised and the overall examination took leLs time as patient positioning was less critical. The effect of CSF motion on image quality and volume measurement was studied. It was found that CSF motion could result in "background" blurring of the image and errors of approximately 5% could occur. The amount of background blurring was related to the amount of motion within the fluid filled phantom. Background blurring was seen most frequently in patients with obstructive hydrocephalus and in patients with normal pressure hydrocephalus. It is possible that in future the background signal level may be used as an index of CSF motion "in vivo". Cranial CSF volumes increased significantly with age and males had more cranial CSF than females, but there was a wide variation of normal. Total cranial and cortical sulcal CSF volumes increased more significantly than ventricular and posterior fossa volumes reflecting age related cortical atrophy. The normal ratio of ventricular CSF volume to cortical sulcal volume (V:CS ratio) was less than 0.33. The CSF volume measurements were found to be highly reproducible in the short term, but there was a significant increase in CSF volume premenstrually when compared with the mid-cycle CSF volumes. The premenstrual increase may have a hormonal basis or reflect reciprocal changes in intracranial blood volume. Further work is needed to examine intracranial pressure and blood volume changes related to the menstrual cycle and measure CSF volumes in patients with pre-menstrual syndrome, catamenial epilepsy and premenstrual migraine. Total cranial CSF volume decreased during hypercapnia and increased during hypocapnia. This reflected the reciprocal changes in cerebral blood volume and thus provided confirmation of the modified Monro-Kellie Doctrine. Large amounts of CSF were often lost following lumbar puncture and the reduction in CSF volume was related to the presence of headache 24 hours after LP. The preliminary clinical studies included patients with dementia, normal pressure hydrocephalus (NPH), obstructive hydrocephalus and benign intracranial hypertension (BIH). Patients under 70 years of age who had dementia had more cortical sulcal and ventricular CSF than healthy elderly controls, but there was no clear separation between patients over 70 years of age with dementia and healthy elderly subjects. The V:CS ratio separated patients with NPH from patients with dementia of other causes and from healthy volunteers. This measurement may be valuable in the diagnosis of NPH. Patients with a V:CS ratio greater than 1.00 had a significant clinical improvement after V-P shunting but further work is necessary to compare the predictive value of this test with the numerous other tests that claim to be able to predict a good post-operative outcome. The degree of reduction in ventricular CSF volume post-operatively was readily measured and these measurements may be important in the management of patients if a shunt blockage is suspected. (Abstract shortened by ProQuest.).

• Item Type: Thesis (MD)
• Qualification Level: Doctoral
• Keywords: Neurosciences, Medical imaging
• Date of Award: 1987
• Depositing User: Enlighten Team
• Unique ID: glathesis:1987-77592
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 14 Jan 2020 09:04
• Last Modified: 14 Jan 2020 09:04
• URI: https://theses.gla.ac.uk/id/eprint/77592