Hui-Feng Ho1,2, Hung-Chieh Chen2, Hsin Tung3, Yi-Hsin Tsai4, Yi-Ying Wu2, Jyh-Wen Chai2,4, Clayton Chi-Chang Chen2, Shin-Lei Peng1, Wu-Chung Shen1, and Tzu-Ching Shih1
1Biomedical Imaging and Radiological Science, China Medical University, Taichung, Taiwan, 2Radiology, Taichung Veterans General Hospital, Taichung, Taiwan, 3Neurological Institute, Taichung Veterans General Hospital, Taichung, Taiwan, 4College of Medicine, China Medical University, Taichung, Taiwan
Synopsis
Cerebrospinal fluid (CSF) is the fluid circulating through the subarachnoid space and providing a neuroprotective function as a hydraulic cushion for the brain and the spinal cord. CSF not only plays a vital role for normal brain function, but also servers numerous important functions in the central nervous system. However, CSF leaks are a key cause of new-onset headaches. Despite numerous reports characterizing CSF and its circulation in subarachnoid space, our understanding of CSF leakage remains limited. In this study, we calculate the CSF volumes in the brain and the spine of SIH patients using T2-weighted MRI images.
PURPOSE
Spontaneous intracranial hypotension (SIH), resulted from CSF leaks in spine, leads to alterations in the equilibrium between the volumes of intracranial CSF, blood and brain tissue [1]. We believed that morphometric analysis of CSF would be crucial in diagnosis and follow-up of SIH patients. However, there was a lack of literatures in quantification of whole body CSF volume. In this experiment, we attempted to calculate the CSF volume in brain and spine using heavily T2WI MRI for SIH patients before and after treatments.
Six healthy controls and seven SIH patients were excluded if they had arrhythmia, primary headache, structural brain lesions, and lumbar punctures. Brain MRI data were acquired from a whole body 1.5 T MRI system with three sets of brain and spine heavily T2WI MR myelography and voxel size of 1x1x1 mm. Spine MRI data were reconstructed with image distortion correction algorithm to a whole spine MR myelography. The flowchart for CSF volume calculation was displayed in Fig.1(A). We used the entropy threshold and distance regularized level set evolution(DRLSE) to segment the CSF portion from MR images [2]. Fig. 1(B) shows 3D volume of CSF in brain MRI.RESULTS
For six healthy subjects, we evaluated the CSF volume in whole brain, upper and lower spine (listed in Table1). The averaged volumes in the brain, upper spine, and lower spine were 259.8, 50.1, and 46.7 cm3. The averaged total CSF volume was about 365.6cm3.
|
Table 1 CSF volumes in brain, upper and lower spine of six healthy subjects
|
|
Subjects
|
Gender
|
Age
|
Brain
(cm3)
|
Upper spine
(cm3)
|
Lower spine
(cm3)
|
Total volume
(cm3)
|
|
#1
|
F
|
27
|
171.5
|
60.1
|
59.0
|
290.6
|
|
#2
|
M
|
26
|
255.4
|
39.9
|
43.9
|
339.2
|
|
#3
|
M
|
33
|
335.5
|
52.8
|
50.5
|
438.8
|
|
#4
|
M
|
31
|
294.6
|
56.5
|
46.8
|
397.8
|
|
#5
|
M
|
28
|
264.4
|
48.2
|
49.3
|
361.9
|
|
#6
|
F
|
30
|
237.5
|
42.9
|
30.7
|
311.2
|
For treated patients #1-3, their total CSF volumes increased (Table 2). For patient #2 with SIH, the CSF volumes in brain were 226.6 and 289.0 mm3 before and after treatments. Total CSF volume of patient #4 reduced from 327.5 to 240.5 cm3. Indeed, we found the patient #4 associated with the brain hemorrhage. With the brain hemorrhage, regional brain CSF volume of Patient #5 was small as 181.7 cm3. After treatment, the brain CSF volume significantly increased and reached 330.7 cm3. With an unknown CSF leakage, the CSF total volumes of patients # 6 and 7 after treatments were reduced to 17.37% and 3.54%, respectively. CSF volume in brain of patient #7 almost kept the same before and after treatment, but the spinal CSF volume of patient #7 decreased with about 12.1 cm3.
|
Table 2 CSF volume alternations in the brain and spine for seven SIH patients before and after treatments
|
|
|
|
|
|
Before Treatment
|
After Treatment
|
|
Patients
|
Gender
|
Age
|
BMI
|
CSF volume in Brain
(cm3)
|
CSF volume in Spine
(cm3)
|
Total CSF volume (Brain+Spine)
(cm3)
|
CSF volume in Brain
(cm3)
|
CSF volume in Spine (cm3)
|
Total volume (Brain+Spine)
(cm3)
|
|
#1
|
M
|
33
|
32.4
|
226.6
|
75.6
|
302.2
|
235.3
|
93.1
|
328.4
|
|
289.0
|
73.7
|
362.7
|
|
#2
|
M
|
34
|
23.6
|
212.1
|
88.1
|
300.2
|
275.4
|
65.9
|
341.3
|
|
#3
|
F
|
57
|
24.8
|
215.4
|
47.9
|
263.3
|
257.9
|
96.4
|
354.3
|
|
#4
|
F
|
41
|
19.2
|
221.9
|
105.6
|
327.5
|
133.0
|
107.5
|
240.5
|
|
#5
|
M
|
44
|
22.5
|
181.7
|
109.3
|
291.0
|
330.7
|
89.5
|
420.2
|
|
#6
|
M
|
48
|
23.3
|
316.6
|
121.4
|
438.0
|
235.3
|
126.6
|
361.9
|
|
#7
|
F
|
31
|
20.4
|
278.2
|
62.6
|
340.8
|
278.3
|
50.5
|
328.7
|
DISSCUSION
CSF not only plays a key role for brain function, but also servers numerous important functions in the central nervous system [3-4]. CSF volumes in the brain of patient #4 were 221.9 and 133.0 cm3 before and after treatment, respectively. We found that patient # 4 has a subdural hemorrhage. A regional brain CSF volume leakage may be associated with occurring the brain hemorrhage. Leakage of the spontaneous spinal CSF, with resultant CSF hypovolemia and intracranial hypotension, is the most common cause of SIH associated with low CSF pressure. Even after treatment the CSF volume in brain remains almost the same, the total volume of CSF is decrease for the SIH patient due to CSF leakage in the spine (such as patient #7). MR T2-weighted images can use to evaluate the CSF volume alteration in the brain and spine for SIH patients.CONCLUSION
As instead of lumbar puncture, this study used T2-weighted magnetic resonance images to evaluate the CSF volume in the brain and the spine of SIH patients. The MRI T2-weighted findings represent the alternation of CSF volume before and after treatment and compensatory changes in response to the leakage.Acknowledgements
References
[1] Rai A, et al. Epidural blood patch at C2: diagnosis and treatment of spontaneous intracranial hypotension. Am J Neuroradiol 2005;26:2663-6. [2] Chen SH. Segmentation parameter optimization and quantification of cerebrospinal fluid leakage in magnetic resonance myelography, mater thesis, National Yunlin University of Science and Technology, Taiwan. June 2015. [3] Schievink W. Spontaneous spinal cerebrospinal fluid leaks and intracranial hypotension, JAMA 2006;295(19):2286-2296. [4] Ducros A, Biousse V. Headache arising from idiopathic changing in CSF pressure, Lancet Neurol. 2015;14(6):655-668.