Neurology -- Correspondence for Miyazawa et al., 60 (6) 941-947

HOME

HELP

FEEDBACK

SUBSCRIPTIONS

Correspondence published:

CSF hypovolemia vs intracranial hypotension in "spontaneous intracranial hypotension syndrome": Gordon Kelley (15 September 2003)

Reply to Letter to the Editor: Y Shiga, K Miyazawa, T Hasegawa, and Y Itoyama (18 June 2003)

CSF hypovolemia vs intracranial hypotension in "spontaneous intracranial hypotension syndrome": Sun J Chung (18 June 2003)

CSF hypovolemia vs intracranial hypotension in "spontaneous intracranial hypotension syndrome" 15 September 2003

Gordon Kelley,
Shawnee Mission Medical Center
8800 W 75th Street #100, Shawnee Mission, KS 66204
Send Correspondence to journal:
Re: CSF hypovolemia vs intracranial hypotension in "spontaneous intracranial hypotension syndrome"

dr_gordo{at}yahoo.com Gordon Kelley

Miyazawa et al [1] demonstrate that venous engorgement occurs in association with CSF hypovolemia and that this may occur in the absence of hypotension as measured by routine lumbar puncture. It is likely that progressive CSF hypovolemia leads to intracranial hypotension but that the intracranial pressure is also dependent on other variables. In mild to moderate CSF hypovolemia, the Monro-Kellie doctrine applies and venous dilation compensates for the CSF volume loss. However, with larger deficits in CSF volume, the ability of vasodilation to adequately compensate for the lost volume is surpassed. Upright intracranial pressures will drop proportionately faster and the loss of brain buoyancy will result in greater postural shifts of the posterior fossa contents. When the body is upright in the normal condition, the intracranial pressure ranges from –5 to +5 cm H2O, referenced to the foramen of Monro. It is not until the pressure falls in the range of –30 to –35 cm that patients will start experiencing orthostatic headache. [2]
The extra-axial fluid collections seen in Miyazawa et al’s patients may have minimized their patients’ complaints. Brain buoyancy is the key variable in determining whether patients have clinical symptoms of orthostatic headache or decreased level of consciousness. The Monro-Kellie doctrine has been conceptualized to apply to the intracranial contents. It is usually expressed as “the sum of volumes of brain, CSF, and intracranial blood is constant.” [3] However, the intracranial space is continuous with the intraspinal space. In patients with large spinal extradural CSF collections, the combined volume of intrathecal and extradural CSF may be enough to maintain brain buoyancy in the upright state without a drop in intracranial pressure or shift of intracranial contents. This is probably how the epidural blood patches or intradural saline infusions quickly relieve symptoms. [4]
References
1. Miyazawa K, Shiga Y, Hasegawa T et al. CSF hypovolemia vs intracranial hypotension in “spontaneous intracranial hypotension syndtrome”. Neurology 2002;60:941-947.
2. Chapman PH, Cosman ER, Arnold MA. The relationship between ventricular fluid pressure and body position in normal subjects and subjects with shunts: a telemetric study. Neurosurgery 26:181-189, 1990.
3. Mokri B. The Monro-Kellie hypothesis Applications in CSF volume depletion. Neurology 2001; 1746-1748.
4. Sencakova D, Mokri B and McClelland RL. The efficacy of epidural blood patch in spontaneous CSF leaks. Neurology 2001;57:1921-1923.

Reply to Letter to the Editor 18 June 2003

Y Shiga
Tohoku University School of Medicine Sendai Japan,
K Miyazawa, T Hasegawa, and Y Itoyama
Send Correspondence to journal:
Re: Reply to Letter to the Editor

yshiga{at}em.neurol.med.tohoku.ac.jp Y Shiga, et al.

Dr. Chung points out several differences from previous reports including his article. [1] He has reservations about the diagnostic sensitivity of spinal MRI. We think that spinal MRI using conventional methods, which has high spatial resolution can detect the leaked CSF, although the leaked CSF compresses the subarachnoid space and sometimes causes an effacement of the subarachnoid space. Therefore, a small amount of leaked CSF into the epidural space can be mistaken for the normal subarachnoid space and easily overlooked. In our report, [2] we described the diagnostic tips and demonstrated that spinal MRI was a more useful diagnostic test than had been thought. Diffuse pachymeningeal gadlinium enhancement is the well-known brain-MRI dinsing of so-called spontaneous intracranial hypotension (SIH) syndrome but we must reconfirm the usefulness of spinal MRI. Based on the spatial resolution we found that the diagnostic sensitivity of spinal MRI was superior to that of radionuclide cisternography, especially for the detection of small amounts of leaked CSF.
We emphasize again that spinal MRI is non-invasive test, whereas RC and CT myelography require invasive lumbar puncture to administer the radioisotope and contrast material intrathecally. Furthermore, accurate lumbar punctureis not easy in the case of patients with so-called SIH syndrome. Dr. Chung takes account of technical problems of our radionuclide cisternography but his objection is off the point because experts did our intrathecal injections of radioisotope after obtaining the CSF. However, we examined only four patients. We agree with Dr. Chung that a large-scale study is needed to estimate strictly which examination is superior in terme of the diagnostic sensitivity.
We enrolled 10 consecutive patients examined by spinal MRI who showed the typical orthostatic headache, i.e., a headache that occurred less than 15 minutes after assuming an upright position and disappeared or improved less than 30 minutes after resuming the recumbent position [3] without a previous history of dural tear or lumbar puncture. No patients were specifically selected. Quite recently, we experienced one man whose spinal MRI demonstrated epidural CSF leakage. In our series, the orthostatic headache had improved in all patients within a month with only conservative treatment including strict bed rest. No patients required epidural blood patches or surgical treatment. We cannot account for the difference between our result [2] and Dr. Chung¹s result. [1] The symptoms of our patients might have been milder than those of Dr. Chung¹s patients, but this also is unclear.
References
1) Chung SJ, Kim JS, Lee MC. Syndrome of cerebral spinal fluid hypovolemia: clinical and imaging features and outcome. Neurology 2000;55:1321-1327.
2) Miyazawa K, Shiga Y, Hasegawa T, et al. CSF hypovolemia vs intracranial hypotension in "spontaneous intracranial hypotension syndrome." Neurology 2001;60:941-947.
3) Headache classification committee of the international headache society:classification and diagnostic criteria for headache disorders, cranial neuralgias and facial pain. Cephalgia 1988;8(Suppl 7):1-96.

CSF hypovolemia vs intracranial hypotension in "spontaneous intracranial hypotension syndrome" 18 June 2003

Sun J Chung
Asan Medical Center Seoul South Korea
Send Correspondence to journal:
Re: CSF hypovolemia vs intracranial hypotension in "spontaneous intracranial hypotension syndrome"

sjchung{at}amc.seoul.kr Sun J Chung

In a recent article, Miyazawa et al. [1] investigated the role of CSF hypovolemia in spontaneous intracranial hypotension syndrome. Although they successfully concluded that CSF hypovolemia is the fundamental cause of this syndrome, I have reservations about their interpretations of the diagnostic sensitivity of the spinal MRI and radionuclide cisternography (RC) for detecting CSF leakage. Although it has been reported that the sensitivity of RC for detecting CSF leak was lower than that of CT myelography, [2] systemically randomized large scaled study has not been done. In previous study,[3] 91 % out of 23 patients showed abnormal RC findings compatible with CSF hypovolemia and 52% showed paradural activity suggesting CSF leakage or meningeal diverticula. In a study of Miyazawa et al., [1] they performed RC in four patients out of ten. Although the early imaging findings of RC are very important to detect the abnormalities of patients with CSF hypovolemia caused by CSF leakage, [3] they did not obtain RC less than 2.5 hours after injection of radionuclide. In addition, the imaging quality of RC is dependent on accurate insertion of radioisotope into CSF space by the expert, which is not easy for the patient with CSF hypovolemia because of low CSF pressure. I suspect that this may account for the fact that there was a quite low incidence of abnormalities suggesting CSF leakage on RC in their study. [1] Because the number of subjects who underwent RC was very small, they should have analyzed the diagnostic sensitivity of each imaging study with prudence.
Regarding spinal MRI findings, although it was not performed in a specialized method used in previous report, [4] Miyazawa et al. reported that all patients showed extradural fluid collection, [1] that is in odds with previous study (false negative rate, 33%). [2] Furthermore, the fact that all subjects were woman and improved with conservative management is very unusual. In previous study, the therapeutic efficacy of supportive management for patients with CSF hypovolemia was 29 % at 4 weeks after the treatment. [3] They did not mention whether their patients were specifically selected or included consecutively. Therefore, I suggest that well-designed prospective large scaled study is needed to assess the exact diagnostic sensitivity of spinal MRI and RC.
References
1. Miyazawa K, Shiga Y, Hasegawa T, et al. CSF hypovolemia vs intracranial hypotension in "spontaneous intracranial hypotension syndrome". Neurology 2003;60:941-947.
2. Schievink WI, Meyer FB, Atkinson JLD, Mokri B. Spontaneous spinal cerebrospinal fluid leaks and intracranial hypotension. J Neurosurg 1996;84:598-605.
3. Chung SJ, Kim JS, Lee MC. Syndrome of cerebral spinal fluid hypovolemia: clinical and imaging features and outcome. Neurology 2000;55:1321-1327.
4. Matsumura A, Anno I, Kimura H, Ishikawa E, Nose T. Diagnosis of spontaneous intracranial hypotension by using magnetic resonance myelography. J Neurosurg 2000:92:873-876.