A quantitative study of intracranial hypotensive syndrome by magnetic resonance

A quantitative study of intracranial hypotensive syndrome by magnetic resonance

Clinical Neurology and Neurosurgery 141 (2016) 71–76 Contents lists available at ScienceDirect Clinical Neurology and Neurosurgery journal homepage:...

1014KB Sizes 0 Downloads 41 Views

Clinical Neurology and Neurosurgery 141 (2016) 71–76

Contents lists available at ScienceDirect

Clinical Neurology and Neurosurgery journal homepage: www.elsevier.com/locate/clineuro

A quantitative study of intracranial hypotensive syndrome by magnetic resonance Weizhong Tian a , Ji Zhang a,∗ , Jinhua Chen a , Ying Liu b , Xiaoyun Chen a , Ning Wang a a b

Department of Radiology, Taizhou People’s Hospital, Taizhou 225300, Jiangsu Province, China Department of Neurology, Taizhou People’s Hospital, Taizhou 225300, Jiangsu Province, China

a r t i c l e

i n f o

Article history: Received 21 April 2015 Received in revised form 3 December 2015 Accepted 13 December 2015 Available online 21 December 2015 Keywords: Magnetic resonance imaging Intracranial hypotension syndrome Quantitative assessment

a b s t r a c t Objectives: The study aims to investigate the magnetic resonance imaging (MRI) findings of intracranial hypotension syndrome (IHS) and the change of quantitative indicators, so as to yield a deeper understanding of the disease. Patients and methods: The clinical data and MRI findings of 26 cases of IHS which were confirmed by lumbar puncture were retrospectively analyzed. Two physicians evaluated the MRI findings including thickening and enhancement of dural, pituitary enlargement, subdural effusion (hematocele), venous engorgement and brain sagging, and measured the quantitative indicators including mamillopontine distance and pontomesencephalic angle. The consistency between the two results of the physicians was assessed by Kappa consistency test. The differences of mamillopontine distance and pontomesencephalic angle between the patient group and the control group were determined by paired t-test. The diagnostic efficiency of mamillopontine distance and pontomesencephalic angle was assessed by area under the ROC curve, and their best diagnostic thresholds were also determined, respectively. Age- and sex-matched healthy volunteers controls (n = 26) were recruited and served as the control group. Results: All of the 26 patients suffered from the characterized by orthostatic headache of IHS. The clinical evaluations of dural thickening and enhancement, pituitary enlargement, subdural effusion (hematocele), venous engorgement by the two physicians showed excellent agreements ( = 0.808, 1 and 0.906, P < 0.01), and the clinical evaluations of brain sagging showed medium agreements ( = 0.606, P < 0.01). The mamillopontine distance and pontomesencephalic angle of the patient group were 5.4 ± 1.6 mm and 47.8 ± 8.7◦ , respectively, which were obviously less than those of the control group (6.9 ± 1.1 mm and 61.0 ± 6.1◦ , respectively), and the differences were statistically significant (t = −4.563, P < 0.01; t = −.329, P < 0.01). The area under ROC curve of mamillopontine distance and pontomesencephalic angle were 0.774 and 0.908, respectively, and the diagnostic value of pontomesencephalic angle was higher than that of the mamillopontine distance. The sensitivity and specificity were 73.1% and 73.1%, respectively, when diagnostic threshold of mamillopontine distance was 6.4 mm. The sensitivity and specificity were 76.9% and 96.2%, when diagnostic threshold of pontomesencephalic angle was 51.7◦ . Conclusion: The MRI findings presented characteristic features of IHS. The quantitative indicators including mamillopontine distance and pontomesencephalic angle were helpful for clinical diagnosis of subjective findings of IHS. © 2015 Elsevier B.V. All rights reserved.

1. Introduction Intracranial hypotension syndrome (IHS) is a series of clinical syndrome caused by cerebral spinal fluid leakage, in which the decreased cerebral spinal fluid (CSF) opening pressure is less

∗ Corresponding author at: East Yingchun Road No.210, Hailing district, Taizhou 225300, Jiangsu Province, China. Fax: +86052386606776. E-mail address: [email protected] (J. Zhang). http://dx.doi.org/10.1016/j.clineuro.2015.12.014 0303-8467/© 2015 Elsevier B.V. All rights reserved.

than 60 mm H2 O [1–4]. IHS is characterized by an orthostatic headache, which is worsened in the upright posture and alleviated in the recumbent position, and sometimes accompanied by nausea, vomit, dizziness, visual impairments, hearing disorder, mental disorder and so on [5,6]. Magnetic resonance imaging (MRI) findings of IHS have their own characteristic features. These MRI findings include dural thickening and enhancement, pituitary enlargement, subdural effusion (hematocele), venous engorgement, brain sagging and so on [7–12].

72

W. Tian et al. / Clinical Neurology and Neurosurgery 141 (2016) 71–76

Fig. 1. A: MRI obtained in a 27-year-old female patient with orthostatic headache more than a month and accompanied by nausea, vomiting and nuchal rigidity. A1: dural thickening and enhancement and a smoother finished surface with no nodules of the enhanced dura. A2: arc-shaped short-T1 signal appeared at the left frontotemporal of dura which was caused by subdural blood. A3: brain sagging demonstrating enlargement of the pituitary, narrowing of cisterna chiasmatis, prepontine cistern and the fourth ventricle and descent of pointed cerebellar tonsil. B: MRI obtained in a 17-year-old male patient with orthostatic headache. Enhanced MRI scan suggested thickening and expansion of bilateral sigmoid sinus. C: MRI obtained in a 27-year-old female patient of the control group. C1: mamillopontine distance (8.8 mm). C2: pontomesencephalic angle (65.6◦ ). D: MRI obtained in the same patient featured in A. D1: mamillopontine distance (mamillopontine distance). D2: pontomesencephalic angle (33.9◦ ).

W. Tian et al. / Clinical Neurology and Neurosurgery 141 (2016) 71–76

73

Table 1 Evaluation of MRI findings of 26 patients by 2 physicians. Physician

1 2  P-value

MRI findings (case) Dural thickening and enhancement

Pituitary enlargement

Subdural effusion(hematocele)

Venous engorgement

Brain sagging

23 23 1 <0.01

13 15 0.846 <0.01

10 10 1 <0.01

7 8 0.906 <0.01

3 6 0.606 <0.01

Note: Kappa statistics was used.  values ≤0.4 described poor agreements, 0.4–0.75 described fair agreements, and ≥0.75 described good agreements. Student’s statistics was also used. P < 0.01 indicated very significant when compared with the control group.

However, these MRI findings are subjective judgments of diagnosticians, which are closely related to their clinical experiences. The aim of this present study is to find objective and quantitative indicators and to improve the accuracy and reliability of the diagnosis of IHS.

2. Materials and methods 2.1. Patients With IRB (institutional review board) approval from Taizhou People’s Hospital Medical Research Ethics Committee, 30 cases of IHS from August 2006 to December 2013 in our hospital were collected. All patients underwent a lateral approach for lumbar puncture and their CSF opening pressure were less than 60 mm H2 O. Two of the 30 patients with IHS were not subjected to MRI examination. Another 2 IHS patients were also excluded because of the large artifacts of their dentures in MRI images. Namely, 26 cases were incorporated in the patient group, including 8 males and 18 females whose ages were 18–78 years old (mean age, 38.3 years old), and their clinical course of disease varied from 2 days to 1 years. In the 26 cases, 23 of them were primary and 3 were secondary IHS. While, the IHS of the 3 cases were occurred secondary to spinal surgery, mild traumatic brain injury and severe dehydration. Twenty-six normal cases which exactly match with the gender and the age of the patient group were collected as a control group. The twenty-six cases were incorporated in the control group, including 8 males and 18 females whose ages were also 18–78 years old (mean age, 37.4 years old), and their clinical courses were also 2 days to 1 years. These subjects had no history of specific headache and other related neurological symptoms.

2.2. MRI acquisition and evaluation MRI scanning was performed with Siemens avento 1.5 T and verio 3.0 T scanner. Quadrature coils were used to obtain the brain MRI images. The scanning sequence contained Sagittal T1 WI (TE (time of echo) 7.8 – 8.5 ms, TR (time of repetition) 450–2000 ms), T2 WI (TE 92–98 ms, TR 3800– 6000 ms), T1 WI (TE 7.5–8.5 ms, TR 400–6000 ms), FLAIR (TE 794–109 ms, TR 8500–9000 ms), and DWI (TE 75–86 ms, TR 3100–6400 ms), and the enhancement scanning (contrast agent Gd-DTPA (0.1 mmol/kg) was also applied, and after the injection, axial, sagittal and coronal positions were scanned immediately). MRI images were respectively read by 2 associate chief physicians who had ten years of experiences in MRI clinical diagnosis. Five MRI findings involving dural thickening and enhancement, pituitary enlargement, subdural effusion (hematocele), venous engorgement and brain sagging were determined by both of them, and then the evaluation results were recorded, respectively.

2.3. Measurement of mamillopontine distance and pontomesencephalic angle Mamillopontine distance was defined as the distance between the inferior border of the mammillary body and the superior border of the pons. Pontomesencephalic angle was defined as the angle between the leading edge tangent line of the midbrain and the tangent line of superior border of pons, and the measurement was based on the median sagittal plane (Fig. 1C1, C2). The mamillopontine distance and pontomesencephalic angle of the patient group and the control group were measured by the above two physicians. The results were presented the average of their measurements [13,14]. 2.4. Statistical analyses All statistical data were analyzed by SPSS 16.0 software. The Kappa consistency test was used to assess the agreement between the evaluation results and measured data of the MRI findings of the two physicians. The paired t-test was used to evaluate the difference of the mamillopontine distance and pontomesencephalic angle between the patient group and the control group. The ROC curve was used to assure the diagnostic efficiency of the quantitative indicators, and the optimal diagnostic thresholds were respectively determined according to the maximumYouden’s index. 3. Results 3.1. Clinical characteristics All of the 26 patients had typical orthostatic headache, which would aggravate when standing upright and would be remitted within 30 min after lying down [15]. Pain location of these patients: frontal part (7 cases, 26.9%), tempus (6 cases, 23.1%), frontotemporal part (5 cases, 19.2%), parietal-occipital part (5 cases, 19.2%), and whole headache (3 cases, 11.5%). Besides, there were 8 (30.8%) patients suffered from nausea and vomiting (30.8%), 5 patients suffered from neck rigidity (19.2%), 3 patients suffered from dizziness (11.5%), 3 patients suffered from ambiopia (11.5%), and 1 patient suffered from tinnitus (4%). 3.2. Clinical evaluation results of the MRI findings Both of the two physicians proposed that none patients had these five findings at the same time, and 3 patients had none of the five findings. As shown in Fig. 1A and B, dural thickening and enhancement was the most common sign, sequentially followed by pituitary enlargement, subdural effusion (hematocele), venous engorgement and brain sagging in sequentially. However, the evaluation results between the two physicians were not completely consistent (Table 1). The results of dural thickening and enhancement, subdural effusion (hematocele), pituitary enlargement and venous engorgement were highly consistent ( value = 0.808, 1 and

74

W. Tian et al. / Clinical Neurology and Neurosurgery 141 (2016) 71–76

Table 2 mamillopontine distance and pontomesencephalic angle of the patients group and the control group (¯x ± s). Groups

Mamillopontine distance (mm)

Pontomesencephalic angle (◦ )

The patient group The control group t P-value

5.4 ± 1.6

47.8 ± 8.7

6.9 ± 1.1

61.0 ± 6.1

−4.563 <0.01

−6.329 <0.01

Note: Compared with the control group, P < 0.01 indicated high significant difference.

Fig. 2. The area under ROC curve of mamillopontine distance and pontomesencephalic angle were 0.774 and 0.908, respectively; the diagnostic sensitivity and specificity were respectively 73.1% and 73.1% when the diagnostic threshold of mamillopontine distance was 6.4 mm; the diagnostic sensitivity and specificity were respectively 76.9% and 96.9% when the pontomesencephalic angle was 51.7◦ .

0.906, P < 0.01). And the results of brain sagging were medium consistency ( value = 0.606, P < 0.01). 3.3. Results of the mamillopontine distance and pontomesencephalic angle The mamillopontine distance and pontomesencephalic angle measured by the two physicians were highly consistentency ( value = 0.919, P < 0.01;  value = 0.808, P < 0.01). The mamillopontine distance and pontomesencephalic angle between the patient group and the control group are shown in Table 2 and Fig. 1 (C1, C2, D1 and D2). The mamillopontine distance and pontomesencephalic angle of the patient group were obviously less than those of the control group, and there were statistically significant differences between the two groups (P < 0.01). The area under ROC curve of pontomesencephalic angle was 0.908, and the area under ROC curve of mamillopontine distance was 0.774 (Fig. 2). Therefore, the diagnostic value of pontomesencephalic angle for IHS was higher than that of mamillopontine distance. According to the largest Youden’s index, when the diagnostic threshold of mamillopontine distance was set as 6.4 mm, and the sensitivity and specificity were respectively 73.1% and 73.1%. And, when the diagnostic threshold of pontomesencephalic angle was set as 51.7◦ , the sensitivity and specificity were respectively 76.9% and 96.2%.

4. Discussion 4.1. Possible causes and pathological mechanisms of IHS IHS was firstly reported by Schaltenbrand, a German doctor of Neurology, in 1938. Orthostatic headache, the main clinical manifestation caused by IHS, was usually misdiagnosed as migraine headache and/or mental abnormality which may delay effective treatments. IHS was divided into two types: primary IHS and secondary IHS. The causes of primary IHS are uncertain, but there are three reported possible causes [1,16]: (i) the production of cerebral spinal fluid (CSF) is insufficient due to the paracrisis. (ii) CSF reflux is too much, and/or too fast. (iii) CSF abnormally leaks. To date, the first two causes have not been confirmed. At present, most scholars generally consider that the abnormal leakage of CSF caused by tiny meanings tear is the main reason, which has already been imageologically proved [7,17]. Secondary IHS is commonly seen after lumbar puncture, and commonly caused by shock, dehydration, severe infection, sleeping pill poisoning, craniocerebral trauma, surgery and so on. In the patient group, 23 IHS patients were primary and another 3 cases were secondary, which was respectively caused by lumbar puncture, craniocerebral trauma and serious dehydration. Without sufficient understanding of IHS at the beginning of the research, neither MRI examination of cervical, thoracic and lumbar vertebra of the primary IHS nor determination of CSF leakage was performed. However, these will be explored in the near future. Orthostatic headache, a characteristic symptom of IHS, is generally located in the frontal–temporal–occipital part and sometimes spread to the whole head, or neck and/or shoulder. It is often accompanied by nausea, vomiting, vertigo, tinnitus, hearing loss, blurred vision, double vision, nuchal rigidity and so on. In the patient group, a total of 26 patients had the performance of orthostatic headache. The main pain states of the patients were the frontal–temporal–occipital part, and only 3 of 26 patients were whole headache. Meanwhile, common associated symptoms were also observed, including nausea and vomiting (8 cases, 30.8%), neck rigidity (5 cases, 19.2%), dizziness (3 cases, 11.5%), ambiopia (3 patient, 11.5%) and tinnitus (1 cases, 4%). These associated symptoms was aggravated by standing and/or sitting, and became relieved or disappeared by prostrating, which were similar to reported literature [16]. The mechanisms of the orthostatic headache were as follows: (i) the CSF pressure decreased because of the reduced volume of CSF, which influenced its fluid support to brain tissue, and ten to stretch pain-sensitive structures (meninges, blood vessels, trigeminal nerve, glossopharyngeal, vagus nerve and so on) which were anchored on the brain surface. These led to orthostatic headache. (ii) compensatory intracranial hyperemia could also generate the orthostatic headache [18]. (iii) the presence of the reduction of labyrinth pressure and the changes of cochlear canal pressure, which were caused by lower CSF pressure, could also generate dizziness and tinnitus [16,19]. 4.2. MRI findings and their pathological mechanisms MRI findings are one of the diagnostic criteria of IHS, mainly including dural thickening and enhancement, pituitary enlargement, subdural effusion (hematocele), venous engorgement and brain sagging. According to the Monro–Kellie theory [18], the sum of brain tissue volume, CSF volume and cerebral blood volume is a constant. Because the volume of CSF decrease and the volume of brain tissue is pretty constant, a compensatory increase in cerebral blood volume is inevitably occurred, and the expansion of cerebral venous system is following as the earliest manifestation. The vessel of leptomeninges and brain parenchyma has a defect bloodbrain barrier (BBB), but dura does not. Therefore, the dural vein

W. Tian et al. / Clinical Neurology and Neurosurgery 141 (2016) 71–76

expansion induced by compensatory hyperemia and the increase of permeability of capillaries happen, which will lead to the enhancement of the accumulation of MRI contrast agent in micro-vascular and interstitium of dura. And the enhancement will cause enhancement of dural thickening in MRI examination. Moreover, subdural effusion, cerebral venous sinus dilation and pituitary enlargement were compensatory responses of CSF leakage, and massive CSF leak would lead to brain sagging. The discussions of the different MRI findings in separate compartments were as follows: 4.2.1. Dural thickening and enhancement MRI scan of dura suggested uniform T1 WI signal, and slight increase in T2WI signal. Enhanced MRI scan suggested obviously dural enhancement, and the enhanced dura maintained a smoother finished surface with no nodules. However, the MRI of leptomeninges of the surface of the brainstem and cerebral sulcus showed no enhancement. The finding appeared as the earliest finding of MRI, also owned the highest positive rate, and was reversible [20,21]. It was reported that the incidence of the finding was about 56–80%, and even up to 100% [7]. And the incidence of dural thickening and enhancement was observed in MRI images of 23 patients (88.5%), which was consistent with the above reference. 4.2.2. Subdural effusion (hematocele) Brain MRI scan demonstrated a crescent-shaped area between skull inner plate and the dura, showing low signal intensity on T1WI and high signal intensity on T2WI. The changes of the MRI signal were complex showing high signal intensity on T1WI, when suffered from a bleed. Schievink et al. reported that the incidence of subdural effusion (SDE) was 36–50% [9]. There were 10 of 26 cases developed varying degree of subdural effusion, and 1 of these 10 cases showed hematocele. Pituitary enlargement presented as enlargement of pituitary body [8]. Venous engorgement presented as the engorgement of venous sinus [10]. Brain sagging presented as descent of midbrain, narrowing of cisterna chiasmatis and prepontine cistern, and descent of cerebellar tonsil and the fourth ventricle. The above comparatively rare results of these MRI findings depended on the physician’s experience and subjective judgment, lacking objective diagnostic criteria. The result of the assessment by the two physicians was not completely consistent, especially the finding of brain sagging. 4.3. Diangostic values of quantitative measurement of mamillopontine distance and pontomesencephalic angle in IHS

75

those of the control group were 7 ± 1.3 mm and 65 ± 9.9, respectively [14]. In this research, the mamillopontine distance and pontomesencephalic angle of the patient group were 5.4 ± 1.6 mm and 47.8 ± 8.7, and both of them were obviously decreased compared to the control group whose the mamillopontine distance and pontomesencephalic angle were 6.9 ± 1.1 mm, 61.0 ± 6.1, respectively. Meanwhile, the difference between the two groups in this research was statistically significant, and the results of this research were similar to those of Shah et al. [14]. The diagnostic sensitivity and specificity were respectively 73.1% and 73.1% when the diagnostic threshold of mamillopontine distance was 6.4 mm; the diagnostic sensitivity and specificity were respectively 76.9% and 96.9% when the pontomesencephalic angle was 51.7◦ . The mamillopontine distance and pontomesencephalic angle of 1 of the 3 patients, whose MRI findings of brain sagging were controversial by the 2 physicians, was 6.3 mm and 50.3◦ , respectively, which were lower than the diagnostic threshold. Combined with the MRI, both of the physicians agreed that brain sagging was presented. However, the 2 physicians thought that brain sagging did not appear in another 2 patients whose mamillopontine distance and pontomesencephalic angle were higher than the diagnostic threshold. Accordingly, based on the results of this study that the diagnostic threshold of mamillopontine distance was 6.4 mm and the pontomesencephalic angle was 51.7◦ , brain sagging was cited to exist when mamillopontine distance and/or pontomesencephalic angle approached the diagnostic thresholds. Of the two objective criteria evaluated in this study, the mamillopontine distance and pontomesencephalic angle are useful when the qualitative MRI findings are variable. None of the patients displayed all five qualitative MRI features, in such a situation, physicians could have more confidence to make the diagnosis of IHS on the basis of quantitative data of mamillopontine distance and pontomesencephalic angle. 4.4. The limitations of this study It was necessary to perform double-blind study with large sample size in order to enhance the accuracy, not only because this study was retrospective study and dis-double-blind trail, but also because of its smaller sample size. Besides, no further investigation about the reasons of the existence of CSF leakage in primary IHS patients was performed; hence the causal relationship between CSF and IHS is still in need of further study. Conflict of interest

Brain sagging, which could not be early diagnosed, was a characteristic MRI finding in patients with IHS. Therefore, early detection of brain sagging was of great significance in early diagnosis and treatment of IHS. In this research, 3 patients, whose clinical courses were 8–12 months, had the characteristic features of brain sagging in MRI, which were approved by the 2 physicians. However, it remained controversial between the 2 physicians whether another 3 patients had brain sagging. Early brain sagging might only show mild narrowing of cisterna chiasmatis and prepontine cistern. However, due to different clinical experience and diagnostic criteria of the 2 physicians, the differences of evaluation results might exist. Therefore, objective quantitative indicators could help to reduce the subjective difference. It was reported that mamillopontine distance and pontomesencephalic angle were the indicators of brain sagging [22], and the measurement of the two indicators could contribute to accuracy of the evaluation of brain sagging. Shah et al. had improved the accuracy of early diagnosis of IHS combining with the quantitative indicators, and the mamillopontine distance and pontomesencephalic angle of the patients were 4.4 ± 1.8 mm and 41.2 ± 17.4,

None. Acknowledgement This study is funded by grant H201262 from the Health Services Research Foundation of Jiangsu Province, China. References [1] B. Mokri, Spontaneous cerebrospinal fluid leaks: from intracranial hypotension to cerebrospinal fluid hypovolemia—evolution of a concept, Mayo Clin. Proc. 74 (1999) 1113–1123, Elsevier. [2] E. Eross, D. Dodick, K. Nelson, Orthostatic headache syndrome with CSF leak secondary to bony pathology of the cervical spine, Cephalalgia 22 (2002) 439–443. [3] J. Benito-León, M. Reina, J. Alvarez-Linera, [Intracranial hypotension syndrome], Neurologia (Barcelona, Spain) 16 (2001) 418–426. [4] E. Ferrante, A. Savino, G. Sances, G. Nappi, Spontaneous intracranial hypotension syndrome: report of twelve cases, Headache 44 (2004) 615–622. [5] M. Paldino, A.Y. Mogilner, M.S. Tenner, Intracranial hypotension syndrome: a comprehensive review, Neurosurg. Focus 15 (2003) 1–8.

76

W. Tian et al. / Clinical Neurology and Neurosurgery 141 (2016) 71–76

[6] I.F. Quintero, Á. Candamil, M. Mejía, J.H. antilla, H. Medina, C. Áriza, F. adena, Intracranial hypotension syndrome: Á post dural puncture headache? Revista Colombia na de Ánestesiologia 41 (2013) 57–60. [7] W. Schievink, M. Maya, C. Louy, F. Moser, J. Tourje, Diagnostic criteria for spontaneous spinal CSF leaks and intracranial hypotension, Am. J. Neuroradiol. 29 (2008) 853–856. [8] A.K. Firat, H.M. Karakas, Z.Y. Firat, B. Kahraman, G. Erdem, Spontaneous intracranial hypotension with pituitary adenoma, J. Headache Pain 7 (2006) 47–50. [9] W.I. Schievink, M.M. Maya, F.G. Moser, J. Tourje, Spectrum of subdural fluid collections in spontaneous intracranial hypotension, J. Neurosurg. 103 (2005) 608–613. [10] R. Farb, R. Forghani, S. Lee, D. Mikulis, R. Agid, The venous distension sign: a diagnostic sign of intracranial hypotension at MR imaging of the brain, Am. J. Neuroradiol. 28 (2007) 1489–1493. [11] J. Alvarez-Linera, J. Escribano, J. Benito-Leon, J. Porta-Etessam, A. Rovira, Pituitary enlargement in patients with intracranial hypotension syndrome, Neurology 55 (2000) 1895–1897. [12] K. Miyazawa, Y. Shiga, T. Hasegawa, M. Endoh, N. Okita, S. Higano, S. Takahashi, Y. Itoyama, CSF hypovolemia vs intracranial hypotension in spontaneous intracranial hypotension syndrome, Neurology 60 (2003) 941–947. [13] Y.H. Al-Brashdi, S. Raniga, S.R. Revati, Spontaneous intracranial hypotension with magnetic resonance localisation of spinal cerebrospinal fluid leak, Sultan Qaboos Univ. Med. J. 13 (2013), E611.

[14] L.M. Shah, L.A. McLean, M.E. Heilbrun, K.L. Salzman, Intracranial hypotension: improved MRI detection with diagnostic intracranial angles, Am. J. Roentgenol. 200 (2013) 400–407. [15] T.S. Engedal, H. Ørding, O.J. Vilholm, Changing the needle for lumbar punctures-results from a prospective study, Clin. Neurol. Neurosurg. (2015). [16] B. Mokri, Headaches caused by decreased intracranial pressure: diagnosis and management, Curr. Opin. Neurol. 16 (2003) 319–326. [17] A. Watanabe, T. Horikoshi, M. Uchida, H. Koizumi, T. Yagishita, H. Kinouchi, Diagnostic value of spinal MR imaging in spontaneous intracranial hypotension syndrome, Am. J. Neuroradiol. 30 (2009) 147–151. [18] B. Mokri, The Monro–Kellie hypothesis applications in CSF volume depletion, Neurology 56 (2001) 1746–1748. [19] D. Grimaldi, E. Mea, L. Chiapparini, E. Ciceri, S. Nappini, M. Savoiardo, M. Castelli, P. Cortelli, M. Carriero, M. Leone, Spontaneous low cerebrospinal pressure: a mini review, Neurol. Sci. 25 (2004) s135–s137. [20] O.C. Bruera, L. Bonamico, J.A. Giglio, V. Sinay, J.A. Leston, M.L. Figuerola, Intracranial hypotension: the nonspecific nature of MRI findings, Headache 40 (2000) 848–852. [21] T. Brightbill, R.S. Goodwin, R.G. Ford, Magnetic resonance imaging of intracranial hypotension syndrome with pathophysiological correlation, Headache 40 (2000) 292–299. [22] E.L. Yuh, W.P. Dillon, Intracranial hypotension and intracranial hypertension, Neuroimaging Clin. N. Am. 20 (2010) 597–617.