Imaging, Clinical, and Demographic Differences in Patients With Type III Spinal Cerebrospinal Fluid Leak (Cerebrospinal Venous Fistulas) Compared With Patients With Types I and II Spinal Cerebrospinal Fluid Leak : Journal of Computer Assisted Tomography

Secondary Logo

Journal Logo

Neuroimaging: Spine

Imaging, Clinical, and Demographic Differences in Patients With Type III Spinal Cerebrospinal Fluid Leak (Cerebrospinal Venous Fistulas) Compared With Patients With Types I and II Spinal Cerebrospinal Fluid Leak

Mehan, William A. Jr. MD, MBA; Buch, Karen MD

Author Information
Journal of Computer Assisted Tomography 46(6):p 986-990, 11/12 2022. | DOI: 10.1097/RCT.0000000000001369
  • Free

Abstract

Four types of spinal cerebrospinal fluid (CSF) leaks have been described with fast leaks characterized as type I (dural tears) and type II (meningeal diverticular tears).1–3 Type III CSF leaks related to CSF-venous fistulas (CVFs) are generally considered to be slow spinal CSF leaks.1,2 A type IV CSF leak is considered indeterminate or unknown.1,2 Types I and II leaks are most common, with type III leaks being less common and underrecognized.1–3 Type III spinal CSF leaks were first described in 2014 and are an increasingly recognized cause of spontaneous intracranial hypotension.4–14 Because these type III leaks are considered to be slower in nature, they do not necessarily result in epidural fluid collections and can be more difficult to detect compared with types I and II CSF leaks.15

Much has been previously described in the literature related to use of dynamic myelography, computed tomography myelography (CTM), and digital subtraction myelography to detect types I to III spinal CSF leaks.16–18 Various myelographic findings can be used to help distinguish between types I and III CSF leaks with hyperdense paraspinal veins being a myelographic finding associated with CVFs.17,18 These findings are often subtle, and a high degree of pr-myelographic suspicion for suspected CVF is important when evaluating patients with intracranial hypotension related to suspected CSF leak.

To date, there remain limited data on the intracranial imaging findings associated with type III spinal CSF leaks compared with types I and II spinal CSF leaks. Prior studies have described imaging features associated with spontaneous intracranial hypotension related to spinal CSF leak, which include engorged venous sinuses, pachymeningeal enhancement, presence of subdural fluid collections, decreased pontomammilary distance, and transtentorial herniation.2 In this study, we investigate intracranial and spinal imaging findings in patients with myelographically and surgically proven type III CVFs compared with those of the more common types I and II CSF leaks.

MATERIALS AND METHODS

This was a retrospective, institutional review board–approved study performed at a single institution. Consecutive patients undergoing evaluation of suspected CSF leak were included in this study. Specific inclusion criteria were (1) adult patients with surgically confirmed type I, type II, or type III CSF leak, (2) preoperative brain magnetic resonance imaging (MRI) with intravenous contrast, (3) preoperative total spine MRI, and (4) medical record documentation of symptom onset, spectrum of symptoms, and initial clinical diagnosis provided at the time of symptom onset and final clinical diagnosis. Patients were excluded from this study if they had nondiagnostic brain and/or spine MRI examinations, history of prior craniotomy/craniectomy, history of intracranial neoplasm, history of recent trauma (within the last 2 months), or incomplete medical records. Basic demographic information and clinical information including symptom onset, spectrum of symptoms, degree of disability, and initial and final clinical diagnoses were recorded. Brain MRI examinations were evaluated for the following findings: (1) presence of extra-axial fluid collections, (2) pachymeningeal thickening, (3) reduced pontomamillary distance (<5 mm), (4) low lying cerebellar tonsils, and (5) transtentorial herniation.2 Specific measurements were recorded from the MRI brain examinations including a measurement of inferior cerebellar tonsillar descent with respect to the horizontal plane of foramen magnum and a measurement of the pontomamillary distance (with normal range considered between 6.5 and 7.5 mm).19 Both of these measurements were recorded off the sagittal T1-weighted sequences. Spine magnetic resonance (MR) images were reviewed to detect the presence of spinal longitudinal extradural CSF collection (SLEC).17 All neuroimaging studies were reviewed independently by two board-certified neuroradiologists blinded to the patients' clinical data and surgical findings.

Unpaired, 2-tailed t tests were used to compare the intracranial and spinal manifestations of type III CSF leaks with those of type I and type II CSF leaks.

RESULTS

A total of 23 patients met the inclusion criteria for this study with 7 patients having a confirmed type III CSF leak and 16 patients having a confirmed type I or II CSF leak. There were nearly equal distributions of males and females with type III CSF leak (57% males and 43% females) compared with type I/II CSF leak (44% males and 56% females).

Patients with type III leaks were significantly older (mean, 59.4 years; SD, 7.7 years) than those with types I and II leaks (mean, 40 years; SD, 11 years) (P = 0.0003).

Similar frequencies of clinical symptoms including headaches, positional headaches, vision changes, nausea/vomiting, diplopia, and extremity tingling were seen between type III and type I/II spinal CSF leaks. A higher frequency of dizziness, fatigue, gait imbalance, and inability to work were encountered in patients with type III spinal CSF leak compared with type I/II spinal CSF leaks Table 1.

TABLE 1 - Clinical Symptoms in Patients Type III Versus Type I/II Leaks
Type III CSF Leak Type I/II CSF Leak
Clinical Symptoms No. Cases Percentage No. Cases Percentage
Headache 6 86% 16 100%
Positional changes 5 71% 13 81%
Vision changes 3 43% 6 38%
Nausea 2 29% 6 38%
Numbness 0 0% 0 0%
Weakness 0 0% 1 6%
Diplopia 1 14% 3 19%
Dizziness 3 43% 3 19%
Syncope 0 0% 0 0%
Fatigue 2 29% 0 0%
Gait imbalance 2 29% 2 13%
Bowel/bladder symptoms 0 0% 0 0%
Tingling 1 14% 3 19%
Exercise intolerance 2 29% 0 0%
Inability to work 2 29% 3 19%

Patients with type III CSF leaks had a higher frequency of initial misdiagnosis (100% of cases) compared with types I and II CSF leaks (31% of cases) and had a significantly longer duration of symptoms until accurate diagnosis (1111 days vs 280 days, P = 0.03).

Intracranial MR findings in type III spinal CSF leaks had no instances of extra-axial fluid collections compared with 50% of type I/II spinal CSF leaks (Figs. 1, 2). Similarly, extradural spinal collections were never seen in patients with type III spinal CSF leaks compared with 100% with type I or II spinal CSF leaks (Figs. 1, 2). Pachymeningeal thickening, decreased pontomammilary distance, and transtentorial herniation were seen with nearly equal frequency between patients with type III spinal CSF leak and type I or II spinal CSF leaks (50% vs 57%, respectively). A total of 5 patients with type III spinal CSF leaks had significantly reduced pontomamillary distances with a mean pontomamillary distance of 2.8 mm. This was compared with a total of 5 patients with type I or II spinal CSF leak who had reduced pontomamillary distance with a mean distance of 4.8 mm (P = 0.047). Inferior cerebellar tonsillar descent was exclusively seen in patients with type III spinal CSF leak and was not encountered in any of the cohort cases with type I/II spinal CSF leak. The mean distance of inferior cerebellar tonsillar descent for patients with type III spinal CSF leak was 4.6 mm compared with 0 mm for patients with type I/II spinal CSF leak. This finding was statistically significant (P < 0.002). These findings are described in Table 2.

F1
FIGURE 1:
A 56-year-old woman with positional headaches and diplopia initially diagnosed with Chiari malformation. Sagittal constructive interference in a steady state sequences through the cervical, thoracic, and lumbar spine (images A–C) did not demonstrate a SLEC. In image A, the cerebellar tonsils are noted to lie 7 mm below the foramen magnum (white arrow). Magnetic resonance imaging of the brain was performed with an axial T2-weighted image (D), which demonstrated no intracranial extra-axial collection. Postcontrast T1-weighted image (E) demonstrates diffuse pachymeningeal thickening. A subsequent dynamic CTM demonstrated a CVF associated with the right T9 nerve root shown on the axial CTM image F and parasagittal reformatted image G (black arrows). A CVF at this level was confirmed at the time of operative repair.
F2
FIGURE 2:
A 36-year-old man with sudden onset development of a severe positional headache with blurry vision and diplopia, initially diagnosed with intracranial hypotension. Sagittal tau inversion recovery image (A) demonstrates ventral and dorsal SLECs (white arrowheads). Note that on this same image (A) the cerebellar tonsils are not low lying. A brain MRI performed with a T2-weighted image (B) demonstrates small, bilateral intracranial extra-axial collections, and smooth, pachymeningeal thickening and enhancement seen on the axial postcontrast T1 image (C). A noncontrast spine computed tomography shown in image D demonstrates a right dorsolaterally located bony spicule (black arrowhead) at the T1 and T2 level. The subsequently performed dynamic CTM (image E) demonstrates extradural contrast extravasation (black arrow) on the right aspect of the thecal sac at the T1 and T2 level in the region of the bony spicule. A dural tear was confirmed at this level at the time of surgical repair.
TABLE 2 - Intracranial Imaging Markers of CSF Hypotension in Patients With Type III Versus Type I/II CSF Leaks
Intracranial MRI Findings Type III CSF Leak Type I/II CSF Leak P
No. Cases Percentage No. Cases Percentage
Extra-axial fluid collections 0 0% 8 50%
Pachymeningeal thickening 5 71% 13 81%
Pontomamillary distance <5 mm 5 (mean, 2.8 mm) 40% 5 (mean, 4.8 mm) 30% 0.027
Transtentorial herniation 4 57% 8 50%
Inferior cerebellar tonsils descent 3 (mean descent, 4.6 mm) 43% 0 (all above FM) 0% 0.0021
FM indicates foramen magnum.

DISCUSSION

Several differences were noted between the cohort of patients with type III spinal CSF leaks compared with types I and II spinal CSF leaks, which may potentially be helpful when evaluating a patient with suspected spinal CSF leak. Patients with type III spinal CSF leaks were significantly older compared with types I and II spinal CSF leaks, with the youngest type III spinal CSF leak patient in this cohort being 50 years old compared with 27 years old in the type I/II spinal CSF leak cohort. With respect to intracranial imaging findings, patients with type III spinal CSF leak did not have SLECs but had significantly smaller pontomamillary distances and lower cerebellar tonsils compared with that of types I and II spinal CSF leaks. In this cohort, no patient with type III spinal CSF leak had a SLEC, whereas all patients with type I/II CSF leak had evidence of at least 1 SLEC on spine MRI. This is similar to previous studies demonstrating that SLECs are seen exclusively in the setting of dural tears (types I and II spinal CSF leaks).1,8,15,18,20,21 These findings are potentially important, as patients with suspected spinal CSF leak will frequently undergo brain and total spine MRI before myelography to find the potential site of spinal CSF leak. Different myelographic techniques may be pursued if a type III spinal CSF leak is suspected over a type I/II spinal CSF leak. For example, at our institution, in the absence of a SLEC on MRI, a standard computed tomography myelogram will be performed first. If that standard CTM is negative, lateral decubitus digital subtraction myelography will be performed next for the evaluation of suspected CVF similar to previously described techniques.15–18,20–27 Conversely, patients with a positive finding of a SLEC on MRI may proceed to a prone digital subtraction myelography or dynamic CTM depending on the presence and location of suspicious endplate osteophytes or bony spicules suspicious for the site of dural tear on preprocedural noncontrast computed tomography spine imaging.15,20–22

The lack of SLECs in type III CSF leaks is possibly due to the slow nature of the leak.15 The observation that intracranial subdural collections were absent in all patients with type III leaks may also be related to the slow leak and rate of development of intracranial hypotension. Decreased pontomamillary intervals and lower cerebellar tonsillar position observed in patients with type III CSF leaks is of unclear etiology but could be related to lower CSF volumes and longer disease course before treatment compared with type I/II spinal CSF leaks.23–25 Further investigations are necessary to explain these findings.

With respect to clinical symptoms, similar frequencies of headache, vision changes, diplopia, and other symptoms occurred between type III and type I/II spinal CSF leaks, which is supported by clinical observations in prior studies and case reports.

There are limitations to this study. This is a retrospective study performed at a single medical center. Furthermore, this is a small cohort study, as spinal CSF leaks are an uncommon phenomenon with type I/II CSF leaks being more common than type III spinal CSF leaks. Because of these small cohort numbers, the data presented in this study may not be generalizable to a larger population. In addition, assessment of certain imaging findings including a measurement of the pontomamillary distance is quite a small value and is therefore potentially susceptible to measurement error.

In conclusion, the results of this study suggest demographic and imaging-based differences between patients with type III versus type I/II spinal CSF leaks, noting that type III spinal CSF patients tended to be older in age, lacked intracranial extra-axial fluid collections and SLECs, had smaller pontomamillary distances, and had a greater degree of inferior cerebellar tonsillar descent. To our knowledge, a direct comparison of intracranial imaging manifestations between type III and type I/II spinal CSF leaks has not been previously reported. These findings may potentially be important for myelographic planning and optimization for localization of site of suspected CSF leak in patients with intracranial hypotension.

REFERENCES

1. Schievink WI, Maya MM, Jean-Pierre S, et al. A classification system of spontaneous spinal CSF leaks. Neurology. 2016;87:673–679.
2. Luetzen N, Dovi-Akue P, Fung C, et al. Spontaneous intracranial hypotension: diagnostic and therapeutic workup. Neuroradiology. 2021;63:1765–1772.
3. Kranz PG, Gray L, Malinzak MD, et al. Spontaneous intracranial hypotension: pathogenesis, diagnosis, and treatment. Neuroimaging Clin N Am. 2019;29:581–594.
4. Schievink WI, Moser FG, Maya MM. CSF-venous fistula in spontaneous intracranial hypotension. Neurology. 2014;83:472–473.
5. Kranz PG, Gray L, Malinzak MD, et al. CSF-venous fistulas: anatomy and diagnostic imaging. AJR Am J Roentgenol. 2021;217:1418–1429.
6. Mamlouk MD, Shen PY, Sedrak MF. Spontaneous intracranial hypotension in the critical patient. J Intensive Care Med. 2022;37:618–624.
7. Carlstrom LP, Oushy S, Graffeo CS, et al. Intracranial hypotensive crisis from an insidious spinal cerebrospinal fluid-venous fistula: a case report. Oper Neurosurg (Hagerstown). 2021;21:E283–E288.
8. Schievink WI, Maya M, Prasad RS, et al. Spontaneous spinal cerebrospinal fluid-venous fistulas in patients with orthostatic headaches and normal conventional brain and spine imaging. Headache. 2021;61:387–391.
9. Urbach H, Fung C, Dovi-Akue P, et al. Spontaneous intracranial hypotension. Dtsch Arztebl Int. 2020;117:480–487.
10. Nagesh CP, Devaraj R, Joshi G, et al. The importance of repeat digital subtraction myelography in the diagnosis of cryptogenic CSF-venous fistula causing spontaneous intracranial hypotension. Interv Neuroradiol. 2021;27:727–732.
11. Shlobin NA, Shah VN, Chin CT, et al. Cerebrospinal fluid-venous fistulas: a systematic review and examination of individual patient data. Neurosurgery. 2021;88:931–941.
12. Tatum PS, Anderson E, Kravtsova A, et al. Rapid cognitive decline secondary to csf venous fistula with postoperative rebound intracranial hypertension and a hyperintense paraspinal vein sign seen retrospectively [published online September 12, 2020]. Mil Med. doi: 10.1093/milmed/usaa204.
13. Pope MC, Carr CM, Brinjikji W, et al. Safety of consecutive bilateral decubitus digital subtraction myelography in patients with spontaneous intracranial hypotension and occult CSF leak. AJNR Am J Neuroradiol. 2020;41:1953–1957.
14. Schievink WI, Maya MM, Moser FG, et al. Spontaneous spinal CSF-venous fistulas associated with venous/venolymphatic vascular malformations: report of 3 cases. J Neurosurg Spine. 2019;32:305–310.
15. Farb RI, Nicholson PJ, Peng PW, et al. Spontaneous intracranial hypotension: a systematic imaging approach for CSF leak localization and management based on MRI and digital subtraction myelography. AJNR Am J Neuroradiol. 2019;40:745–753.
16. Piechowiak EI, Pospieszny K, Haeni L, et al. Role of conventional dynamic myelography for detection of high-flow cerebrospinal fluid leaks: optimizing the technique. Clin Neuroradiol. 2021;31:633–641.
17. Madhavan AA, Kim DK, Brinjikji W, et al. Diagnosis of a cerebrospinal fluid-venous fistula associated with a venous malformation using digital subtraction and computed tomography myelography. World Neurosurg. 2020;135:262–266.
18. Kim DK, Brinjikji W, Morris PP, et al. Lateral decubitus digital subtraction myelography: tips, tricks, and pitfalls. AJNR Am J Neuroradiol. 2020;41:21–28.
19. Koehler PR, Haughton VM, Daniels DL, et al. MR measurement of normal and pathologic brainstem diameters. AJNR Am J Neuroradiol. 1985;6:425–427.
20. Hoxworth JM, Trentman TL, Kotsenas AL, et al. The role of digital subtraction myelography in the diagnosis and localization of spontaneous spinal CSF leaks. AJR Am J Roentgenol. 2012;199:649–653.
21. Thielen KR, Sillery JC, Morris JM, et al. Ultrafast dynamic computed tomography myelography for the precise identification of high-flow cerebrospinal fluid leaks caused by spiculated spinal osteophytes. J Neurosurg Spine. 2015;22:324–331.
22. Luetmer PH, Mokri B. Dynamic CT myelography: a technique for localizing high-flow spinal cerebrospinal fluid leaks. AJNR Am J Neuroradiol. 2003;24:1711–1714.
23. Luetmer PH, Schwartz KM, Eckel LJ, et al. When should I do dynamic CT myelography? Predicting fast spinal CSF leaks in patients with spontaneous intracranial hypotension. AJNR Am J Neuroradiol. 2012;33:690–694.
24. Kranz PG, Gray L, Amrhein TJ. Spontaneous intracranial hypotension: 10 myths and misperceptions. Headache. 2018;58:948–959.
25. Mokri B. Spontaneous cerebrospinal fluid leaks: from intracranial hypotension to cerebrospinal fluid hypovolemia—evolution of a concept. Mayo Clin Proc. 1999;74:1113–1123.
26. Kranz PG, Luetmer PH, Diehn FE, et al. Myelographic techniques for the detection of spinal CSF leaks in spontaneous intracranial hypotension. AJR Am J Roentgenol. 2016;206:8–19.
27. Clark MS, Diehn FE, Verdoorn JT, et al. Prevalence of hyperdense paraspinal vein sign in patients with spontaneous intracranial hypotension without dural CSF leak on standard CT myelography. Diagn Interv Radiol. 2018;24:54–59.
Keywords:

cerebrospinal fluid leak; intracranial hypotension; brain sagging; MRI

Copyright © 2022 Wolters Kluwer Health, Inc. All rights reserved.