français

DIAGNOSIS

Reviewed by Dr. Anish Kapadia MD

Last updated: {insert date}

 

Under-diagnosis and/or misdiagnosis of a spinal CSF leak/ Spontaneous intracranial hypotension (SIH) is extremely common, resulting in little or no treatment for some patients suffering from it. Currently, there is not a single diagnostic test that can rule out SIH with a high degree of certainty; therefore, detailed assessment and investigation of symptoms by a specialist is crucial.

 

Possible Misdiagnosis

A CSF leak is often misdiagnosed as one of the following conditions:

  • Chronic migraine

  • Chronic tension-type headache

  • Chiari malformation 1

  • Postural orthostatic tachycardia syndrome (POTS)

  • Dementia (in some very rare cases)

  • Cervicogenic headache

  • New daily persistent headache (NDPH)

  • Meniere's disease

  • Post-concussion syndrome

 

Symptoms

Experts agree that symptoms and history of the patient are extremely important for a successful diagnosis: see SYMPTOMS.

Dr. Carroll has devised a 48-hour Flat Test to assist physicians in determining the extent to which a patient’s headache has a postural component through which symptoms improve or worsen in response to body position (horizontal vs vertical).

 

Neurological and Physical Examination

The results of a neurological examination are usually normal, but may show some signs related to the neurological symptoms caused by a CSF leak.  These signs can vary but can include cognitive, visual, sensory, motor, balance and walking problems.

 

Blood Work

Blood work is usually normal in patients with CSF leaks.  There is currently no routine diagnostic test on blood work for spinal CSF leaks.

 

Non-targeted Epidural Blood Patch (EBP)

Sustained improvement of symptoms after epidural blood patching is a good indication of a spinal CSF leak. If the initial non-targeted EBP is unsuccessful, a second, larger-volume EBP is sometimes considered, and some recommend a third lumbar or even lower thoracic level EBP before proceeding with localization of the leak with more invasive testing (i.e., CT Myelography). The success of the EBP can vary depending on the type of leak and the chronicity of the leak.

 

Imaging Modalities for Diagnosis

Patients with spinal CSF leak typically undergo imaging to diagnose the condition, and may require additional invasive testing, or repeat imaging/testing to make the diagnosis.

There is no diagnostic imaging test that is 100% accurate in determining whether someone is leaking. Clinicians often rely on a combination of symptoms and imaging for diagnosis and further work-up. It is important to note that MRI imaging of patients with spinal CSF leak may sometimes be normal. Spinal CSF leak can exist without any evidence of a leak on MRI imaging, and thus a negative MRI of the brain or spine does not rule out a spinal CSF leak in a patient with high clinical suspicion based on symptoms. A combination of findings from the clinical assessment and MRI imaging determines the next steps in work-up and management.

Different types of imaging include:

  • Brain MRI (with and without contrast) should be done in all cases where a spinal CSF leak is suspected. At a minimum, the scan should include T1 post-contrast, T2 FLAIR, and SWI (susceptibility-weighted imaging) sequences. A T2 fat-suppressed sequence through the orbits, which allows measurement of the optic nerve sheath diameter and the peri-optic subarachnoid space (the fluid-filled space surrounding the optic nerve), has also been shown to be helpful. These measurements can provide additional clues when other imaging of the brain appears normal.

There are 5 main MRI brain abnormalities which can be remembered by the acronym SEEPS:

Subdural fluid collections
Enhancement of pachymeninges (also known as dural enhancement)
Engorgement of venous structures
Pituitary hyperemia
Sagging of the brain

The brain (Bern) SIH-Score (bSIH-Score) is a quantitative scoring system of cranial MRI used to indicate the likelihood of identifying a CSF leak during more specialized spinal imaging. This scoring system relies on the six most pertinent brain MRI findings (with three major factors assigned 2 points each (pachymeningeal enhancement, venous sinus engorgement, and effacement of the suprasellar cistern ≤ 4.0 mm) and three minor factors given 1 point each (subdural fluid collection, effacement of the prepontine cistern ≤ 5.0 mm, and mamillopontine distance ≤ 6.5 mm), enabling an evaluation of the probability of a positive spinal imaging outcome in individuals with SIH ranging from 0 to 9 (higher values are associated with higher possibilities of finding a CSF leak).  However, if the Bern Score suggests a low probability, patients should not be deprived of necessary evaluation and treatment if clinical suspicion remains high.

In simple terms, the Bern SIH Score is a point-based system that uses brain MRI findings to estimate how likely it is that specialized spinal imaging will find a CSF leak. A higher score means a higher chance of finding a leak. However, a low score does not mean there is no leak; clinical symptoms must always be considered.

 

  • A Full Spinal MRI  (Cervical, Thoracic, Lumbar) without contrast

A heavily T2-weighted, fat-saturated, high-resolution 3D sequence, also called non-invasive MR myelography, is considered the most useful spinal imaging sequence for evaluating a suspected CSF leak. This technique provides detailed views of the spinal fluid space without requiring any injection of contrast. It allows for superior detection of subtle epidural fluid, proper assessment of the lateral epidural space (which conventional protocols often miss), and clear visualization of meningeal diverticula (small outpouchings of the dural sac along the nerve root sleeves) and nerve root sleeve cysts along the spine.

A landmark 2019 paper by Farb et al. proposed a systematic imaging approach that categorizes spinal CSF leak based on whether there is a spinal longitudinal extradural CSF collection (SLEC), a collection of CSF fluid visible along the spine on MRI. The paper identified four types of spinal CSF leaks and demonstrated that the presence or absence of SLEC is key to determining the type of leak and guiding the diagnostic and treatment approach.

SLEC-Positive (Extradural Fluid Collection Present): These patients have a visible collection of CSF outside the dural sac along the spine. This typically indicates a mechanical dural tear.

Types include:

  • Type 1 (Ventral Dural Tear): A tear at the front of the spinal canal, often associated with a bone spur (calcified disc or osteophyte). This is the most common type of dural tear.

  • Type 2 (Lateral Dural Tear): A tear occurring at the side of the dural sac, at or near the nerve root sleeve.

 SLEC-Negative (No Extradural Fluid Collection): In these patients, no epidural fluid collection is visible on spinal MRI, even though brain MRI may show clear signs of intracranial hypotension.

Types include:

  •  Type 3 (CSF-Venous Fistula (CVF)): An abnormal direct connection between the spinal subarachnoid space and an adjacent epidural vein, allowing CSF to drain directly into the venous circulation rather than pooling in the epidural space. First described in 2014, CVFs have since become an increasingly recognized cause of SIH.

  • Type 4 (Distal Nerve Root Sleeve Leak): A leak occurring at the far end of a nerve root sleeve. This is the rarest type identified in the study.

 This systematic imaging approach using spinal MRI to first determine whether a SLEC is present predicts the nature of the underlying CSF leak and guides optimal patient positioning for myelography.

R.I. Farb et al. American Journal of Neuroradiology April 2019, 40 (4) 745-753; DOI: https://doi.org/10.3174/ajnr.A6016 (Shared with permission)

  • MR Myelography

MR myelography is a specialized non-invasive MRI that does not rely on the injection of contrast to view the spine.  Sometimes, it can be sufficient to localize a leak, and further invasive testing is not required.

  • Myelography

Digital Subtraction Myelogram (DSM), and Dynamic Decubitus CT Myelography are the study of choice for locating a leak once spinal CSF leak is suspected. Both of these studies involve injection of contrast dye into the spinal subarachnoid space via a lumbar puncture. Both modalities use X-ray radiation, whereas an MRI does not involve any radiation.

                   

  • Digital Subtraction Myelography (DSM)

DSM uses real-time fluoroscopy (a type of live X-ray) to take rapid images as the contrast dye flows through the spinal canal.  The "digital subtraction" part means that a baseline image taken before the contrast is injected is digitally removed (subtracted) from the images taken afterward. This removes the bones and other background structures from the picture, making it much easier to see exactly where contrast dye is escaping from the spinal fluid space.

 

  • Dynamic CT Myelography

Dynamic CT Myelography is an evolving technique in which the contrast injection and imaging is performed under CT. The main advantage is its ability to provide high-resolution cross-sectional images of the anatomy.

Patient positioning during DSM and dynamic CT myelography depends on the suspected type of leak. Patients with a positive SLEC (Type 1) are best positioned prone (face down).  Patients without a SLEC (Types 2, 3 and 4) are best evaluated in the lateral decubitus (lying on their side) position during myelography. 

 

  • Photon-counting CT myelography (PCCT)

PCCT is a next-generation scanning technology that is thought to offer sharper image detail through higher spatial and contrast resolution, while also exposing patients to less radiation than standard dynamic CT myelography. 

 

  •  Cone-beam CT (CBCT) myelography

CBCT, including newer ultra-high-resolution versions, is another advancing technology that can produce highly detailed cross-sectional images and may be performed immediately after digital subtraction myelography (DSM) to confirm or further clarify findings that are uncertain. 

 Key Notes:

  • Head CT scans have low sensitivity for detecting signs of spontaneous intracranial hypotension (SIH) and are not sufficient to properly evaluate for this condition.

  • About 15-20% of patients with spinal CSF leak(s) have normal MRI brain imaging.

  •  Radioisotope Cisternogram is rarely used in current practice, and it is not recommended as a diagnostic tool.

  •  A Lumbar puncture is not a helpful diagnostic tool and is not required to make a diagnosis. The majority of patients with SIH exhibit opening CSF pressures within the normal range. A normal opening pressure does not exclude the diagnosis.

  • For any imaging requiring a lumbar puncture, smaller size, pencil point atraumatic type needles, versus conventional cutting needles, can minimize the risk of creating further spinal CSF leaks during testing.

  •  While diverticula can occur in people without CSF leaks, a greater number or larger diverticula may be associated with an increased likelihood of a CSF-venous fistula. Some leaks (particularly Type 2 and Type 3) originate from or near these diverticula.

 

Recorded video from the 2nd Canadian SIH Conference: “MRI Brain & Spine: Findings, Bern score, Optic Nerve Sheath Diameter, Subtle Findings, Optimal Protocol.”

Recorded video from the 2nd Canadian SIH Conference: “Digital Subtraction Myelography: Utility/optimal technique for different CSF leak types, tricks and pitfalls.”

Recorded video from the 2nd Canadian SIH Conference: “Static CT Myelography, Dynamic CT Myelography & Photon Counting CT Myelography: When to use each, optimal techniques, pros and cons.”

References and Suggested Readings: