The term spontaneous intracranial hypotension (SIH) describes a clinical condition characterized by an incapacitating orthostatic headache syndrome caused by a spontaneous leakage of the cerebrospinal fluid (CSF). More often than not, it is accompanied by stigmata of findings on intracranial magnetic resonance imaging (MRI). In 1939, Schaltenbrand wrote about the incidence of spontaneous CSF hypotension leading to various clinical symptoms, most notably positional headache. He also denoted that the entity itself had been discussed in the French literature twenty years earlier, using several terms such as ‘hypotension of the spinal fluid’ and ‘ventricular collapse’ [1]. The availability of MRI has raised the interest in SIH in the past few decades. While historically perceived to be a rare entity, SIH is now being diagnosed more often, with the incidence currently estimated at 5 per 100,000 annually [2]. However, considering the high occurrence of misdiagnosis—particularly at the initial stage of clinical presentation—the true incidence is likely higher [3]. Although rarely encountered in children, SIH may occur at any age, with the mean age of diagnosis being at around 40 years. Relating this condition, there is female predilection, with a female–male ratio of 2:1 ^[4][5][6][4,5,6]. Additionally, a higher incidence of SIH is observed in certain connective tissue disorders, particularly the Marfan syndrome and the Ehlers–Danlos syndrome ^[7][8][7,8]. According to the International Classification of Headache Disorders (ICHD) 3rd edition, SIH is diagnosed when there is a spontaneous onset headache, which develops in temporal relation to the occurrence of CSF hypotension (opening CSF pressure < 60 mm H₂O); or evidence of CSF leakage (proven through imaging) [9]. However, more recent evidence has challenged the traditional conception of SIH as a condition caused by low CSF pressure.

2. Pathophysiology

The syndrome of SIH, in general, is a consequence of spontaneous CSF leaks along the neuraxis, almost invariably from the thecal sac of the spinal column ^[10][11][12][15,16,17]. Spontaneous leaks from the skull base are extremely rare ^[10][13][15,18]. Importantly, it has a distinct pathology from postdural puncture headaches or CSF hypovolemic conditions due to postoperative CSF loss or trauma [9]. Although the exact mechanism of spontaneous CSF leaks remain unknown, it is suspected that an underlying weakness of the spinal meninges is an imperative factor ^[14][19]. It creates a predisposition for the development of meningeal diverticula or perineural cysts, and subsequently spontaneous dural dehiscence. The presence of calcified degenerative thoracic disc protrusions may cause trivial traumatic events leading to progressive dural tears, which can be either ventral or lateral ^[7][14][15][7,19,20]. The CSF–venous fistula (CVF) is an important cause of SIH, which pose a diagnostic challenge ^[16][21]. Another rare, but interesting cause is the “nude nerve root” phenomenon, which describes the congenital absence of the entire nerve root sleeve ^[14][19]. The prevalence of SIH in certain connective-tissue disorders, particularly the Marfan syndrome, the Ehlers–Danlos syndrome (type II), and autosomal dominant polycystic kidney disease (ADPKD), are presumably due to the pre-existing dural fragility ^[7][8][13][7,8,18]. The old theories of SIH as a condition caused by CSF hypotension have been found to be inaccurate. A series of recent studies on larger samples by Luetmer et al. (2012), Yao et al. (2016), and Kranz et al. (2017), involving 388 patients, proved that only 21% to 55% of SIH patients had a low CSF opening pressure ^[17][18][19][22,23,24]. Following the data, it has been postulated that the reduction in CSF volume—rather than CSF pressure—is the primary pathogenetic factor ^[10][13][20][15,18,25]. Kranz et al. further advocated the concept of tissue compliance that determines the physiological relationship between pressure and volume ^[10][15]. Compliance with regards to the CSF component is highly dynamic and variable between individuals, as it is influenced by posture (upright vs. recumbent), body habitus, and the extent of the spinal epidural venous plexus ^[21][22][23][26,27,28]. It has also been demonstrated that the removal of CSF (via lumbar puncture) changes the spinal compliance due to the reduced spinal CSF volume ^[24][29]. Considering the fact that the rate of CSF loss is also different between individuals depending on mechanical factors, this may explain why not all SIH patients will have a low CSF opening pressure, despite a common unifying problem of low CSF volume. In 2019, an international jury of experts awaiting new radiological criteria confirmed the historic radiological definition of CM1 (cerebellar tonsillar position greater than or equal to 5 mm below the level of the foramen magnum) (agreement 83.7%) ^[25][32]. Different mechanisms may underlie the pathogenesis of CM1, primarily the underdevelopment of the posterior fossa bony structures. Several studies have shown that many, but not all, patients with CM1 have a small posterior fossa ^[26][33]. Another mechanism is represented by the presence of hemodynamic alterations that increase intracranial pressure. According to some authors, the delayed opening of the membrane covering the outlet of the fourth ventricle during fetal development could cause transient obstructive hydrocephalus with a consequent hernia of the hindbrain and cerebellar tonsils ^[27][34]. CM1 should be differentiated from cerebellar tonsil herniation secondary to space-occupying lesions (hydrocephalus, arachnoid cysts, brain tumors), and termed ‘acquired tonsillar ectopia’ by an international jury of experts (agreement 95.8%) ^[25][32]. Acquired tonsillar ectopia has an identical shape to the shape of the tonsils associated with idiopathic CM1, and may also be secondary to the downward displacement of the central nervous system due to decreased intrathecal pressure (either secondary to the over-drainage of lumboperitoneal shunting or a spontaneous CSF leak) ^[25][27][28][32,34,35]. In the latter case, careful clinical assessment, ascertaining postural headache, and thorough radiological assessment for intracranial CSF hypotension can help differentiate acquired Chiari malformation from true Chiari I malformation, and guide appropriate treatment (see paragraph 6). Although these mechanisms that act on the normal cerebellar tonsils, causing their deformation and herniation through the foramen magnum, are well understood, there is no consensus on the pathophysiology of CM1. For example, hydrocephalus has been proposed as both an etiologic cause and a consequence of a Chiari malformation.

3. Classification of SIH

Several classification systems have been proposed as guidance for clinical management, as well as for academic discussion ^[29][30][36,37]. The most widely used classification of SIH was introduced by Schievink et al. in 2016 ^[29][36]. This classification is based on the three recognized morphological types of leaks—namely the dural tear, the meningeal diverticulum, and the CSF–venous fistula (CVF)—in combination with the findings of a presence (or absence) of extradural CSF on spinal imaging ^[31][32][30,31] (Table 1).

4. Clinical Presentation of SIH

In a recent meta-analysis of 1694 SIH patients by D’Antona et al., headaches were almost invariably present—being reported in 97% of patients—and in the vast majority, they were orthostatic in nature [6]. This classic headache that characteristically worsens with an upright posture was most pronounced at the occipital region, although frontal and holocephalic headaches were also common. It may be attributed to the traction of the cranial nerves and the pain-sensitive dura mater, caused by the sagging of the brain in the CSF hypovolemia ^[3][33][34][3,38,39]. Typically, the headache is alleviated within 15–30 min after lying in a recumbent position ^[5][10][5,15]. However, the posture-related component of the headache may also vanish or dampen in a chronic condition ^[3][35][3,40]. Traction, distortion, or compression of some of the cranial nerves, certain structures of the brain, brainstem, mesencephalon, and diencephalon are believed to be the cause of many central nervous system manifestations, as well as the various cranial nerve palsies seen in this disorder ^[34][39]. Common associated symptoms include nausea/vomiting, neck pain/stiffness, tinnitus, dizziness, and hearing disturbances. Apart from that, there are a myriad of less frequently reported presentations which include other visual symptoms (photophobia, diplopia, blurred vision, nystagmus, visual loss), back pain, cognitive symptoms (including cognitive impairment, behavioral changes, memory, slow thinking), ear-related symptoms (vertigo, aural fullness, hyperacusis, or unspecified), reduced level of consciousness, and movement disorders (gait disorder, ataxia, dysarthria, tremor, bradykinesia, poor balance) [6]. Other rare, but peculiar symptoms, such as galactorrhea and hyperprolactinemia, have been attributed to the stretching or distortion of the pituitary stalk and/or hypothalamic region ^[36][37][41,42]; meanwhile, the dilatation of the epidural venous plexus or traction and compression of nerve roots are speculated to be the cause of radicular symptoms and incontinence ^[38][43]. Other less common, innumerable symptoms have also been described, including dysgeusia, sleepiness, other cranial nerve palsy, fever, fatigue, vocal tics, convulsions, facial spasms/numbness/pain, and dysphagia [6].

5. SIH: Diagnostic Workup and Imaging Strategy

As lumbar puncture is an invasive procedure, in addition to the fact that only about half of the demographic shows a low CSF opening pressure, imaging emerges as the key aspect in the management of SIH. Furthermore, CSF analysis often exhibits considerable variability and is nonspecific for SIH ^[13][18]. Imaging plays a pivotal role both in confirming the diagnosis of SIH and localization of the leak point, which allows for the subsequent targeted therapy. A head MRI is the mainstay of initial imaging of SIH, as it is the most sensitive tool in detecting the signs of a CSF leak [6]. Nevertheless, while useful for confirming the diagnosis of SIH, brain MRIs are usually unable to localize the exact site and morphology of the CSF leak. Often, they need to be followed by other myelographic imaging techniques which have the advantage of depicting the direct evidence of CSF leakage (Figure 1), rather than the indirect signs seen with a cranial MRI ^[15][23][20,28]. Several myelographic imaging techniques have been developed in response to the daunting task of identifying the site of the CSF leakage, which can be anywhere along the neuraxis and with various rates of leakage. These include a spinal MRI, computed tomography myelograms (CTM), dynamic CTM, digital subtraction myelography (DSM), MR myelography, and radionuclide cisternography. Choosing the correct myelographic imaging technique from an array of imaging modalities available today is fundamental for the successful treatment of SIH. A proposed imaging strategy starts with an MRI of the brain, with a contrast to look for signs of a CSF leakage ^{[15][33][39][40]}[20,38,44,45]. Patients showing a high possibility of SIH features should be subjected to a non-directed epidural blood patch (EBP). Those whose head MRI findings were equivocal would benefit from a spinal MRI, which is a non-invasive investigation. A spinal MRI is also indicated in patients who are refractory to non-directed EBP. The findings of a spinal longitudinal extradural collection (SLEC) within the spinal MRI would determine the best myelographic techniques for further delineation of the leaking site. Either DSM or CTM/dynamic CTM performed in certain recommended positions demonstrate a higher yield of CSF leak point detection ^[40][45]. If no leak is identified, further exploration with nuclear medicine myelography or MRI myelography (with intrathecal Gadolinium) may reveal an intermittent leak or very slow leaks.

6. Treatment and Prognosis of SIH

Large-scale data and randomized trials are still lacking to guide the standard management of SIH. Current practice of conservative treatment, such as bed rest, high oral fluid intake, caffeine, and an abdominal binder, are often the first-line therapy ^[41][47]. Subsequently, non-directed EBP is the treatment of choice if all conservative measures fail ^[13][18]. In non-directed EBP, patients’ autologous blood is injected into the epidural space, usually at the lumbar region. This will induce fibrin clot formation at the site of the dural tear, acting as a “patch” to stop the leakage. Symptomatic relief is sometimes immediate, with the majority of patients only requiring one or two sessions of EBP to achieve symptomatic relief ^[6][41][6,47]. For patients with recurrent symptoms after non-directed EBP, further investigations to identify the exact site and type of the CSF leak are necessary. A more targeted approach to the causative pathology can be offered, such as a directed EBP, microsurgical repair of the leak or endovascular treatment for CVF ^[42][43][44][62,63,64]. The latter is a relatively new treatment of CVF and can be performed by venous embolization of the suspected fistula site: a diagnostic and therapeutic work-up have recently been proposed, showing their feasibility and short-term effectiveness ^[45][65]. One complication of a successful CSF leak treatment is the development of rebound high intracranial pressure, which results in reverse orthostatic headaches; the high pressure is generally self-limiting.