Case
A 35 year old obese woman presents with headache, pulse synchronous tinnitus, and blurred vision. Visual acuity is 20/20 (6/6) OU. The pupil exam was normal OU. Automated perimetry showed a superior and inferior nerve fiber layer defect OU with a mean deviation of −27 decibels (dB) OD and −14 dB OS. Fundus exam showed Frisen grade 3 papilledema OU. Cranial magnetic resonance imaging (MRI) with and without gadolinium was negative for intracranial lesion but MR venography showed marked transverse venous sinus stenosis bilaterally. Lumbar puncture (LP) showed a markedly elevated opening pressure of 40 cm of water but normal cerebrospinal fluid contents.
Introduction
Idiopathic intracranial hypertension (IIH) is a disease predominantly of women of childbearing age with elevated body mass index in which there is elevated intracranial pressure (ICP) without tumor or meningitic inflammation as its cause. Although Quinke first described IIH in 1893 [1], it was Walter Dandy who in 1937 catalogued the condition in 22 patients ranging from age 9 to 48 [2]. In his seminal paper, Dandy explained that the cause was not known, but theorized that it could result from changes in cerebrospinal fluid (CSF) dynamics , and interestingly, due to “variations in the intracranial vascular bed probably by vasomotor control.” Based on Dandy’s description, J. Lawton Smith articulated 4 “modified Dandy criteria” in 1985, which included signs and symptoms of elevated ICP (headache, nausea, transient visual obscurations from papilledema), no other localizing signs other than abducens palsies, no abnormalities of the CSF other than elevated ICP, and normal or small ventricles.
Typical symptoms include position-dependent headache, a pulsatile whooshing, horizontal diplopia (from abducens palsies) and transient visual obscurations, blurry vision or visual field defects due to papilledema. Weight loss offers long term control of IIH, but faster control can be achieved with diuretics such as acetazolamide, which also reduces cerebrospinal fluid production, furosemide or topiramate. Acetazolamide has been validated in a large multicenter trial, which demonstrated slightly better visual field outcomes in patients with mild visual field loss, treated with weight loss and acetazolamide vs. those with only weight loss [3].
Unfortunately, up to 10% of IIH patients do not respond or do not tolerate medical therapy [4], and therefore require surgical treatments for fast and dependable control of ICP to prevent progressive vision loss and manage symptoms. Furthermore, up to 2.9% of patients present with fulminant vision loss, such that they cannot wait for a gradual reduction in ICP with medical treatment and must be treated from the start with surgery [5]. This can include a ventriculoperitoneal shunt, lumboperitoneal shunt, optic nerve sheath fenestration or bariatric surgery. Each procedure its advantages and specific risks, with shunting offering the best control of headaches, and fenestration the most effective protection from vision loss related to papilledema [6]. Since the late 1990s, it has been increasingly recognized that the majority of IIH patients have stenosis along the junction of the transverse sinus (TS) and sigmoid sinus (SS). This observation of venous sinus stenosis (VSS) has led to numerable studies evaluating treatment with venous sinus stenting. In this chapter, we will take a close look at the role of venous sinus stenosis in IIH, and our experience so far with venous stenting around the world.
Venous Stenosis and IIH
In 1995, King et al. demonstrated elevated venous pressures in the superior sagittal sinus and transverse sinus in patients with IIH [7]. Interestingly, this was not observed in minocycline induced IIH, suggesting that in “secondary IIH,” other factors were at work. As Farb et al. later showed in 2003, stenosis at the junction of the TS and SS is found in the majority of patients with IIH, suggesting that the stenosis causes the elevated venous pressure [8]. Using the auto-triggered elliptic-centric-ordered three-dimensional gadolinium-enhanced MR venography (ATECO MRV) technique, they found that a combined venous conduit score (CSS, accounting for the degree of stenosis bilaterally) of <5.0 predicted IIH with high sensitivity and specificity (both 93%).
Venous stent placement in the transverse sinus-sigmoid sinus junction demonstrated in the context of the greater cortical venous system
The Collapsible Model Theory Linking Extrinsic Stenosis and Elevation of ICP
As we shall see below, there now exists a considerable literature demonstrating successful treatment of IIH with stenting of extrinsic stenosis. Thus the question is raised: if extrinsic stenosis is secondary to elevated ICP and not the cause, then why would its eradication be therapeutic? The prevailing theory answering this question has been coined the “self-limiting venous collapse feedback loop model” [14]. Normally, CSF pressure must be higher than venous pressure in order to maintain drainage of CSF into the arachnoid granulations. Thus, the wall of the venous sinuses must remain rigid to avoid compression by higher CSF pressures that surround it. In certain individuals with compressible regions of the sinus, an inciting event causing a mild elevation in CSF pressure (weight gain or obstructive sleep apnea for example) might therefore result in stenosis of the susceptible region, in turn leading to localized venous hypertension. CSF drainage into the venous system is therefore compromised, resulting in elevation of ICP, which in turn compresses the susceptible sinus even more. The resulting positive feedback loop can ultimately result in the clinical entity of IIH. This system of a partially compressible sinus surrounded by a compartment whose pressure is related to the flow through that sinus has been mathematically modelled as a Starling-like resistor [15].
Self-limiting venous collapse feedback loop model. (a) A Starling Resistor may serve as a model for the relationship between blood flow through a collapsible segment of vein. The flow is proportionate to the difference between the proximal (transverse sinus) and distal (sigmoid sinus) segment pressures, also known as the “trans-stenotic gradient” multiplied by the fluidity. Fluidity, the reciprocal of resistance, is proportional to the diameter of the collapsible segment, which decreases when Pcsf > Pvenous, especially when the degree of collapsibility is high. (b) Decreased flow leads to higher venous pressure in the superior sagittal sinus and therefore elevated ICP, according to the formula: ICP = Rout * Formation Ratecsf + PressureSSS, where Rout is the resistance to outflow of CSF and PressureSSS is the pressure in the superior sagittal sinus. Increased ICP leads to greater venous stenosis and therefore an even greater trans-stenotic gradient, leading to a positive feedback loop. A modest increase in CSF pressure due to an inciting event such as weight gain or obstructive sleep apnea may lead to compression of a collapsible segment of venous sinus, leading to sinus stenosis, outflow obstruction and venous hypertension. CSF absorption is thereby decreased, leading to even further elevation of CSF pressure and so on. Note that in this model, venous sinus stenosis allows this vicious cycle but is not the inciting event. TS transverse stenosis, SS sigmoid sinus, SSS superior sagittal sinus, ICP intracranial pressure, CSF cerebrospinal fluid
The Role of Dural Incompetence and Venous Sinus Pulsatility
Lazzaro et al. described three patients with IIH and VSS in which direct manometry revealed an increase in the pulsatility amplitude, in addition to an increase in the mean venous pressure [17]. They postulated that this increase in pulsatility reflected a greater transmission of the CSF waveform through the CSF-venous interface (i.e. the dural wall). They further theorized that this resulted from acquired dural incompetence due to progressive thinning of the wall in the face of chronic intracranial hypertension. With this theoretical framework in mind, venous stenting treats the acquired dural incompetence and reduces the transmission of CSF waveform dynamics to the venous sinus. It also suggests that the measurement of pulsatility may aid in predicting which patients have dural incompetence and would therefore benefit from stenting.
Quantitative Assessment of VSS
MRV has the advantage of being an accurate, noninvasive test that avoids exposure to radiation and iodinated contrast medium. Typically, the assessment of dural vessel size, shape and volume is qualitatively assessed by the radiologist. Recently however, Lublinsky et al. developed a computer assisted detection (CAD) method for vessel cross section analysis to enable accurate evaluation of the shape and size of the dural sinuses in a quantitative manner [18]. In addition to confirming a generalized narrowing of the transvers sinuses in IIH patients, they also identified highly stenotic segments and found a significant difference between the minimal cross sectional area in IIH patients vs. controls. Using shape analysis, they also found that narrower vessels were more triangular, a finding whose significance needs to be investigated further. This group used CAD to create dural sinus normograms in healthy individuals to enable more accurate evaluation of sinuses in suspected IIH patients [19]. More recently, CAD was used to demonstrate a generalized stenosis of both transverse sinuses clustering near the junction with the sigmoid sinus [20], confirming previous reports of systemic venous sinus narrowing in IIH patients [21].
Venous Stenosis and Obesity
Although CAD did not demonstrate a correlation of VSS with BMI, a more recent study did demonstrate a correlation between BMI and mean intracranial venous pressure, as well as with the trans-stenotic gradient [22]. This finding offers more evidence in support of the theory that weight gain serves as an inciting event that raises the venous pressure to the point where the collapsible sinus feedback model takes over. The study also showed that patients with higher BMI were more likely to respond to VSS; obesity should therefore not be looked at as a contraindication to stenting.
Outcomes of Stenting
Map of the world with select locations featuring research regarding venous sinus stenosis and stenting in IIH
Case study of a 28 year old woman who presented with severe headaches, blurry vision, transient visual obscurations and pulsatile tinnitus. Exam revealed severe papilledema and Humphrey visual fields demonstrated enlarged blind spots and dense nasal depressions. MRV demonstrated bilateral venous sinus stenosis and lumbar puncture (LP) showed normal contents but elevated opening pressure to 77 cm of H2O. One month of increasing doses of acetazoalmide failed to affect her symptoms, the papilledema or visual field loss. A conventional venogram was performed and demonstrated a trans-stenotic gradient of 22 mmHg. A stent was placed across the stenosis and the gradient reduced to 1 mmHg. Within days, headaches, tinnitus and visual obscurations resolved. Repeat funduscopy within 12 weeks demonstrated resolution of papilledema with mild disc pallor in the left eye only. OCT showed near total resolution of thickening of the retinal nerve fiber layer (RNFL). Repeat LP showed opening pressure of 24 cm of H2O
Intracranial Pressure
The success of any treatment for IIH depends on its ability to reduce intracranial pressure quickly, dependably and durably. Out of 22 studies looking at venous stenting, 17 included pre-stent ICP data in the majority of the subjects, while 5 studies did not. In some cases, the last pre-stent ICP was measured prior to another intervention such as VPS, and could not be compared with post-stent ICP. Only 7 studies at the time of this writing included assessment of post-stent ICP in the majority of treated patients [25–31], possibly reflecting patient preference and concerns about performing LP while the patient is on antiplatelet therapy. In some studies, pre-stent mean ICP is reported for the majority of patients while post-stent ICP is reported for only a few patients, making a meaningful comparison challenging. Excluding ICP data from these studies, and including all other case reports and studies, there were 90 patients with both pre and post assessment of ICP. Using a weighted average reflective of the number of patients with pre and post ICP assessment in each study, mean estimates for pre and post-stent ICP for this population may be calculated: mean pre-stent ICP was 32.8 cm H2O, dropping to 17.6 cm H2O, yielding a 46.3% drop in ICP.
It should be noted that post-stent assessment was performed at a wide range of times following the procedure. Dinkin et al. reported post-stent ICP assessment at 3 months, but while a dependable drop in ICP was demonstrated, the speed of the ICP reduction was not clear [27]. However, Fargen et al. demonstrated an immediate reduction of ICP (using a parenchymal ICP monitor) from 70 to 20 cm H2O in one patient following stenting, and further reduction over the next 24 h [32]. The question was addressed more systemically by Liu et al. who utilized ICP monitoring in all ten patients and demonstrated an immediate reduction in ICP with stenting from 37.4 to 17 cm H2O, yielding an immediate mean reduction of ICP by 53.5% [30]. Matloob et al. placed ICP bolts in ten patients and assessed the mean ICP in the 24 h pre and post stenting and found a mean reduction from 20 to 12.2 cm H2O [31]. Of note, this group used an unconventional cut-off for elevated ICP of 6.1 cm H2O over a 24 h period, based on their own experience monitoring patients electively. The majority of their patients had a pre-stent ICP that most authors would consider borderline (8/10 patients had an ICP ≤ 21 cm H2O) and yet a significant reduction in mean ICP was still demonstrated. One patient did not show a reduction in ICP in the 24 h following stenting, suggesting that in a minority of patients, venous sinus stenosis may be purely a secondary phenomenon or that its effect on ICP in those patients may be delayed. This study also demonstrated a reduction in pulse amplitude and number of ICP spikes in the 24 h following stenting, suggesting a broader physiological response in CSF hemodynamics than just mean ICP.
Symptoms
The majority of studies assessing venous stenting for IIH have shown an improvement in patient symptoms following the procedure. Out of 464 patients, 424 presented with headaches. Following stenting, headaches completely resolved in 98 patients and improved in an additional 210. With 72.6% of patients presenting with headache experiencing either resolution of improvement, the remaining fifth either found no relief in headache, or only a transient relief. Given the higher rates of improvement in ICP (in those who underwent post-stent lumbar puncture) and papilledema, the persistence of some degree of headache in 73% of patients may reflect the multifactorial nature of headaches in this patient population, with migraines, tension headaches and medication-rebound headaches playing a role. In some cases of stenting, the incomplete reduction in ICP may have been sufficient to eradicate papilledema but not headache. We believe that the headache response rate after stenting is comparable with that of other therapies, including acetazolamide and shunting, and higher than optic nerve sheath fenestration [6].
Out of 102 patients presenting with transient visual obscurations (TVOs), 68 (66.7%) experienced complete resolution following stenting, reflecting an improvement in papilledema. The discrepancy between this outcome and the higher rate of papilledema resolution may reflect alternative contributors to visual obscurations including dry eye. Of 36 patients presenting with diplopia, 33 (91.7%) experienced resolution following stenting, likely reflecting resolution of ICP-related abducens nerve palsies. 126/417 (30.2%) patients were reported to have pulsatile tinnitus, with a resolution reported by 112 (88.9%).
It is likely that many specific symptoms were under-reported in the 464 patients reviewed here, as many studies simply referred to “visual symptoms” or even simply “symptoms.” The incidence of pulsatile tinnitus at presentation for example, was 29/37 (78.4%) in one study that specifically questioned patients about this symptom [33], more than double the rate reported overall. In that study, which assessed a tinnitus handicap inventory in each patient, there was a 97% resolution of PT immediately after stenting, supporting the concept that the venous sinus stenosis is the direct cause of this symptom (the one patient who did not experience resolution had minimal tinnitus to begin with). Of note, PT recurred in the three patients who experience recurrent tandem sinus stenosis.
Papilledema
Papilledema was present in at least 308/464 (66.4%) patients, with an additional 18 presenting with “either papilledema or high intracranial pressure” [34] and no discussion of optic nerve appearance at presentation in an additional 40 eyes [31, 35]. Out of the 308 patients with definite papilledema, 174 (60%) experienced complete resolution and an additional 83 (26.9%) experienced improvement in grade. Optic atrophy was noted in 27 patients (8.8%) in follow up. In total, these results suggest that the drop in ICP following stenting is sufficient to reverse optic disc swelling, with a relatively low rate of secondary optic atrophy.
Optical Coherence Tomography
By analyzing the interference pattern of low coherent light as it reflects off the posterior pole, optical coherence tomography (OCT) is able to construct three-dimensional images of the retina and optic nerve and to measure the thicknesses of the various retinal layers, including the retinal nerve fiber layer (RNFL) and ganglion cell layer (GCL) which contains the neurons of the optic nerve. Thinning of the RNFL indicates optic atrophy while thickening is found in cases of active papilledema. Thus, using OCT, papilledema in IIH can be quantified and followed over time. A total of four studies included OCT in the assessment of papilledema pre and post-stenting [27, 30, 36, 37] for a total of 73 eyes. Combining their results, the mean pre-stent RNFL thickness was 225.4 μm and the post thickness was 87.6 μm, with a mean reduction in RNFL thickness of 137 μm. Out of the 73 eyes, 36 improved to <110 μm. These OCT results support the fundus observations indicating resolution or improvement in papilledema in patients who undergo stenting for IIH. However, RNFL thicknesses always need to be interpreted with caution, since pseudo-normalization may results from a mixture of residual RNFL thickening (papilledema) and atrophy. Future studies would benefit from analyzing the GCL as well, since atrophy of the GCL would not be obscured by residual papilledema, and could therefore serve as a quantitative marker of optic nerve damage from papilledema.
Visual Acuity
No treatment option for IIH is of high value if it does not prevent vision loss from papilledema. A therapy which reduces ICP too slowly, too mildly, or too fleetingly, might not prevent the secondary atrophy that results from prolonged papilledema and the permanent visual field and acuity loss that may accompany it. As such, we turn now to the available data regarding outcomes in visual assessment in patients undergoing stenting. Unfortunately, objective assessments of visual parameters have inconsistently been included in stent studies, with many simply reporting an improvement in vision or visual symptoms.
Including case reports, initial visual acuity (VA) loss was reported in 177/833 eyes (21.2%), out of which 105 (59.3%) improved. A total of eight studies included quantitative data on VA pre and post stenting, totaling to 160 eyes [24, 25, 27, 29, 30, 38–40]. In one study, only an average of VA in the right and left eyes was reported, in which case that value is estimated for the purposes of this review as the acuity in each eye [29]. Looking at all the eyes from patients who underwent stenting in which quantitative data was available, mean VA prior to stenting was LogMAR 0.25 (20/36) and improved to 0.136 (Snellen 20/27). Since papilledema only rarely causes severe VA loss (even when there is visual field loss), assessment of change in mean VA is less useful than visual field assessment.
Visual Fields
Qualitative visual field data was available in six studies [24, 27, 29, 30, 40, 41]. A combined analysis reveals that VF defects were reported in 172 eyes (assuming visual field loss in both eyes in studies that reported visual field loss by patient [29]) out of which there was improvement in 122, no change in 38 and a worsening in 12. Quantitative analysis of mean deviation (MD) was performed in three of the studies [27, 29, 40] comprising 71 eyes. The combined results showed an improvement in average MD by 3.29 dB (from −10.35 to −7.05 dB). Hopefully further studies comparing venous stenting to other treatment modalities in larger populations will be able to show a similar effect on visual fields.
Stenting for Fulminant Vision Loss in IIH
In most studies evaluating venous stenting for IIH, indications included a lack of response to ICP-lowering medications such as acetazolamide or medication intolerance. However, more established surgical options such as optic nerve sheath fenestration and CSF shunting are also utilized prior to medication trials, in patient with IIH with severe vision loss due to papilledema at presentation. In such cases, there is no time to wait for the relatively slow effect of medications, which might not lower the ICP in time to prevent permanent vision loss. Elder et al. showed significant recovery in visual fields in 2/4 patients stented for fulminant vision loss from IIH [28]. However, one patient who presented already with some degree of optic atrophy descended from 4/200 right eye and 2/200 left eye to blindness in both eyes. Four of the patients in Ahmed et al’ study presented with severe vision loss, but a precedent ONSF in all 4 prevents a complete understanding of the stent’s effect on visual improvement. Three patients (6 eyes) presented with severe visual field loss in one study and were stented prior to any medication trial [27]. All three experienced improvement in visual fields.
Side Effects and Complications
Similar to alternative surgical treatments for IIH, venous stenting may result in side effects and complications, some of which have the potential to be serious. By the far the most frequent adverse symptom is headache ipsilateral to the stent, which was reported in 104/464 (22.4%) patients in the literature, likely due to dural stretch, and typically lasting for less than a week. The true incidence is likely higher since such pain was likely lumped in with post-stent headaches presumed to be related to IIH in some studies. In Dinkin et al., for example, >50% complained of this type of headache, which differed from their pre-stent headaches as it was typically unilateral and non-positional [27].
In the first demonstration of venous stenting for IIH, Higgins et al. [23] reported two cases of in-stent thrombosis, but attributed this complication to the fact that they did not start antiplatelet therapy until after stenting in their first few patients. Intracranial hemorrhages have been reported in four cases (0.96%). Venous stasis and a contralateral venous sinus thrombosis led to a combined subdural hematoma (SDH) and subarachnoid hemorrhage (SAH), complicated by mild foot weakness in one case, but this was in the setting of a parasagittal arteriovenous malformation, as opposed to truly idiopathic intracranial hypertension [42]. Of the two patients who suffered SDH in the study by Ahmed et al. (one followed a guidewire perforation), both made full recoveries [24].
Other serious complications occurring at the time of stenting included femoral pseudoaneurysm (1.2%), transient hearing loss, presumably from cochlear vein obstruction (0.96%), allergic reaction to contrast (0.72%), retroperitoneal hematoma (0.48%), neck hematoma (0.48%), contrast extravasation (0.24%), stent migration (0.24%), and ruptured ovarian cyst (0.24%). Later complications were mostly a consequence of anti-platelet therapy, with one patient experiencing each of the following: menorrhagia, epistaxis and melena.
We found two reports of death related to venous stenting. In the first case, uncontrolled cerebral edema at the time of stenting was attributed to hypercapnia in the setting of hypoventilation while under generalized anesthesia [43]. However, given the rarity of anesthesia-associated death, it is conceivable that stent-associated venous stasis (possibly by blocking the output of a cortical vein), contributed to the cerebral edema in this case. Indeed, blockage of the ostium of the vein of Labbe was found in 92.1% of patients in one study, although the authors also found no immediate effect on venous outflow [44]. Raper et al. found similar results in 56 patients, noting no neurological sequelae [45]. The practice at Cornell (in two similar cases) has been to treat with several days of heparin to avoid secondary thrombosis. The second case of stent-associated mortality resulted from a severe cerebellar hemorrhage, possibly from guidewire perforation [46].
Satti et al. reported that the rate of major complications with venous stenting (2.9%) was higher than following ONSF (1.5%) but lower than in patients who are status post CSF diversion (7.6%) [6]. Complication rates for all of these procedures are likely much higher however, due to reporting bias, as major complications occurring in patients outside of studies may not be reported.
Recurrent Stenosis and Contributing Factors
Recurrent stenosis, sometimes accompanied by a clinical relapse, was reported in a total of 35 patients (5.52%), 8 of whom developed recurrent in-stent stenosis and 27 of whom developed stent adjacent stenosis (SAS). In most of the cases of SAS, a new stent was placed within the narrowed region and was accompanied by clinical improvement [24, 27, 47]. Raper et al. analyzed the pattern of resolution of the trans-stenotic gradient to see if it was related to the risk of SAS [48]. They described three patterns of resolution—in the first (38.3%), the mean venous pressure (MVP) , distal to the stenosis (superior sagittal, torcula and transverse sinus) was reduced, in the second type (14.9%), the proximal MVP increased, and in the third (46.8%), both occurred. The authors found that the incidence of SAS (which was 14.9% in total) was significantly higher in patients with patterns 2 and 3 as compared to pattern 1, and they theorized that those in group 2 and 3 were “non-responders” since the distal MVP did not decrease, and were therefore more prone to SAS. Kumpe et al. found that female gender and a purely extrinsic compression of the TS-SS junction were risk factors for SAS, the latter relationship presumably resulting from the fact that extrinsic stenosis is secondary to elevated ICP and is thus prone to occur again in a separate location, (even as it also feeds back causing even greater ICP elevation), while intrinsic stenosis is a primary anatomical problem which presumably causes the high ICP in the first place [49]. Goodwin et al. compared three patients who failed stenting therapy and required shunting with 15 who were successfully treated with stent alone, and found a higher mean opening pressure pre-stenting in the failure group (50 cm H2O) vs. the stent-only group (37 cm H2O) [50]. They suggest that severely elevated ICP signifies a form of IIH in which stenting alone might not be sufficient to control the disease. This conclusion needs to be explored further in larger studies.
Stenting for Venous Stenosis in the Absence of IIH
Venous stenosis may occur in the absence of classic IIH in the setting of prior venous sinus thrombosis, hydrocephalus, meningitis or compression by tumor, some of whom suffer from constant pulsatile tinnitus. Levitt et al. published a series of nine such patients, six of whom had confirmed elevated ICP, all of whom experienced resolution of symptoms except for those with congenital hydrocephalus [51]. Our group has also treated patients with isolated pulsatile tinnitus, some of whom have not had any other symptoms or signs of IIH, with complete resolution after stenting. The role of stenting in these cases needs to be investigated in larger series.
Global Perspective
The apparent efficacy and safety of venous stenting as a treatment for IIH is, to date, based on retrospective studies and two uncontrolled and un-blinded prospective studies. As such, reported results must be looked at in that context, taking into account the placebo effect on both patients and physicians. It is also quite likely that some complications have not been published in the literature, as we have been made aware (personal communication). Several groups are therefore preparing prospective head to head trials comparing stenting vs. shunting in IIH patients [52]. It is our hope that such studies will help confirm the utility of stenting for medically-refractory patients with IIH and allow the proper identification of appropriate patients for this emerging therapy.
Comments by Andrew G. Lee, MD
Although there has been increasing evidence and enthusiasm for venous sinus stenting in IIH, I agree that a prospective, head to head controlled clinical trial will be necessary to answer some of these nagging questions. In the interim, it seems reasonable to consider venous sinus stenting in the armamentarium for IIH. I personally consider venous sinus stenting in IIH patients who have failed or are intolerant of maximum medical therapy (e.g., progressive visual loss or intractable headache), have a documented venous sinus pressure gradient on manometry, and who have failed, declined, or are not a candidate for optic nerve sheath fenestration or CSF diversion procedure.