Article Text
PDF
Statistics from Altmetric.com
Introduction
In recent years, venous sinus stenting (VSS) has emerged as an effective surgical treatment for idiopathic intracranial hypertension (IIH) with concomitant venous sinus stenosis. Meta-analyses of small series of VSS have demonstrated an excellent safety profile with improvement in headaches, pulsatile tinnitus, papilledema, and visual symptoms in the majority of patients.1 This has led to a rapid growth of VSS being performed with few practitioners having substantial experience. This fact is concerning for several reasons. First, little is known about the pathophysiology of the stenosis or why it might recur following stenting, other than untested theories.2 3 Second, more realistic series suggest that as many as 60% of patients with IIH have symptoms that persist or recur after VSS.4 Third, major complications, including death, occur in about 2% of treated patients.5 These complications are largely avoidable and can potentially be minimized by thoughtful patient selection and safe procedural techniques. Yet recommendations for the selection of patients and performance of VSS are quite limited,6 with little published advice or educational resources available for surgeons interested in performing these procedures.
The purpose of this manuscript is to present lessons learned from a high-volume VSS practice to aid neurointerventionalists in caring for patients with IIH. My personal experience includes over 150 VSS stent procedures and over 400 cerebral venogram procedures over a 5-year period (roughly 30 and 80 procedures annually, respectively). Admittedly, there is little scientific evidence supporting most of the principles presented, but for those learning points where some data exist, references are included. In many instances, alternative strategies not discussed in this paper may be equally (or more) effective; there are many ways to take good care of patients. As such, the views expressed in this commentary are strictly my own and should not be considered ‘evidence-based,’ but instead represent ‘pearls’ from an experienced practitioner to aid physicians with limited experience with this procedure.
Patient selection
IIH is a spectrum
First and foremost, IIH should be thought of as a spectrum of intracranial pressures. The symptoms experienced vary markedly among patients, with certain patients having higher sensitivity to pressures than others. About 15–20% of my patients have opening pressures (OPs) on lumbar puncture (LP) that do not meet the classic criteria for IIH (OP of ≤24 cm of water), but clearly demonstrate temporary symptomatic improvement following cerebrospinal fluid (CSF) removal.2
Patients with connective tissue disorders (most notably Ehlers-Danlos syndrome) have a classic phenotype that includes IIH. Anecdotally most of these patients are Caucasian, young, light-haired, and non-obese. Many are hyperflexible. These patients often have severe symptoms with OP in the 15–24 cm of water range and frequently have CSF rhinorrhea, internal jugular vein (IJV) stenosis, craniocervical instability, postural orthostatic tachycardia syndrome, and mast cell activation syndrome with numerous allergies. They are often challenging to treat. They are also more likely to develop CSF leaks following LP, so using a small gauge (22 or higher) atraumatic needle is advised. Particular attention should be placed on potential IJV stenosis near C1 in these patients.
Intracranial pressure and stenosis
The higher the LP opening pressure, the more likely it is that a stentable stenosis is present. For instance, roughly 25% of patients with an OP of ≤24 cm of water have a pathologic stenosis, whereas about 75% of patients with an OP of ≥35 have a pathological stenosis and are candidates for VSS.7
Non-invasive imaging
Many doctors continue to use non-invasive imaging (MR venography or CT venography) for patient screening prior to venography. I have found this imaging to have low negative predictive value8 and therefore do not routinely order these studies. Radiologists often focus on the medial transverse sinus when evaluating for stenosis and may miss severe stenosis near the transverse-sigmoid junction. Additionally the IJVs are often ignored and clear-cut pathologic stenosis may be missed if attention is not placed on the entirety of the visualized venous outflow pathways (figure 1). Should one decide to screen with non-invasive imaging, be sure to evaluate the caliber of the superior sagittal sinus all the way down to the lowest visualized portion of the IJV.
Patient referred for symptoms of idiopathic intracranial hypertension with normal MR venography radiology report. Note the intracranial venous sinuses appear normal caliber without stenosis, but there is severe stenosis of both internal jugular veins (IJVs) near the transverse process of C1 (green arrow) with pathologically dilated suboccipital venous plexi (yellow arrows) indicating severe, pathologic venous congestion (A). Anteroposterior right IJV venogram (B) confirms severe stenosis (green) with a 6 mm Hg pressure gradient and pathological collateralization (yellow).
Symptoms
One interesting question to assess whether symptoms are related to IIH is to ask if the symptoms worsen with weather changes (specifically barometric pressure changes). Almost uniformly, patients with IIH report feeling worse with changes in weather patterns. One reliable way to assess whether symptoms are pressure related is to assess for temporary symptomatic relief following CSF removal with LP. Failure of symptoms to at least temporarily improve after LP should lead you to strongly question the diagnosis. Do, however, keep in mind that many patients with elevated OP will leak CSF following LP and may develop intracranial hypotension (sometimes requiring blood patching), which can obscure the subjective assessment of symptom improvement.
Most patients with IIH experience tinnitus. High-pitched ringing is common as a global intracranial hypertension phenomenon, but pulsatile tinnitus that becomes louder when the headache is worse is a reliable indicator of contributing pathologic venous stenosis. This observation is consistent with studies strongly linking venous stenosis or other anomalies to pulsatile tinnitus.9 10 Additionally, many authors now think that most patients with spontaneous CSF otorrhea or rhinorrhea have underlying or contributing IIH.11
VSS is highly effective in treating papilledema and visual loss1 and has even shown promise in the setting of acute or fulminant visual loss.12 Headache is the most prominent and burdensome symptoms to patients, however, and is particularly prone to persistence or recurrence even after successful stenting.4 Patients with the predominant symptom of headache should be appropriately counseled about the high likelihood that some degree of headache will persist even after stenting.
Venogram
Planning
The most widely accepted theory explaining the pathophysiology of VSS is that of a positive feedback loop linking elevated intracranial pressure to extramural sinus compression and resultant venous congestion. Studies have demonstrated that lowering intracranial pressure can temporarily eliminate the visualized stenosis and its associated gradient.13–15 Therefore any intervention that lowers intracranial pressure will potentially invalidate the diagnostic venogram procedure. I do not perform venograms within 2 weeks of a prior LP to eliminate the potential confounding from ongoing spinal CSF leak. Furthermore, I ask patients to stop taking acetazolamide, topiramate, or diuretics 48–72 hours before the venogram so as to increase the diagnostic yield by ensuring patient physiology is in an unadulterated state and not confounded by medication that lowers intracranial pressure.
Sedation
Diagnostic venography should be performed with the patient under minimal or conscious sedation.6 Two studies have clearly shown that general anesthesia invalidates (and usually underestimates) trans-stenosis gradients.16 17 I believe that performing a diagnostic procedure first, followed by stenting under anesthesia at a later time, provides the best way to obtain the necessary information and then counsel the patient and their family about the risks and benefits of treatment, except in pediatric patients where a combined diagnostic procedure under conscious sedation, followed immediately by VSS under general anesthesia, seems to be the best strategy.18
Access
There has been a recent push to transition to radial access for neurointerventional procedures. In my practice I have not found this conversion to be necessary for patients with IIH and continue to use right femoral access with placement of 5F arterial and venous sheaths. Access in this manner is straightforward, allows for rapid performance of the procedure from a single access site which is important for patient comfort due to the lack of sedation, and I have yet to identify an iatrogenic arteriovenous fistula when using this strategy. Additionally, femoral access into the right and left IJV is quite straightforward even with troublesome venous valves, and a full evaluation can be performed successfully in most cases. However, femoral arterial access in this predominantly overweight population, often taking dual antiplatelet agents, must be weighed against the known increased risk of access site complications compared with upper extremity access.19 Some practitioners have advocated direct IJV puncture, for either venography or for stenting. Although no complications have yet been reported from directly accessing this site, I would strongly advise against this approach due to potential injury to the IJV (which is known to occur with direct access),20 especially given that the IJV is an essential part of the venous outflow pathway of the head.
Intracranial catheterization
Navigation of devices within the jugular bulb and sigmoid sinus (SS) is painful for patients. Sometimes, navigation of a microcatheter into the SS can be challenging. Using a standard 5F primary curve diagnostic catheter as a guide catheter is helpful in obtaining sinus access, and given that venous access is retrograde and there is no need to inject around the microcatheter, a larger guiding catheter has never been necessary in my experience. Start by directing the 5F catheter medially (figure 2A) and then advancing the microwire; if this fails, rotate the diagnostic catheter and try from alternate positions. Navigation into the SS, transverse sinus (TS), and then the superior sagittal sinus (SSS) is aided by a primary microwire tip curve followed by a gentle secondary bend (figure 2B).
Catheterization of the sigmoid sinus is aided by aiming the catheter medially (A) and with gentle microwire primary and secondary curves (B). During venous sinus stenting, stenting is aided by placing a guide catheter in the jugular vein (blue arrow) and advancing an intermediate catheter (yellow arrow) over a 5F insert catheter and glidewire that is advanced into the superior sagittal sinus (SSS). The glidewire is aimed inferiorly through the lateral TS to avoid the vein of Labbé and then aimed superiorly in the medial TS to access the SSS (C).
Microcatheter
One published study compared pressure transduction through multiple microcatheters and suggested that certain small-bore microcatheters are probably inadequate for use.21 In my experience 0.027 inch microcatheters are easily navigable, allow for adequate venous injections, and capture waveform and pressures reliably.
Venous manometry
Always zero the pressure transducer in the mid-axillary line to ensure accurate and reliable pressure measurements. Given the preponderance of concomitant or isolated SSS and IJV stenosis,3 22 pressures should be recorded from the mid-SSS through central venous pressure using distinct landmarks: S1–2 junction, torcula (confluens), TS, transverse-sigmoid junction, SS, IJV above C1, IJV below C1 (at C3–4), and central venous pressure. Usually unilateral, dominant pathway pressure measurements are sufficient, except in select cases where co-dominance is present or where venous collateralization suggests pathology on the non-dominant side.
Paying attention to the venous waveform is also helpful in confirming the accuracy of the pressures being recorded. High-amplitude waveforms with limited respiratory variability are usually present when upstream of a significant hemodynamic stenosis, while low-amplitude waveforms with pronounced respiratory variability are often present in the absence of stenosis or when caudal to a stenosis.23
Jugular vein stenosis
Published data on selecting patients for IJV stenting or on IJV stenting outcomes are limited. Measured gradients across severe IJV stenoses are usually much smaller than those seen in the TS (usually 3–6 mm Hg). After catheterization of the IJV, provocative injection into the rostral jugular vein can be helpful in replicating or worsening symptoms to confirm that venous hypertension is contributing to the underlying condition. With the catheter above C1, perform a strong injection and ask the patient if their symptoms acutely worsened afterwards. Note pathologic suboccipital venous plexus collaterals (figure 1B). Rotational angiography also can be helpful, as the IJV frequently further stenoses with head turning.
If considering IJV stenting, it is necessary to determine the anatomical relationship between the styloid process, transverse process of C1, and IJV on non-invasive imaging (usually CT of neck with contrast). IJV stenosis as a result of a single bony structure (C1 transverse process, for instance) is often amenable to stenting. Bony constraint of the IJV from both C1 and the styloid process will not improve with stenting and styloidectomy can be considered as a potential treatment option.
Venous sinus stenting
Patient selection
The commonly used 8 mm Hg pressure gradient for selection of patients for VSS has not been rigorously studied. Some studies have suggested that the higher the gradient, the more pronounced the benefit to stenting.24 This intuitively makes sense given the usual 1:1 relationship between venous pressures and CSF pressures,25 suggesting that the higher the gradient, the lower venous pressures (and OP) will be following successful resolution of the gradient. However, many practitioners, including this author, occasionally stent for gradients between 4 and 8 mm Hg in select cases.6 Appropriately counseling of the patient about the lower chance of a good outcome seems warranted when stenting for smaller gradients.
Anesthesia
Changes in blood pressure and end-tidal carbon dioxide are known to influence venous pressure measurements.26 27 Asking the anesthesia team to maintain end-tidal carbon dioxide in the near-normal range (38–40 mm Hg) with stable blood pressures will make pre- and post-stenting manometry more accurate.
Internal carotid artery injection
I believe it is helpful to perform ipsilateral internal carotid artery injection before and after stenting to evaluate venous outflow, in-stent thrombus, and identify potential complications, such as venous injury and extravasation.5 This is probably especially helpful for neurointerventionalists inexperienced with VSS but arguably less so for those more familiar with the procedure. The risk of internal carotid artery injection in this predominantly young patient population, especially those taking antiplatelet agents and heparinized, is low and in most cases can be done safely. Slow flow in the ipsilateral vein of Labbé can be seen on injection of the internal carotid artery after stenting without complication,28 and in my experience when this does occur it has always been clinically silent.
Pre- and post-stenting manometry
Determining baseline pressures under anesthesia and then repeating manometry after stenting is highly recommended to ensure adequate treatment of the gradient and identify de novo gradients that may occur.3
Guide catheters
Accessing the TS for stenting often requires substantial force, which can be daunting and seem excessive to physicians inexperienced with intracranial venous access. Often there is significant resistance to advancing stents through the IJV into the SS due to an acute turn. There are a number of constructs and catheters that can be used to successfully implant stents; different stents (biliary vs carotid) may require different systems. With that being said, I have found access into the TS for stenting is most easily performed using a triaxial system without intervening heparinized flush valves (0.088–0.091 90 cm guide catheter, 105 cm 0.070 inch navigable intermediate catheter, and then 5F angled insert catheter) and 0.038 inch glidewire (figures 2C and 3). The larger guide catheter is placed in the IJV near the bulb and the intermediate catheter is then navigated over the wire and insert catheter into the ipsilateral TS. The insert and wire are then removed and a stent is advanced over a microwire. Having the intermediate catheter in the TS usually allows easy navigation of the stents intracranially. It is easy to access the SS with the 0.038 inch glidewire through a 5F insert (usually easier than with a microcatheter) and often this wire can easily be advanced through the TS and into the SSS for support when climbing with the intermediate catheter. The wire should be turned downward as it travels along the TS (to avoid inadvertent Labbé puncture) and then turned upward near the medial TS to access the SSS (figure 2C). The intermediate catheter will occasionally get hung up on the sinus wall; advancing the glidewire and 5F insert beyond the point of resistance usually allows the intermediate catheter to overcome this resistance and safely advance. Delivery of stents into the SSS can be difficult unless the intermediate catheter is positioned in the medial TS. Be mindful that use of considerable force can result in prolapse of the proximal system into the right atrium, which can be detected due to resultant ectopy or by checking fluoroscopy over the chest if the catheter is not advancing appropriately with additional load.
Stenting of the superior sagittal sinus (SSS) and TS for multifocal stenosis with gradients (A). Stents can be easily navigated by placing a 0.088–0.091 inch guide catheter in the jugular vein near the bulb (blue arrow) and a 0.070 inch intermediate catheter in the TS (yellow arrow). Place distal stents first. The torcula (green triangle) seems resistant to extramural compression and usually stents do not need to cross this site. Undersizing of stents can cause untoward migration (B). in this image an undersized right TS stent (yellow dots) was pushed forward into the torcula accidentally when placing a second proximal TS stent. This could not be corrected even with pushing from contralateral access (shown) but was tolerated without complication, although future access to the SSS is in jeopardy should further stenting in the SSS be necessary in the future.
Stents
Little published guidance on stent constructs is available.3 22 Because of concern about stent-adjacent stenosis if only a segment of the TS is stented, I always perform stenting from the torcula to the TS–SS junction, which usually requires 70 mm of length. If there is SSS stenosis present with even a small gradient (≥3 mm Hg), I will often place an additional stent in the caudal SSS at this location to prevent worsening stenosis that may occur at this site in the future, although there are no published data to support this practice. The torcula appears to be relatively resistant to extramural compression,3 and stents usually do not need to cross this region (figure 3A). If using multiple stents in sequence, place the most distal stents first to avoid inadvertently pushing the previously implanted stent out of position while traversing it. Avoid stent oversizing, which may increase the chance of adjacent stenosis29; usually 6–8 mm stents are appropriate for the TS and 5 mm in the SSS. Undersizing can also be problematic and result in inadvertent stent migration (figure 3B).
Post-stent angioplasty
A constrained stent without an associated persistent gradient usually does not require angioplasty. If a post-stenting gradient exists at the site of residual stenosis, submaximal balloon angioplasty can be safely performed and will usually result in resolution of the gradient (figure 4).
Severe stenosis (red arrows) of the TS is present on pre-stent venography with a 17 mm Hg gradient (A). after stent placement (ends marked with yellow arrows), residual stenosis is present (red arrow) with a 9 mm Hg gradient (B). Angioplasty was performed (C), resulting in elimination of the gradient (D).
Antiplatelet agents
Peri- and post-operative dual antiplatelet agents are recommended as there are reported cases of stent thrombosis with aspirin alone.6 Duration is highly variable among surgeons, but to my knowledge there are no reported or anecdotal cases of complications arising from stopping dual antiplatelets after 4–6 weeks. Heparinization for VSS with confirmation of an appropriate activated clotting time is also recommended due to the rare risk of intraprocedural thrombosis.5
Post-stent management
Hospital care
The overwhelming majority of patients do not require intensive care unit admission and spend a single night in hospital for control of pain and nausea. Stent pain (stabbing pain behind the ipsilateral eye and behind the ear) is common after VSS, is worst in the first 8 hours, and usually resolves over the next 3–5 days in the majority of patients.
Symptom persistence or recurrence
Failure to improve symptoms, or delayed symptom recurrence, are fairly common even after successful VSS and in the presence of lower intracranial pressures or with resolution of papilledema.4 Non-invasive imaging performed for subacute worsening headache days to weeks after VSS is almost always unremarkable except for occasional new stenosis that can be visualized; delayed stent thrombosis is extraordinarily rare.5
Often headaches and even ipsilateral tinnitus recur roughly 3–6 months after stenting.4 Lumbar puncture can be helpful in guiding management of patients by showing the degree of pressure reduction following stenting and by confirming that ongoing symptoms are, in fact, IIH-related if they improve with CSF removal. Those patients with dramatic reductions in OP on LP after VSS rarely have ongoing venous stenosis, while those with OP approximating or slightly lower than pre-stent OP are occasionally discovered to have new stenoses that may be amenable to VSS.
New stenosis and repeat stenting
New stent-adjacent or remote pressure gradients can occur upstream of the stent, either in the TS or SSS.3 30 31 In my experience, the highest yield stenting procedure is the first. Additional procedures can be performed safely but have diminishing effect on improving symptoms.
Conclusion
Few published recommendations for the selection and treatment of patients with IIH by VSS are available to assist neurointerventionalists in the care of these patients. This manuscript highlights important ‘pearls’ learned from a high-volume VSS practice, in the absence of robust scientific evidence, which might help neurointerventionalists in the selection, management, and treatment of patients with IIH and associated venous stenosis.
Supplemental material
Data availability statement
No data are available. Not applicable.
Ethics statements
Patient consent for publication
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Contributors The author is the sole contributor of the manuscript.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.