Introduction Venous outflow obstruction is recognized as a contributing factor in a subset of patients with idiopathic intracranial hypertension (IIH). Little is known about venous sinus waveform morphology or how it changes after stenting.
Methods Fifteen patients with IIH underwent waveform recording during catheter venography and manometry. Ten patients (Group A) with venous sinus stenosis and pressure gradient ≥7 mm Hg underwent waveform recording during awake venography and during stenting under general anesthesia. Five control IIH patients (Group B) without a gradient underwent awake recording only.
Results Group A patients underwent successful stenting with reduction of their gradient from 15.1±6.19 mm Hg to 1.2±0.60 mm Hg. This resulted in an amplitude reduction from 8.3 mm Hg to 2.8 mm Hg (P=0.02). Qualitative evaluation of the waveform yielded a number of novel findings. In Group A before stenting, the observed waveform progressed from an intracranial pressure (ICP)-dominated to central venous pressure (CVP)-dominated waveform. Stenting abolished the high amplitude waveform and smoothed the transition from the intracranial to central venous measurement points. Group B displayed primarily CVP-influenced waveforms distal and proximal to the transverse-sigmoid junction along with respiratory variability of the waveform, absent in 8/10 Group A patients. General anesthesia appeared to blunt the waveform in 5/10 Group A patients.
Conclusion The cerebral venous waveform appears to be influenced by both the ICP and CVP waveforms. As measurement moves proximally, the waveform progressively changes to mirror the CVP waveform. Venous sinus stenosis results in a high amplitude waveform which improves with treatment of the stenosis.
- blood flow
Statistics from Altmetric.com
Idiopathic intracranial hypertension (IIH) is a poorly understood disease characterized by chronic headaches, tinnitus, and ophthalmologic manifestations including diplopia and papilledema leading to vision loss. IIH was first described by Walter Dandy and, in perhaps a rather telling fashion, he did not include any diagnostic criteria for the disease in his initial report.1
The pathophysiology of IIH remains poorly understood, however many experts postulate that impaired CSF resorption underlies the condition.2 Recently, clinicians have found an increasing number of patients in whom cerebral venous sinus stenosis with resultant venous sinus hypertension is implicated as a causative mechanism. In these patients, cerebral venous manometry during venography often identifies physiologic venous outflow obstruction at the transverse–sigmoid sinus junction which manifests as a pressure gradient across the stenotic segment of vein. Relief of the venous bottleneck by implanting stents at the site of stenosis is associated with good clinical results.3 4
Most surgeons assess candidacy for stenting based on the magnitude of the venous pressure gradient across the stenosis. Little is currently known about the venous sinus waveform, its component peaks, or the influence of stenosis or anesthesia on its morphology. We report the results of a prospective study of IIH in patients in which venous manometry waveforms were recorded and analyzed to provide insight into the waveform and pathological morphologic changes seen with sinus stenosis.
A prospective, single-center observational study of patients with IIH was performed following institutional review board approval. All patients with IIH undergoing awake diagnostic angiography and venography to determine candidacy for venous sinus stenting were prospectively enrolled in this observational pilot study. Patients who were evaluated with catheter angiography and venography and were found to have a trans-stenosis gradient of ≥7 mmHg were assigned to Group A, and subsequently underwent venous sinus stenting under general anesthesia. Group A patients underwent venographic waveform recording during the awake diagnostic procedure as well as before and after stenting under general anesthesia. Those without a trans-stenosis gradient underwent waveform recording during the diagnostic procedure only and did not undergo stenting (Group B). All candidates for inclusion were diagnosed with IIH based on clinical examination in accordance with published diagnostic criteria.5 The relatively small sample size is due to the pilot nature of this study, designed to assess the changes in the cerebral venous waveform associated with treatment of venous sinus stenosis.
Diagnostic cerebral arteriography, venography and venous manometry
Diagnostic cerebral arteriography, venography, and venous manometry were performed in all patients under mild conscious sedation using fentanyl only. In all patients, the femoral artery and vein were both accessed and 5 F sheaths were placed in each vessel. A 5F diagnostic catheter was used to perform a cerebral arteriogram to evaluate venous outflow pathways and rule out arteriovenous fistulas. Next, the 5F diagnostic catheter was advanced into the dominant internal jugular vein (IJ). A Rebar-27 microcatheter (Medtronic, Minneapolis, Minnesota, USA) was navigated over a 0.014 inch microwire into the superior sagittal sinus (SSS). A diagnostic cerebral venogram was then performed followed by serial venous manometry measurements in the SSS, torcula, transverse sinus (TS), transverse–sigmoid junction (TSJ), sigmoid sinus (SS), IJ, and central venous pressure (CVP) in the right atrium. When possible, bilateral transverse–sigmoid pathways were accessed and manometry and waveform recording were performed on both sides. All patients (Groups A and B) underwent waveform recording during this procedure.
Venous sinus stenting
Venous sinus stenting was offered to patients with a clinical diagnosis of IIH as well as a demonstrated pressure gradient ≥7 mm Hg across a venous sinus stenosis. Prior to stent placement, all patients were started on 325 mg aspirin and 75 mg clopidrogrel daily for 1 week prior to the procedure.
Following anesthetic induction and endotracheal intubation, femoral venous access was obtained. An 8 F sheath was placed in the femoral vein. Weight-based intravenous heparin was then administered. A 5 F Berenstein catheter was placed into a 0.070 inch Neuron catheter inside a 0.088 inch Neuron Max guiding catheter (Penumbra, Alameda, California, USA) and navigated over a 0.038 glidewire into the IJ ipsilateral to the stenosis. Following this, pre-stenting diagnostic cerebral venography and manometry were performed from the SSS to the ipsilateral IJ in the previously described fashion using the Rebar-27 microcatheter. Then, venous stenting was performed using an appropriately-sized Precise stent (Cordis, Milpitas, California, USA). Depending on the length of the venous sinus stenosis, two stents might be deployed to fully address the stenotic vessel segment. Following stenting, the Rebar-27 microcatheter was advanced through the stents into the SSS and post-stenting venography and manometry were performed. For both pre- and post-stenting manometry, the guide catheters were positioned in the IJ to ensure no confounding from potential outflow obstruction due to the guide catheters. Only patients from Group A underwent waveform recording during this procedure. All patients were maintained at an end-tidal carbon dioxide of 38–42 mmHg during the procedure.
Manometry and waveform recording were performed through the Rebar-27 microcatheter in all patients. The microcatheter was flushed and then attached to an arterial pressure monitoring line positioned at the level of the patient’s heart. The system was then opened to air and zeroed. Next, the system was closed and pressures were transduced. At each site at least four respiratory cycles were recorded prior to advancing to the next anatomical location. Venous manometry data were continuously collected using MediCollector (Boston, Massachusetts, USA). These data were then exported into discrete files and displayed graphically in order to facilitate analysis and interpretation of cerebral venous waveform characteristics while awake and while under general anesthesia before and after stenting.
All analyses were conducted using ‘R: A language and environment for statistical computing’ (R Foundation for Statistical Computing, Version 3.3.2, Vienna, Austria) and RStudio: Integrated Development for R., Version 1.1.383 (RStudio, Boston, Massachusetts, USA). Descriptive statistics were calculated such that mean (SD) were used for normally distributed variables and median (range) for non-parametric data. For all analyses, two-tailed hypothesis testing was used with P<0.05 interpreted for statistical significance. Prior to conducting analyses, histograms and boxplots of continuous outcome variables were studied to assess normality. Due to the small sample size and non-normality of data, Wilcoxon signed rank tests were conducted to test hypotheses between and within groups on continuous outcomes of waveform frequency, amplitude, and mean venous sinus pressure.
Patients with venous sinus outflow obstruction (Group A)
A total of 10 patients underwent venous manometry recording during an awake diagnostic angiogram and both pre- and post-stenting while under general anesthesia. The mean age of these patients was 31.2 years (range 18–47 years) and 90% were female. Median body mass index (BMI) was 33.8 kg/m2 (range 26.6–54.3 kg/m2). Median opening pressure on lumbar puncture was 30.8 cm H2O (range 26–43 cm H2O). All patients underwent angiography while awake, and all patients had a significant pressure gradient across the transverse sinus with mean and median pressure gradients of 15.1 mm Hg and 13.5 mm Hg, respectively (7–31 mm Hg).
Control patients (Group B)
Five patients who underwent venous manometry recording during an awake diagnostic angiogram did not demonstrate a significant pressure gradient and therefore did not have venous sinus outflow obstruction. These patients were included as control patients. The mean age of these patients was 32.8 years (range 23–56 years) and 40% were female. Median BMI was 29.2 kg/m2 (range 25.5–43.1 kg/m2). Median opening pressure on lumbar puncture was 26 cm H2O (range 24–35 cm H2O).
Mean venous sinus pressure measurements are shown in table 1. In control IIH patients, a normal decrement is seen as the microcatheter is withdrawn through the venous system, with approximately 1–2 mmHg change at each position (SSS to torcula, torcula to TS, etc). The average gradient observed across the transverse and sigmoid sinuses in control patients was 1.8 mm Hg (0–4 mm Hg).
Group A patients demonstrated a statistically significant reduction in median TS-SS pressure gradient from 13.0 mm Hg to 1.0 mm Hg following stenting (P=0.01). The median post-stent gradient was 1.0 mm Hg (1.0–1.8 mm Hg).
All 10 patients from Group A (100%) had higher waveform amplitudes centrally that decreased when the microcatheter was withdrawn proximally through the stenosis while awake and when under general anesthesia (table 2). The amplitude measured at the TS showed a statistically significant decrease from 8.3 mm Hg to 2.8 mm Hg (P=0.02) following stenting of the stenosis (figure 1). Following stenting, the amplitude of the SS and IJ did not show significant changes.
Waveform frequencies were recorded, and no statistically significant difference was observed between Group A and Group B (P=0.99). There was a small statistically significant change in the value of the frequency pre- and post-stenting in Group A patients (P<0.05); however, given that the frequency is primarily determined by heart rate, the difference is likely of little clinical significance.
Qualitative waveform evaluation
Awake group A venous waveform morphology
The venous sinus waveform was recorded in real-time while venous manometry was being performed (figure 2). Depending on the measurement site, the cerebral venous sinus waveform represents a summation of the two waves influencing cerebral venous drainage: the intracranial pressure (ICP) waveform as well as the CVP waveform. The classical ICP waveform is composed of three peaks: P1, P2, and P3.6 7 P1, or the percussion wave, is the first and highest amplitude peak and is thought to represent the initial arterial impulse. P2, the tidal wave, is representative of intracranial compliance and the amplitude is typically 80% of P1.8 P3, the dichrotic wave, likely represents the aortic valve closure. At distal venous sites within the calvarium, the venous waveform appears similar to a traditional ICP waveform, with a triplet peak. Interestingly, the triplet in a number of Group A patients, all with elevated opening pressures on LP, shows a larger P2 than P1, which is traditionally thought to indicate reduced brain compliance secondary to raised ICP.6 This shape is continued with minor change through the SSS, torcula, and TS. Proximally, at the SS and IJ, the venous pressure waveform appears to be predominantly influenced by the CVP waveform. At the level of the TSJ, the waveform becomes increasingly influenced by the CVP waveform, resulting in a relatively chaotic pattern as the two waveforms are both constructively and destructively interfering with each other.
Extrinsic influences on cerebral venous waveform morphology
The cerebral venous waveforms of patients in Groups A and B who were undergoing initial venography under conscious sedation were compared (figure 3). In four of the five patients in Group B there was obvious variation of the cerebral venous waveform with the respiratory cycle. In contrast, eight of the 10 patients in Group A demonstrated a qualitatively less variable baseline, with only a modest (0–1 mm Hg) change with respiration.
All 10 Group A patients underwent venous manometry and venous waveform recordings during their initial diagnostic venogram under conscious sedation as well as under general anesthesia prior to venous sinus stenting. Of these 10 patients, five had blunting of their venous waveform, with loss of the previously observed triple peak. Of note, four of the five patients who did not show this phenomenon had decreases in their trans-stenosis gradient associated with the induction of general anesthesia (range 4–18 mm Hg).
Venous sinus stenting alters the waveform morphology
In Group A patients under general anesthesia prior to stent placement, the venous waveform is characteristically a high-amplitude wave distal to the stenosis (figure 4). This wave often resembles a pathologic ICP waveform which is characteristic of reduced brain compliance, which often has a pronounced P2 and is otherwise featureless. Proximal to the stenosis, the waveform abruptly takes on a distinctively CVP-like morphology, which evolves towards the CVP waveform morphology as the measurement point progresses more proximally. Following stenting, all sites contain elements of CVP morphology. The distal waveform at the SSS demonstrates significant influence from the ICP waveform and this morphology progresses very rapidly to CVP morphology as the measurements are taken more proximal.
In this prospective study, patients with medically refractory IIH undergoing stenting for venous sinus stenosis underwent concurrent venous manometry as well as venous waveform recording. Overall, the waveform morphology appears to represent contributions from the ICP and CVP waveforms, with a transition from ICP to CVP-dominated waveform when progressing from distal to proximal. Venous outflow obstruction causes an ICP-dominated waveform with pronounced P2 peaks and significantly increased waveform amplitude. Venous stenosis stenting resulted in statistically significant reductions in the trans-stenosis pressure gradient as well as the amplitude of the associated venous waveform at the level of the stenosis. The use of general anesthesia was associated with blunting of the venous waveform. Additionally, the presence of a trans-stenosis gradient was associated with less respiratory variation of the cerebral venous waveform. This study is the first of its kind to describe the qualitative features of the venous pressure waveform in patients with IIH before and after treatment of their venous stenosis as well as a group of controls without significant venous pressure gradient.
The cerebral venous waveform appears to be primarily influenced by two factors along its course. Distally, within the calvarium, the waveform is almost indistinguishable from the ICP waveform, including the three peaks traditionally associated with an ICP waveform. This influence of ICP on cerebral venous sinus pressure has been described previously by Pickard et al, who noted co-variation of cerebral venous sinus pressure with drainage of CSF in patients with IIH.9 Interestingly, in a number of the patients who all had elevated opening pressures on lumbar puncture, the second peak (P2) was the highest amplitude, which is known to be a sign of increased ICP and reduced brain compliance when related to the ICP waveform.6–8 10 Prior to stenting, the waveform proximal to the stenosis resembles a CVP waveform. There is a relatively abrupt transition from CVP to ICP dominance of the waveform with marked elevation in waveform amplitude on crossing the venous sinus stenosis. Following stent placement, the amplitude distal to the stent returned to normal. Additionally, the transition from SSS to IJ occurred progressively, with the CVP elements evident at the SSS and becoming more prominent from distal to proximal. Qualitatively, the highest peak at the distal measurement sites (SSS and torcula) tended to change from P2 to P1 following stenting. This change in waveform morphology could be a qualitative correlate to the immediate decrease in ICP following stenting, which has been previously reported in prospective studies with concomitant ICP monitoring.11
A number of factors appear to influence the cerebral venous waveform. In Group B patients who did not have a pressure gradient across their stenosis, respiratory variation of the waveform was observed at all measurement sites. In Group A patients, the sites distal to the stenosis did not show consistent variation with the respiratory cycle. Presumably, the respiratory variation is related to the effect inspiration has on CVP. The presence of a hemodynamically significant stenosis with an associated pressure gradient likely reduces the transmission of the negative pressure distal to the gradient, resulting in the observed blunting of the respiratory variation.
In addition to observed respiratory variation, general anesthesia appears to have an impact on the cerebral venous waveform. All 10 Group A patients had diagnostic venograms under conscious sedation as well as pre-stenting measurements repeated under general anesthesia. Overall, five of the 10 patients had blunting of their distal (SSS, torcula, and TS) measurement waveforms when under anesthesia compared with awake. The majority (4/5) of the patients who did not show blunting of their waveform with general anesthesia showed a decrease of their pressure gradient from conscious sedation to general anesthesia (range 4–18 mm Hg). Although the etiology of waveform blunting is unclear, previous reports have described this phenomenon in the ICP waveform and concluded that it represented a decrease in intracranial compliance secondary to increased ICP.6–8 10 This finding is of potential clinical significance given the fact that many of the patients who are treated for venous sinus stenosis have documented increased ICP as well as likely decreased reserve for accommodating subtle changes in ICP. Determining the exact cause of this finding is outside the scope of this investigation. However, other reports have detailed the variable impact of general anesthesia on the trans-stenosis venous pressure gradients.12 13
Few data have been published regarding the venous waveform, especially in pathologic states such as IIH with venous sinus stenosis and pressure gradients. There are obvious and conserved changes that occur in the waveform that have implications for treatment. Given the previous reports of changes in gradient size in relation to general anesthesia, measurement of the gradient at the time of stenting is unreliable.12 13 This presents a problem for the interventionalist, as an objective measurement of treatment of the hemodynamically significant stenosis is unreliable at the time of stenting. Change in the distal waveform pattern (return of peaks which mimic the ICP waveform, return of evidence of CVP waveform influence distal to the treated stenosis and reduction in waveform amplitude) may be indicative of successful treatment of a hemodynamically significant venous sinus stenosis when pressure gradients disappear across the stenosis under anesthesia.
This study is limited by its small cohort size. Additionally, it is a single institution study subject to the biases associated with referral patterns as well as a single treating physician. The qualitative nature of the analysis precludes significant statistical inferences regarding the impact of potentially important co-variables. The venous waveform may also be influenced by technical factors, such as the use of general anesthesia or respiratory variations.12 13 Previous reports have investigated the effect of general anesthesia on pressure measurements, but the effects of various anesthetic agents including sedatives, paralytics, or inhalational anesthetics are unknown. Additionally, throughout the procedure, end-tidal carbon dioxide was kept at 38–42 mm Hg to try to limit variability secondary to changes in cerebrovascular tone. Only one study has evaluated the impact of microcatheter on venous manometry accuracy, but this study suggests 0.027 inch catheters are a reasonable means of performing manometry accurately.14 Additionally, control patients (Group B) had IIH and therefore their data should not be assumed to be representative of the normal population, but instead to be representative of patients with elevated ICP in the absence of venous outflow obstruction.
The cerebral venous waveform has morphologic characteristics influenced by both the ICP waveform as well as the CVP waveform. Venous sinus stenosis with venous outflow obstruction alters the venous waveform causing high amplitude ICP-influenced waveforms proximal to the stenosis. Alterations in the waveform return to match IIH controls following stenting. The clinical and hemodynamic ramifications of these changes are not well understood and warrant further investigation.
Contributors JLW: conception, design, data collection, data analysis, manuscript preparation. RMG: data collection, manuscript preparation. GPG: data collection. JRT: data collection, manuscript preparation. CAA: design, data analysis. JS: conception, design, data collection, manuscript preparation. SQW: conception, design, data collection, manuscript preparation. KMF: conception, study design, data collection, data analysis, manuscript preparation.
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.
Patient consent Not required.
Ethics approval Wake Forest Baptist Health IRB.
Provenance and peer review Not commissioned; externally peer reviewed.
Data sharing statement Unpublished data are available upon request from the corresponding author.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.