Introduction Venous sinus stenting is a popular treatment strategy for patients with high venous sinus pressure gradients across a site of outflow obstruction. Little is known about the effect of anesthesia on venous sinus pressure measurements.
Objective To compare venous manometry performed in patients under general anesthesia and while awake.
Methods A prospective database was accessed to retrospectively identify patients who had undergone venous sinus stenting. Pressure gradients were compared between those patients who underwent manometry while awake and before stenting under general anesthesia.
Results Thirty patients with both general anesthesia and awake pressure recordings were identified. Pressure measurements were highly variable but overall were higher under general anesthesia by an average of 5.8 mm Hg (1.7; p=0.002). A significant difference between awake and general anesthesia pressure measurements was detected in the sigmoid sinus (5.8 mm Hg (2.0); p=0.005) and the jugular vein (8.1 mm Hg (3.9); p=0.040). Only 11/30 (36.7%) pressure gradients remained within 5 mm Hg of the original awake gradient when repeated under general anesthesia; 9/30 (30%) patients had gradients that were at least 10 mm Hg different across procedures.
Conclusions Calculated pressure gradients were markedly affected by anesthesia. These findings suggest that candidacy for stenting should be determined with venous manometry while patients are awake owing to the unpredictable and highly variable effect of general anesthesia on pressure measurements and an apparent tendency to underestimate the degree of venous outflow obstruction.
- Blood Pressure
Statistics from Altmetric.com
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.
Venous sinus stenting has become a popular treatment strategy for a subset of patients with idiopathic intracranial hypertension (IIH).1 ,2 Certain patients with symptomatic venous sinus stenosis without IIH may also benefit from stenting.3 Although the pathophysiology of IIH and venous sinus stenosis remains poorly understood, the available literature suggests that some patients have venous sinus outflow obstruction from sinus stenosis or occlusive arachnoid granulations that manifest as central cerebral venous hypertension. These phenomena most commonly occur in the transverse sinus (TS) or at the transverse-sigmoid sinus (SS) junction. Venous outflow obstruction is suggested by the presence of both an elevated central cerebral venous pressure and a pathologic venous pressure gradient across an area of obstruction using a microcatheter. General candidacy for stenting includes a central cerebral venous pressure of ≥22 mm Hg and a pressure gradient of at least 8 mm Hg.2 ,4 Successful patient selection therefore depends on accurate venous pressure measurements.
Patients with IIH typically have elevated intracranial pressure and headache, which often becomes worse when patients lie flat for cerebral angiography. Further, headaches and auricular pain are commonplace with navigation of catheters through the jugular bulb and into the venous sinuses. This may encourage practitioners to use general anesthesia for venography and manometry, with stenting performed if a pressure gradient exists. However, little is known about the effect of anesthesia on venous sinus pressure measurements. At our center, we perform awake angiography with venous manometry and then ask those patients who are candidates to return later for stenting under general anesthesia. We sought to study the correlation between venous sinus pressures obtained in patients while awake with venous sinus pressures obtained before stenting while under general anesthesia.
Institutional review board approval was obtained to perform a retrospective review of all patients who underwent venous sinus stenting between January 2010 and June 2016 at our center. A prospectively collected database was examined to identify all patients who underwent venous sinus stenting during this time period and who had also undergone an awake diagnostic cerebral angiogram with manometry before the stenting procedure. Next, procedure reports were reviewed to identify all patients who had venous manometry performed both during the awake angiogram and before the stent was placed while under general anesthesia. Patients meeting both these criteria were included in the analysis.
Chart reviews were then performed to access demographic data (diagnosis, age, gender), procedural data, anesthetic agents, and venous sinus measurements. The number of days between awake angiography and stenting under general anesthesia was recorded. Occasionally, multiple venous sinus pressures were recorded at the same location during the same procedure. In such cases, these values were averaged to obtain a single representative pressure measurement. End-tidal carbon dioxide was reported at the approximate time that pressure measurements were recorded.
Awake venous manometry
Angiography, venography, and venous sinus manometry are performed at our center on patients who are awake without conscious sedation or with mild analgesia (IV fentanyl and midazolam) during a single session. 5 F sheaths are placed into both the femoral artery and femoral vein under local anesthesia. First, patients undergo cerebral arteriography with a 5 F diagnostic catheter to evaluate venous sinus stenosis. Next, the arterial catheter is removed and placed into the venous system. The diagnostic catheter is advanced into the right and/or left internal jugular vein (JV) near the jugular bulb. A 0.027 inch Renegade Hi-Flo microcatheter (Boston Scientific; Marlborough, Massachusetts, USA) is then navigated over a microwire into the superior sagittal sinus (SSS) and venography is performed. The microwire is removed, the microcatheter is flushed and attached to an arterial pressure line, and venous manometry is then performed in the SSS, torcula, ipsilateral TS, ipsilateral transverse-SS junction, ipsilateral SS, and ipsilateral JV. At each location, the pressure is allowed to stabilize before pressure recordings are determined. Depending on venous anatomy, the microcatheter is then either advanced across the torcula into the contralateral venous sinuses or the 5 F diagnostic catheter is withdrawn and navigated into the contralateral JV for contralateral venous sinus manometry via the Renegade microcatheter. The access sites are then sealed with a closure device and the patients are discharged to home after a brief recovery period. Those who are deemed candidates for stenting are scheduled electively to return for stenting under general anesthesia at a later date.
Manometry before venous sinus stenting under anesthesia
Venous sinus stenting is performed under general anesthesia. Patients are loaded with aspirin and clopidogrel before the procedure. After induction, the right femoral vein is accessed, an 8 F sheath is placed, and IV heparin is administered. A 0.070–0.088 guide catheter (most commonly a Neuron MAX (Penumbra; Alameda, California, USA)) guide catheter) is navigated into the ipsilateral JV near the jugular bulb. Pre-stenting manometry is variably performed. When pre-stenting manometry is performed, a Renegade Hi-Flo microcatheter is used to measure ipsilateral venous pressures across the site of outflow obstruction. After manometry, stenting is performed in the standard fashion.
Before conducting any formal hypothesis testing, descriptive statistics were generated and presented as frequency (%) and mean (SD, range) for categorical and continuous measures, respectively. To test pressure differences between awake and general anesthesia by location and in total, sides were pooled and a repeated-measures, mixed-effects linear regression model, controlling for all relevant covariates, was constructed. Model pressure estimates (SE) are presented by location for awake, general anesthesia, and the difference between them. To test for the pressure gradient, a linear model controlling for the same covariates identified in the aforementioned model was constructed and results are presented similarly. Statistical significance was assessed at α=0.05. All hypothesis testing was performed using SAS V.9.4 (c).
During the study, 72 patients underwent venous sinus stenting. Of these, 30 patients had venous sinus pressure measurements obtained while both awake and before stenting while under general anesthesia. The majority of the 42 excluded patients underwent stent placement without pre-stent manometry being repeated.
Demographic and procedural details for the 30 included patients are shown in table 1. A total of 326 pressure measurements were recorded; more than two-thirds of the pressure recordings were obtained from the awake procedure (219; 67.2%) than while under general anesthesia (107; 32.8%). One hundred and four locations contained pressure measurements recorded during both procedures in the same patients allowing for comparison (mean 3.5 comparable pressure readings per patient, SD 0.93, range 2–5).
Details of IV medications administered during the awake procedure were available in 29/30 patients (97%). Ten of 29 patients (34%) had no IV medications administered, 1 (3%) had 75 μg of IV fentanyl and 0.5 mg of meperidine, 4 (14%) had IV fentanyl only (mean 56.3 μg, range 25–100 μg), and 14 (48%) patients had both IV fentanyl (mean 68.8; range 25–150 μg) and midazolam (mean 0.9 mg; range 0.5–2 mg) administered during the procedure.
Specific anesthetic details were available for 29/30 procedures (97%). Endotracheal intubation was used in 28 of the 29 procedures (97%) with laryngeal airway used in 1 patient (3%). IV midazolam, fentanyl, and propofol were administered to all patients during the procedure. Inhalational anesthetic was administered in all but two patients (93%): sevofluorane in isolation in 21 patients (72%), isofluorane in 4 (14%), and a combination of the two in 2 patients (7%). Muscle paralytic agents, including rocuronium, vecuronium and/or succinyl choline, were used in all but five patients (83%). Adjunctive ketamine was used during one procedure (3%). End-tidal carbon dioxide was reported during 28 of the 30 procedures (93%), with a mean of 34.5 mm Hg (SD 3.2, range 29–40).
Comparison by location
Pressure measurement comparisons are shown in table 2 and figure 1. A significant difference between awake and general anesthesia pressure measurements was detected in the SS (5.8 mm Hg (2.0); p=0.005), the JV (8.1 mm Hg (3.9); p=0.040), and overall (5.8 mm Hg (1.7); p=0.002). Though not statistically significant, all other locations and the pressure gradient had greater pressure estimates for patients under general anesthesia than when awake. Covariates identified and controlled for included age, where older age was significantly associated with higher venous pressures (p<0.001), and midazolam use (p<0.001). Furthermore, the use of muscle relaxants (p=0.002) and midazolam (p<0.001) also had significant effects on pressures measured while under general anesthesia and awake, respectively. End-tidal carbon dioxide was not correlated with pressures (p=0.75) measured under general anesthesia.
Table 3 shows the absolute magnitude of pressure differences for patients awake and under general anesthesia, according to location. Only 30–50% of measurements performed under anesthesia were within 5 mm Hg of the data obtained while awake. Forty-three per cent of SSS, 25% of torcula, 25% of TS, and 19% of SS measurements were different by ≥10 mm Hg. Although differences occurred in both directions, pressure measurements obtained under general anesthesia were more likely to be ≥6 mm Hg higher rather than ≥6 mm Hg lower for the SSS (43% vs 26%), torcula (30% vs 25%), and TS (35% vs 18%). This was not seen in the SS, where no pressure measurements were more than 5 mm Hg smaller under general anesthesia compared with awake. In fact, 80% of SS readings were higher by at least 10% when performed under general anesthesia compared with awake.
Pressure gradient comparison
Table 3 compares the calculated pressure gradients obtained in awake and anesthetized patients. Only 11/30 (36.7%) pressure gradients remained within 5 mm Hg of the original awake gradient when repeated under general anesthesia; 9/30 (30%) patients had gradients that were at least 10 mm Hg different across procedures. Almost half (47%) of the gradients were at least 50% smaller when obtained under general anesthesia compared with the awake setting, with nearly three-quarters (73%) being at least 10% smaller.
Of the 30 awake procedures, 24 (80%) pressure gradients were ≥8 mm Hg. Only 15/30 (50%) patients had pressure gradients of ≥8 mm Hg while under general anesthesia.
This study demonstrates marked differences between venous manometry performed during awake cerebral angiography compared with measurements obtained before venous sinus stenting under general anesthesia. Overall, venous pressure measurements were higher in patients while under general anesthesia, but pressure recordings were highly variable. General anesthesia did have a consistent effect on SS measurements, with 80% of patients having at least a 10% increase compared with those obtained while awake. Consequently, calculated pressure gradients were markedly affected by anesthesia, with nearly three-quarters (73%) being reduced by at least 10% and nearly half (47%) decreased by at least 50%. This resulted in 40% of gradients being at least 6 mm Hg lower when repeated under general anesthesia.
Although venous pressures obtained under anesthesia were generally higher, the effect on individual central cerebral venous pressures (SSS and torcula) appears unpredictable. Roughly half (45–52%) demonstrated an increase in pressures of >10%, whereas over one-quarter (26–30%) of patients had a reduction in pressures >10% under anesthesia compared with awake. The reasons for this variability are unclear, but do not appear to be related to different anesthetic agents or carbon dioxide goals during anesthesia. A larger guide catheter (6 F) was used during pre-stent measurements compared with a 5 F catheter used during the awake procedures, which might have increased pressures by slightly affecting venous outflow through the JV, although this is unlikely to cause significant changes. The preferential increase in pressure occurring at the SS is interesting and its cause is unclear. The non-significant increase seen at the JV in four patients suggests that this might be a central venous phenomenon, but central systemic venous pressures were not recorded and therefore this relationship cannot be confirmed by this study.
Most neurointerventionalists consider the pressure gradient across the region of obstruction as the principal factor determining candidacy for venous sinus stenting, with a threshold of ≥8 mm Hg indicating candidacy.4 Our study suggests that gradients determined while a patient is under anesthesia are likely to vary greatly from those obtained while patients are awake. Most commonly, anesthesia appears to exaggerate the SS pressure, more so than the TS pressure, resulting in a consequent reduction in the calculated gradient. In the 24 patients who had a gradient of ≥8 mm Hg determined while awake, only 15 had a gradient of ≥8 mm Hg that persisted under anesthesia. These data therefore suggest that pressure measurements obtained under anesthesia may falsely rule out patients for stenting by underestimating the degree of venous outflow obstruction. Conversely, nearly one-quarter of patients (23%) showed a >5 mm Hg increase in their gradient under anesthesia, which might falsely suggest that stenting would be of benefit.
Overall, these findings argue that candidacy for stenting should be determined with venous manometry while patients are awake owing to the unpredictable and highly variable effect of general anesthesia on pressure measurements. Furthermore, patients receiving midazolam for the awake procedure exhibited higher pressures than those who did not receive this medication (p<0.001). Midazolam has been shown to decrease cerebral blood flow, cerebral perfusion pressure, and mean arterial blood pressure while increasing arterial carbon dioxide tension after IV administration.5 ,6 The relevance of this finding in our patient population is unclear but might be explained merely by an increased requirement for sedation in those with more pain. This finding does suggest that midazolam should be avoided, if possible, when performing awake venous manometry.
This study has a number of limitations, including its retrospective design and the intrinsic variability that occurs during venous sinus pressure recordings. Previous studies have demonstrated large interpatient variability in blood flow velocity and other cerebrovascular hemodynamic metrics,7 which might explain the high degree of variability between awake and asleep measurements. The venous sinus stenting procedure occurred within 1 week of the original diagnostic angiogram in only 8 patients (27%), but within 30 days in 22 (73%; mean 25.8 days). It is possible that physiologic changes that occurred during this period, unrelated to anesthesia, might have affected the patient's pressure recordings. End-tidal capnography measurements were recorded at the approximate time of venous manometry and may be slightly inaccurate. Blood pressure recordings were highly variable and could not be accurately obtained from the time of manometry, so this factor was not analyzed.
This retrospective study of 30 patients, which compared venous sinus manometry measurements obtained while awake with those under general anesthesia, demonstrates significantly higher pressure measurements in patients while under general anesthesia but with considerable variability across locations. A more consistent effect of anesthesia was seen at the SS, with 80% of patients having at least a 10% increase in pressure measurements compared with those taken while awake. Consequently, calculated pressure gradients were markedly affected by anesthesia, with 40% being reduced by at least 6 mm Hg. Overall, these findings argue that candidacy for stenting should be determined with venous manometry while patients are awake owing to the unpredictable and highly variable effect of general anesthesia on pressure measurements and an apparent tendency to underestimate the degree of venous outflow obstruction.
Contributors Each author listed above should receive authorship credit based on the material contribution to this article, their revision of this article, and their final approval of this article for submission to this journal.
Competing interests None declared.
Ethics approval Medical University of South Carolina institutional review board.
Provenance and peer review Not commissioned; externally peer reviewed.