Introduction Venous outflow obstruction has been implicated in the pathophysiology of a subset of patients with idiopathic intracranial hypertension (IIH), and venous sinus stenting (VSS) has emerged as an effective treatment. However, the effect of anesthesia on venous sinus pressure measurements is unpredictable. A more thorough understanding of the effect of the level of anesthesia on intracranial venous pressures might help to better define patients who might benefit most from stent placement.
Objective To compare, in a retrospective cohort study, intracranial venous pressures measured under conscious (CS) sedation versus general anesthesia (GA) and to assess the relationship between anesthetic-dependent venous pressures and outcomes after VSS.
Methods We performed a retrospective review of a prospectively maintained database to identify patients undergoing angiographic evaluation and VSS for intracranial venous stenosis. Mean venous pressures (MVPs) and trans-stenosis pressure gradients obtained under CS were compared with those measured under GA.
Results The maximal MVP was significantly lower under GA (19.8 mm Hg) than CS (21.9 mm Hg; p=0.029). The MVPs in the superior sagittal sinus, torcula, and transverse sinus were lower under GA, but were significantly higher in the sigmoid sinus and jugular bulb under GA (p<0.001). The mean trans-stenosis pressure gradient was also significantly lower under GA (8.6 mm Hg) than CS (12.1 mm Hg; p<0.001). Patients with a larger difference between maximum MVP under GA versus CS were more likely to have normalization of the MVP after VSS (p=0.0008).
Conclusions Intracranial venous pressures are markedly affected by GA. In order to obtain an accurate measurement of MVPs and trans-stenosis gradients, patients undergoing investigation for IIH should undergo cerebral angiography and venous manometry under CS, which provides more reliable data for outcomes after VSS.
- Intracranial Pressure
Statistics from Altmetric.com
Intracranial venous sinus stenosis has been associated with idiopathic intracranial hypertension (IIH), either as a causative factor or as a consequence of external compression from increased intraparenchymal pressure.1–5 Venous sinus stenting (VSS) for patients with concomitant IIH and intracranial venous stenosis with a trans-stenosis pressure gradient has been shown in retrospective series to reduce intracranial pressure (ICP), improve visual outcomes, and ameliorate headaches and tinnitus.4–21 For patients not requiring urgent intervention for rapid deterioration of vision, the typical practice is to evaluate venous pressures on an elective basis under light conscious sedation (CS). The mean venous pressure (MVP) at various locations in the intracranial venous sinus system is recorded, and the trans-stenosis gradient is used as a basis for treatment decision-making. A pressure gradient of 8–10 mm Hg across an intracranial venous sinus stenosis has been the threshold for intervention in most series.1 ,9 Owing to the patient discomfort associated with manipulation of the dural venous sinuses, VSS is typically performed under general anesthesia (GA). In some centers, practitioners may elect to combine procedures to perform both venous manometry and stenting in a single session for the sake of convenience. Since our practice has been to repeat venous pressure measurements under GA before and after VSS to quantify the effect of stent placement on the MVP and trans-stenosis pressure gradient, we have been able to compare these measurements under differing levels of anesthesia.
The effect of CS or GA on venous pressure measurements appears to be heterogeneous, but has not been widely reported.18 ,19 The aims of this study are to (1) characterize differences in venous pressure measurements under CS, compared with GA, and (2) assess the relationship between anesthetic-dependent venous pressures and outcomes after VSS.
Institutional review board approval was obtained to perform a retrospective review of a prospectively maintained database of patients who had undergone diagnostic angiography and venous manometry for investigation of possible intracranial venous sinus stenosis at our institution. The prospective database has been maintained since February 2014, and this was reviewed on July 1, 2016. Primarily, these studies were performed as a part of the evaluation for IIH, which was diagnosed according to the modified Dandy criteria,22 and has been described in detail in a prior publication.8 Patients who had both cerebral angiography with venous manometry under CS and a VSS procedure performed under GA were included in the study. Patients in whom repeat venous manometry was not performed under GA before VSS were excluded.
Medical charts, imaging findings, and anesthetic reports were reviewed to access patient demographics, intraprocedural findings, and details of administered anesthetic medication. Imaging reports were reviewed to assess MVPs recorded at various locations throughout the venous sinus system for comparison between cases performed under CS versus GA.
Venous manometry under CS
Diagnostic cerebral angiography and venous manometry were performed in a single session under CS. IV sedation was titrated by the procedure nurse, according to patient needs. Typically, 25–50 µg of fentanyl and 1–2 mg of midazolam were administered IV before groin access. Arterial and venous access is gained at the start of the case, and a diagnostic angiogram was obtained in a standard fashion. Next, a 5 French (Fr) diagnostic catheter was advanced into the dominant internal jugular vein (IJV) without a sheath, and an 0.027 inch microcatheter was advanced over a microwire into the anterior superior sagittal sinus (SSS). Venous manometry was then performed, with recording of MVPs in the SSS, torcula, transverse sinus (TS), sigmoid sinus (SS), jugular bulb, and IJV.
Our VSS procedure has been previously described in detail.8 Briefly, after premedication with aspirin and clopidogrel, an assessment of aspirin reaction units (ARU) and P2Y12 reaction units (PRU) levels was made on the morning of the procedure. For patients with ARU >551 or PRU >208, additional antiplatelet loading was administered. VSS cases were performed under GA in all cases. IV heparin was administered on a weight-based protocol (100 units/kg). Arterial and venous access was obtained, and a shuttle was placed in the IJV ipsilateral to the stenosis. Following pre-VSS angiography and venous manometry, stenting was performed in a standard fashion.
Assessment of outcomes and statistical analysis
Descriptive statistics for comparison of maximum MVP and pressure gradients between groups (CS vs GA) at various locations were first generated. Data are presented as mean and SD for continuous variables, and as frequency for categorical variables. Categorical variables were compared using Pearson's χ2 and Fisher's exact tests, as appropriate; continuous variables were compared using the Wilcoxon matched-pairs signed rank test, assuming a non-Gaussian distribution of variables (eg, MVP and pressure gradient).
Since a threshold of 8 mm Hg is generally used for determination of qualification for stent placement,1 ,9 ,19 an assessment of the proportion of patients who initially met this threshold, but then fell below the threshold under GA, was performed. p Values <0.05 were considered statistically significant. Statistical analysis was carried out using GraphPad Prism V.7 (GraphPad, San Diego, California, USA).
During the study period, 89 patients underwent angiographic investigation for evaluation of intracranial venous sinus stenosis. Of these, 61 had venous manometry recorded under both CS and GA. The mean age was 34.2 years (range 14–62 years, SD 12.3 years), and 60 (98.4%) were female. The mean body mass index was 33.7 kg/m2 (SD 10.1 kg/m2). A total of 192 comparable MVP measurements were obtained in patients under CS and GA (mean 3.1 comparable MVP readings per patient, range 2–7). Anesthetic agents used included sevoflurane in 53 patients (86.9%), propofol infusion in seven patients (11.5%), and dexmedetomidine in one patient (1.6%).
MVPs and pressure gradients under CS versus GA
For all patients, matched-pair analysis of maximum MVP at various points in the intracranial venous circulation is shown in figure 1. MVPs were higher under CS than GA in the SSS, torcula and TS; whereas MVPs were higher under GA in the SS and jugular bulb. MVPs were significantly different between CS and GA in the TS (p=0.026) and SS (p<0.001). The maximum MVP was significantly lower under GA (p=0.029; figure 2).
The trans-stenosis pressure gradient was also significantly lower under GA (p<0.001). The pressure gradient was 12.1 mm Hg under CS, compared with 8.6 mm Hg under GA. Pressure gradients of ≥8 mm Hg are considered significant under CS, and of these patients, 10/38 (26.3%) had pressure gradients <8 mm Hg under GA. The pressure gradient measured under GA was within 5 mm Hg of that measured under CS in 66.7% (40/60 patients).
Outcomes after VSS
We performed a matched-pair analysis of patients in whom MVPs were recorded immediately after VSS under GA, and at follow-up under CS (n=27). Among patients with an initial maximum MVP >20 mm Hg who achieved a normal maximum MVP (<20 mm Hg) after VSS (n=21), the pre-VSS maximum MVP was 24.6 mm Hg under CS and 18.3 mm Hg under GA (p=0.0001). Among patients with maximum MVP that remained elevated after VSS (n=6), the pre-VSS maximum MVP was 31.2 mm Hg under CS and 34.5 mm Hg under GA (p=0.625). For all 27 patients, the degree of difference between maximum MVP values under CS and GA was significantly associated with immediate post-stent maximum MVP: specifically, a larger absolute difference between maximum MVP under CS versus GA was more likely to result in MVP normalization after VSS (p=0.0008).
Measurements of MVP are an integral component of the diagnostic evaluation of patients under consideration for VSS, and provide information that is the basis of therapeutic decision-making about the potential benefit of endovascular treatment.1 ,9 Among patients with IIH who are refractory to medical therapy, indications for VSS generally include: (1) radiographic evidence of intracranial venous sinus stenosis; and (2) a trans-stenosis pressure gradient of at least 8–10 mm Hg.1 ,9 ,19 VSS in these patients has been shown to obliterate the pressure gradient, normalize ICP, and improve ophthalmologic and neurological outcomes.1–4 ,6–8 ,17 ,18 ,23 ,24 In this study, we have shown that MVP measurements are significantly affected by the level of anesthesia. Specifically, overall maximum MVP is significantly reduced by GA; MVPs in the SSS, torcula, and TS are decreased; and MVP in the SS is increased under GA. This results in a significantly lower trans-stenosis pressure gradient, which is typically located at the TS–SS junction.
Fargen and colleagues examined the effect of anesthesia on venous pressure readings among patients undergoing VSS.19 In their series, a high variation in venous pressure measurements between CS and GA was reported, but overall measurements were higher under GA. In contrast, we have found that maximum MVP consistently decreases under GA compared with CS. The differences between cohorts are not entirely clear, and require further investigation. In the prior study, a wide variability was seen across different patients. For example, for pressure readings in the SSS and torcula, the prior study reported that 42–52% of patients had ≥10% increase in MVP under GA, while 26–30% had a reduction of ≥10% under GA. In contrast, 21% of patients in our series experienced an increase of ≥10%, whereas 42% had a decrease of ≥10% in the SSS and torcula under GA. In the SS, 59% of our patients had a ≥10% increase in MVP, whereas only 20% had a reduction of ≥10% in MVP under GA. The change in trans-stenosis pressure gradient was smaller under GA compared with the prior series.19 In the prior study, the pressure gradient increased by ≥5 mm Hg in 7/30 patients (23%), was unchanged in 11 (37%), and decreased in 12 (40%).19 In our study, ≥5 mm Hg increase was seen in 4/61 patients (6.6%), was unchanged in 40 (65.6%), and decreased by ≥5 mm Hg in 17 (27.9%).
Despite these differences between cohorts, it is clear that a number of factors could potentially affect venous pressure, causing differences between anesthetic level and between patients. Obstructive sleep apnea and other respiratory disorders can lead to significant variation in ICP, owing to hypoxic and hypercapnic cerebral vasodilatation.25 Under GA, oxygenation and ventilation are monitored and maintained in a controlled manner, but other factors causing variation between anesthetic levels are unclear.
Our patients generally underwent CS and anesthesia using similar agents to those reported by Fargen et al.19 Our VSS practice varies by practitioner, but most often uses a larger sheath system with a 12F sheath in the ipsilateral IJV, which could contribute to higher MVPs under GA. Other differences in results may be attributable to differences in interpatient variability between the cohorts. Interestingly, those patients who demonstrated greater changes in maximum MVP between CS and GA were more likely to achieve normalization of maximum MVP after VSS (p=0.0008). This normalization was maintained at follow-up in 85% of cases. The group of patients that did not normalize maximum MVP had an overall higher maximum MVP (31 vs 25 mm Hg), and still achieved a reduction in maximum MVP after VSS, despite not reaching the 20 mm Hg threshold. This finding may therefore be related to refractory elevated ICP among patients with higher maximum MVPs before VSS, in whom there may be a limited capacity for MVP reduction.
Limitations of this study include its retrospective design, although the data were prospectively collected. Although the approach to CS and GA was consistent, GA was provided by a number of different anesthesiologists, and agents and dosage varied somewhat by practitioner. Data regarding end-tidal CO2 throughout VSS were not available for review, and might have affected MVP findings. Patient-related factors, such as obstructive sleep apnea, might have contributed to variations in maximum MVP, but were not able to be objectively accounted for in our series. Although we have classified the patient sedation states as ‘CS’ and ‘GA’, there may be differences between CS, which may be considered a form of ‘light anesthesia’, and no sedation. Cerebral angiography without sedation, although used in many cases in which patients have prior documentation of allergy or adverse effects related to benzodiazepine administration, has not been systematically reported in the literature to our knowledge, and requires further characterization. There may be similar differences between CS and no sedation states from those reported in our study. Nevertheless, the consistency of our findings, despite a modest variation in anesthetic and sedative medication usage, suggests that the observations are reliable. Finally, follow-up angiography for assessment of outcomes after VSS were only available in a minority of patients included in this cohort, and results may be significantly different after completion of follow-up in all patients.
Our series of patients undergoing venous manometry under both CS and GA demonstrates that MVP is significantly affected by anesthetic level. In general, the use of GA reduces the observed pressure gradient by decreasing MVP in the SSS and TS, and increasing MVP in the SS. Since diagnostic and follow-up angiography under CS more accurately represent the awake patient's physiology, MVP measurements taken under CS should be used as the basis for decision-making about a patient's suitability for VSS, and for evaluation of technical post-stenting outcomes.
Contributors DMSR conceived the study, performed data analysis, drafted the manuscript, and approved the final version of the manuscript. TJB, DD assisted in data analysis, critically revised the manuscript, and approved the final version of the manuscript. C-JC, RMS critically revised the manuscript, and approved the final version of the manuscript. KCL contributed to the design of the study, oversaw data collection and analysis, critically revised the manuscript, and approved the final version of the manuscript.
Competing interests None declared.
Ethics approval University of Virginia Institutional Review Board for Health Sciences Research.
Provenance and peer review Not commissioned; externally peer reviewed.
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.