Anesthesia modality does not affect clinical outcomes of intra-arterial vasodilator treatment in patients with symptomatic cerebral vasospasms

Introduction

Cerebral vasospasms (CVS) and delayed cerebral ischemia still remain among the leading causes of morbidity and mortality in survivors of aneurysmal subarachnoid hemorrhage (aSAH). Up to 40% of aSAH patients experience symptomatic CVS, resulting in disability in up to 50% thereof.¹ CVS, a narrowing of cerebral arteries thought to be caused by blood breakdown products, mostly develop between 5 to 14 days after aSAH.² So far, there is no therapy known, which was shown in randomized trials to improve cerebral perfusion and thus to avoid brain ischemia and infarction in symptomatic patients. Commonly used rescue treatments for symptomatic CVS include induced hypertension, and in refractory CVS angioplasty or intra-arterial application of vasodilators, e.g. nimodipine or papaverine.^1,3,4 Both have been shown in case series to improve neurological outcome in said patients.^5–7

In recent years, several studies investigated the best method of anesthesia for endovascular treatment in acute ischemic stroke.^8–20 While initially in mostly retrospective studies, data showed conscious sedation (CS) to be superior,^8,9 a recent meta-analysis showed no significant difference in outcomes for CS and general anesthesia (GA)¹⁷ if only randomized controlled trials were considered.^15,16,18 To the best of our knowledge, no studies comparing CS and GA in endovascular treatments for refractory CVS after aSAH have been performed.

The aim of this study is to compare six-month outcomes for choice of sedation in patients treated with endovascular vasodilators for CVS after aSAH.

Methods

This is a single-center retrospective case-control study analyzing clinical outcomes in patients with symptomatic CVS after aSAH treated with endovascular vasodilators at the University Hospital Bern, Bern, Switzerland.

Results

In total, 109 patients with refractory CVS treated by intra-arterial admission of either nimodipine and/or papaverine between September 2011 and October 2019 at the University Hospital Bern, Bern, Switzerland were included. Of those, 11 patients had to be excluded. Reasons for exclusion were incomplete data (n = 2), re-rupture of aneurysm during the hospital stay (n = 2), continuous intra-arterial nimodipine treatment (n = 1) and loss of follow-up at six months (n = 6). The final study population consisted of 98 patients, 23 patients in the “CS” group, 53 patients in the “GA” group and 22 patients in the “both” group. In the “CS” group, 16 patients received a single intervention (“single CS”) and seven patients received up to five interventions (“multiple CS”). In the GA group, 26 patients received one intervention (“single GA”) and 27 patients received 2-10 treatments (“multiple GA”). As per definition, all patients in the “both” group received more than one and up to 10 interventions.

Primary outcome

mRS at six months is displayed in Figure 1a. There was a tendency for a slightly higher percentage of functionally independent patients (mRS 0-2) in the “CS” group (78.3%) at six months. However, this difference did not prove to be statistically significant (p = 0.109). The subgroup analysis comparing single and multiple intra-arterial interventions separately for each anesthesia modality is displayed in Table 2. This analysis also revealed no significant difference in functional outcome at six months between “single CS”, “single GA”, “multiple CS”, “multiple GA” and “both” groups (p = 0.089).

Figure 1. Distribution of modified Rankin Scale (mRS) categories according to anesthesia modality (Conscious Sedation = “CS”, General Anesthesia = “GA” and Conscious Sedation as well as General Anesthesia = “both”).

Numbers represent the percentages for each mRS category per group. a) “CS”, “GA” and “both” at six-month follow up. b) “CS”, “GA” and “both” at discharge from hospital.

Table 2. Subgroup analysis of modified Rankin Scale (mRS) at 6 months and at discharge.

		single CS	multiple CS	single GA	multiple GA	both	p-value
mRS 6 months	mRS 0-2	11 (68.8%)	7 (100%)	16 (61.5%)	12 (44.4%)	14 (63.6%)	0.089+
	mRS 3-6	5 (31.3%)	0 (0%)	10 (38.5%)	15 (55.6%)	8 (36.4%)
mRS discharge	mRS 0-2	9 (56.3%)	4 (57.1%)	9 (34.6%)	6 (22.2%)	7 (31.8%)	0.156+
	mRS 3-6	7 (43.8%)	3 (42.9%)	17 (65.4%)	21 (77.8%)	15 (68.2%)

+ Chi-squared test.

Secondary outcomes

Figure 1b shows mRS at discharge from hospital. This analysis displays no significant difference between “CS”, “GA” and “both” groups (p = 0.056). The subgroup analysis for single and multiple interventions separately also revealed no statistical significance (p = 0.156), as listed in Table 2.

The NIHSS time course analysis is presented in Figure 2. ANOVA for repeated measures displayed a significant interaction of “time*treatment” (p = 0.008) and of “time” (p < 0.001). Post-hoc analysis revealed that significant group differences only occur when comparing “GA” to the two other groups. All of these significant differences were between t₁ and t₅, meaning between 24 hours before (first) intervention and discharge from the hospital. At the six-month follow up appointment, “CS”, “GA” and “both” did not differ significantly regarding NIHSS.

Figure 2. Bar graph depicting National Institutes of Health Stroke Scale (NIHSS) scores of the three anesthesia groups.

Patients were treated in conscious sedation (“CS”), general anesthesia (“GA”) or a combination of both (“both”). Each bar represents a different group. Each cluster represents a different time point. Time points are 24 hours before (first) intra-arterial vasodilator intervention (t₁), immediately before (t₂), immediately after (t₃), 24 hours after (last) intervention (t₄), at discharge from the hospital (t₅) and at six-month follow up (t₆). Data are presented as mean + 1 Standard Deviation (SD) as error bars. Significant inter-group differences are highlighted with asterisks. One asterisk represents p < 0.05, two asterisks represent p < 0.01 and three asterisks represent p < 0.001.

Discussion

This retrospective study found no significant differences in functionally independent outcomes (mRS 0-2) six months after aSAH in patients who were treated with intra-arterial vasodilators in CS, GA or a combination of both. While NIHSS was significantly higher in patients undergoing endovascular therapy in GA compared to patients of the “CS” or “both” group in the time window 24 hours before intervention up to discharge from the hospital, this difference was no longer found at six months.

To the best of our knowledge, the effect of anesthesia modality (CS versus GA) on functional outcome has not yet been studied for intra-arterial admission of either nimodipine and/or papaverine in patients with refractory CVS after aSAH. Albeit, there have been several research papers published regarding anesthesia in aneurysm treatment.^21–23 Most articles describe both CS and GA to be generally safe for treatment of unruptured aneurysms or aSAH with no clear recommendation for either one.

Additionally, multiple studies have had similar research questions in relation to the choice of anesthesia during endovascular therapy of acute ischemic stroke. Abou-Chebl et al. (2010) showed in their analysis of the “North American SOLITAIRE Stent-Retriever Acute Stroke” (NASA) registry, that patients treated in GA experienced poorer neurologic outcome at 90 days and higher mortality rates than patients treated in CS.⁸ Berkhemer et al. (2016) reported similar results in a post-hoc analysis of a prospective trial.¹⁵ Correspondingly, a recent analysis of the “Endovascular Therapy Following Imaging Evaluation for Ischemic Stroke 3” (DEFUSE 3) trial by Powers et al. (2019) showed higher rates of functional independence (mRS 0-2) and a lower NIHSS score at 24 hours for patients treated in CS. At discharge they did not find a statistically significant difference in NIHSS scores anymore.¹¹

In contrast, most recent randomized controlled trials found a non-inferiority of GA when compared to CS. Hendén et al. (2017) reported no difference in outcome at three months or NIHSS after 24 hours in their Anesthesia During Stroke (AnStroke) Trial.¹⁶ Schönenberger et al. (2016) and Simonsen et al. (2018) report, that GA produced better 3-month outcomes, with the former even finding this result to be significant.^18,19 Finally, a recent meta-analysis determined no significant difference between GA and CS if only randomized controlled trials were considered.¹⁷

Our results are in line with these recent randomized controlled trials published for endovascular treatment in acute ischemic stroke. Similar to Schönenberger et al. (2016), Hendén et al. (2017), Simonsen et al. (2018) and Kim et al. (2019), we also found no significant difference in functional independency (mRS 0-2) at discharge or six months.^16–19

Similar considerations regarding choice of anesthesia hold true in intra-arterial vasodilator therapy after aSAH and treatment for acute ischemic stroke alike. Disadvantages of GA may be a delay in treatment, hemodynamic changes and complications associated with intubation such as an increased risk for pneumonia.⁸ Disadvantages of CS may be procedural discomfort for patients, more difficult interventions because of patients’ movements, emergency conversion to GA or increased risk for aspiration.⁸ Many of the conceived disadvantages of GA have also been analyzed in prospective studies concerning ischemic stroke treatment. For example, Berkhemer et al. (2016) as well as Hendén et al. (2017) found no treatment delay in the GA group.^15,16 Schönenberger et al. (2016) reported no difference in feasibility, safety and intra-interventional complication between the groups.¹⁸ They did however discover more postprocedural complications after GA such as delayed extubation and pneumonia. Different authors mention the possibility of blood pressure drops and decreased cerebral blood flow as possible additional complications of GA, which could potentially worsen outcomes because of an increase of the ischemic area.¹⁵ Others argue that GA on its own has a neuroprotective effect by lowering the neuronal oxygen need.^24,25 Overall, the similar functional outcome in our study as well as in the above cited studies in ischemic stroke suggest that these factors are of minor relevance.

The major limitations of this study are its retrospective design and the single-center approach. A potential bias could lie in the choice of sedation for treatment. Even before intra-arterial vasodilator treatment, patients who would go on to receive treatments in GA showed significantly higher NIHSS scores when compared to the “CS” and “both” groups. This indicates that patients in clinically and neurologically worse conditions were more likely to be treated in GA. However, this bias was not reflected in our results, as we did not find a significant difference in mRS scores between the anesthesia groups at discharge or at six months and also no significant difference in NIHSS scores at six months. The “GA” group were therefore in an initially worse state but still managed to reach similar outcomes in the long-term clinical course. This suggests to an even greater degree, that GA will not negatively affect long term outcome in these patients.

Furthermore, some subgroups consisted of a small number of patients. The results of this study will have to be replicated by a larger prospective trial in the future.

Conclusion

Our preliminary results indicate that choice of anesthesia method does not negatively affect six-month outcome in aSAH patients who undergo intra-arterial vasodilator treatment for CVS. Treating physicians should therefore decide between CS and GA individually based on patient characteristics and circumstances.

Data availability