Abstract
Background and Purpose:
To identify the specific post-endovascular stroke therapy (EVT) peak systolic blood pressure (SBP) threshold that best discriminates good from bad functional outcomes (a priori hypothesized to be 160 mmHg), we conducted a prospective, multi-center (12 comprehensive stroke centers across the United States), cohort study with a prespecified analysis plan.
Methods:
Consecutive adult patients treated with EVT for an anterior ischemic stroke were enrolled from 11/2017 to 07/2018. All SBP values within 24 hours post-EVT were recorded. Using Youden’s index, the threshold of peak SBP that best discriminated primary outcome of dichotomized 90-day mRS (0-2 vs. 3-6) was identified. Association of this SBP threshold with the outcomes was quantified using multiple logistic regression.
Results:
Among 485 enrolled patients [median age 69 (IQR 57,79) years; 51% females], a peak SBP of 158 mmHg was associated with the largest difference in the dichotomous mRS (absolute risk reduction of 19%). Having a peak SBP >158 mmHg resulted in an increased likelihood of mRS 3-6 (OR 2.24 [1.52-3.29], p<0.01; aOR 1.29, [0.81-2.06], P=0.28, after adjustment for prespecified variables).
Conclusion:
A peak post-EVT SBP of 158 mmHg was prospectively identified to best discriminate good from bad functional outcome in EVT-treated stroke patients. Those with peak SBP>158 had an increased likelihood of having a bad outcome in unadjusted, but not in adjusted analysis. The observed effect size was similar to prior studies. This finding should undergo further testing in a future randomized trial of goal-targeted post-EVT antihypertensive treatment. .
Keywords: Thrombectomy, Blood Pressure, Stroke, Outcome
Subject Terms: High Blood Pressure, Ischemic Stroke
BACKGROUND
Nearly half of the large vessel occlusion strokes (LVO) stroke patients successfully treated with endovascular therapy (EVT) remain disabled at 90 days, and blood pressure (BP) after EVT is a readily modifiable parameter with the potential to improve these outcomes.1-3 With the ever-increasing use of EVT, evidence-based guidelines on the ideal post-EVT BP management are urgently needed. Although the current guidelines recommend maintaining systolic BP (SBP) at ≤180mmHg for 24-48 hours post-EVT, this recommendation is not substantiated with strong evidence, such as randomized trial data.4 Prior retrospective observational studies5-9 have demonstrated an association between a higher peak (i.e., maximum) SBP following EVT and worse functional outcomes, but an inconsistent association between higher-post EVT SBP and hemorrhagic complications. They raise the hypothesis that stricter SBP control with lower targets might further improve patient outcomes.10 In fact, the DAWN trial protocol recommended lower than SBP target (<140 mmHg) after EVT.3 Prior to testing the efficacy of lowering BP below the current standard-of-care in a prospective trial, validated BP targets must be generated. Prior studies5-7 addressing this question are limited by retrospective design, post-hoc choice of BP thresholds, and inclusion of single or only few centers, thus restricting the generalizability of their results. We have found substantial heterogeneity in the management of SBP in post-EVT stroke patients across the US in a prior survey.11 The majority of surveyed institutions targeted lower SBPs after EVT than recommended by guidelines. Additionally, reperfusion status was taken into account by most responders when deciding the individual patient’s post-EVT SBP target. To overcome some of the limitations of the existing literature on the ideal post-EVT BP target, we undertook a prospective, multicenter study.
Therefore, we designed the collaborative, multi-center Blood Pressure after Endovascular Stroke Therapy (BEST) study with the primary aim to determine the specific post-EVT peak SBP threshold that discriminates good vs bad functional outcomes in EVT-treated stroke patients. Based on our prior retrospective findings, our prespecified hypothesis was that this value will be around 160 mmHg. Our long-term goal is to inform the design of a future randomized trial.
METHODS
Study Design
BEST was designed as a prospective, multi-center cohort study with a pre-specified analysis plan submitted to Institutional Review Boards and shared with investigators. Twelve comprehensive stroke centers across the United States participated to enroll patients from November 2017 to September 2018. Prior to initiation, this study was approved by the Institutional Review Board of eleven sites and was deemed exempt by one site and informed consent was waived. The study protocol is provided as a supplement. A REDcap database was utilized for a streamlined and secure online data collection with inbuilt quality checks for data quality and integrity.12 The data that support the findings of this study are available from the corresponding author upon reasonable request.
Patient Selection
This study enrolled consecutive, adult patients who suffered an acute ischemic stroke from an occlusion in the internal carotid artery or the M1 or M2 segments of the middle cerebral artery and were treated with an EVT. Patients with a 1) known pre-stroke modified Rankin score (mRS) > 2, 2) terminal medical diagnoses such as a stage IV cancer, 3) left ventricular assist device, or 4) stroke in the perioperative or inpatient setting were excluded. Each participating site pre-specified their method to identify and include all consecutive patients that met the study’s eligibility criteria.
Study Variables and Covariates
BP variables included consecutively-recorded 24-hour post-EVT systolic (SBP), diastolic (DBP), and mean arterial (MAP) BP readings. Invasive BP readings were preferentially recorded. Peak SBP during this period was the primary study variable. When available, arterial line readings were preferentially recorded. Missing MAP readings were estimated by the (2*DBP+SBP)/3 formula. During the study period, none of the participating institutions had a written standardized protocol to guide BP management specifically in EVT-treated patients.
The following variables were also collected: use of a continuous infusion of intravenous (IV) hypertensive and/or pressor agents in the 24 hours post-EVT; administration of >2 boluses of an IV antihypertensive agent (if no continuous infusion was administered); The Alberta Stroke Program Early Computed Tomography score (ASPECTs) available from the radiology report of the first non-contrast brain computed tomography (CT) or calculated by a trained local study personnel; core and penumbra volumes automatically calculated by the iSchemaView RAPID software from the initial pre-EVT CT perfusion scan; time from last known well to final recanalization assessed as a modified thrombolysis in cerebral ischemia (mTICI) score (or angiogram completion if mTICI was 0); and, presence of any intraparenchymal hemorrhage (ICH; excluding subarachnoid hemorrhage) on either MRI or CT scan within the 72 hours post-EVT.
Outcomes
The primary study outcome was the dichotomized 90-day mRS (0-2 vs. 3-6 representing good vs. bad outcome, respectively). Outcome data was collected as routine care at all institutions via telephone call or in-person encounter with the patient. Secondary outcomes consisted of any ICH, symptomatic ICH (associated with ≥4 points increase in NIHSS from baseline to 24 hours), change in the mRS distribution, and early neurologic recovery (NIHSS 0-1 or 8 points decrease in NIHSS from baseline to 24 hours).13, 14
Statistical Analysis
Power calculations were performed prior to initiating the study using data from our prior retrospective tri-center study.7 A sample size of 340 was calculated to provide 80% power to detect an odds ratio of 1.36. This odds ratio corresponded to an 8% absolute difference in the proportion of patients with a 90-day mRS 3-6 at a two-sided 0.05 significance level between the following two patient groups: 1) those with a certain mean peak SBP and 2) those having a mean peak SBP at least one standard deviation above that in the first group. Accounting for a 15% loss to follow up at 90 days, we initially aimed to enroll 400 patients. The study sample size was increased to 485 at the time of enrollment of the 200th patient, to account for a 25% loss to follow-up rate, which was higher than expected. While we planned a priori to impute any missing 90-day mRS data, we also wanted to insure minimal dilution of treatment effect associated with this imputation.
The threshold of peak SBP that best discriminated good vs. bad outcome was determined using Youden’s index. Association of SBP, dichotomized at the threshold that best discriminated the primary outcome, with primary outcome of binary 90-day mRS (0-2 vs. 3-6) was then quantified using logistic regression models. This quantification was adjusted for a priori selected covariates that included age, baseline NIHSS, glucose, time to recanalization, ASPECT score, recanalization status (successful defined as mTICI 2b-3 vs unsuccessful defined as mTICI 0-2a), SBP on admission, and history of hypertension in a pre-specified model. The association of peak SBP with secondary outcomes were also quantified using logistic regression.
Sensitivity analyses included testing the association of SBP with outcomes according to the recanalization status. An institution-wise subgroup analysis and that based on the type of anesthesia, age, baseline NIHSS, ASPECT score, IV thrombolytic administration, administration of IV anti-hypertensive agent, and history of hypertension to evaluate heterogenous effects of BP on outcomes were also conducted to account for heterogeneity in effect. The association of clinical outcomes with diastolic (DBP), mean (MAP), variability (quantified using mean to peak ratio) of SBP, and the proportion of peak SBP values above the discriminating threshold were explored. Missing 90-day outcome data was imputed by multiple imputations using the predictive mean matching method. A fitted generalized additive model was generated to visualize any time-dependent BP trends according to the 90-day mRS and study site. Lastly, propensity score matching (using both exact and nearest neighbor methods) was attempted to balance covariate differences in patients dichotomized at the discovered SBP threshold.
All analyses and graphical depictions were generated using SAS version 9.4 (SAS Institute, Cary, North Carolina, USA) and R version 3.5 (R Foundation for Statistical Computing, Vienna, Austria). Statistical significance α was set at 0.05 and all p-values are 2‐sided. All effect sizes are reported with 95% confidence intervals in addition to the p-values.15
RESULTS
Description of Patients and BP values
From November 2017 to September 2018, 485 patients were enrolled at 12 comprehensive stroke centers across the United States. Demographics and baseline characteristics are outlined in Table 1. Baseline perfusion scan was performed on 221 (46%) of patients. A total of 17,850 BP readings were recorded for the 485 patients. Of the recorded BP variables, the following readings were excluded from the analysis as likely spurious values: SBP>300 mmHg (n=1), DBP <20 mmHg (n=1), DBP ≥ SBP value (n=1). For each patient enrolled in the study, a median of 34 [interquartile range (IQR): 29-42] BP readings were recorded during the 24 hours post-EVT.
Table 1.
Demographics, Comorbidities, Stroke Characteristics, Stroke treatment, and Outcomes
| Age (median with IQR) | 69 (57-79) |
| Female Sex, n (%) | 249 (51%) |
| Comorbidities, n (%) | |
| Hypertension | 364 (75%) |
| Diabetes | 139 (29%) |
| Atrial Fibrillation | 176 (36%) |
| Current Smoking | 114 (24%) |
| Antiplatelet Drug Use | 165 (34%) |
| Anticoagulant Use | 86 (18%) |
| Baseline Lab Values | |
| Glucose mg/dL, (Mean, SD) | 139 (58) |
| Platelet (Mean, SD) | 230 (75) |
| INR (Mean, SD) | 1.14 (0.30) [n=480] |
| Stroke Characteristics | |
| Baseline NIHSS (median with IQR) | 16 (11-20) |
| ASPECT score (median with IQR) | 8 (7-10) [n=484] |
| Penumbra Volume (median with IQR) | 114.5 (67.4, 162.0) [n=176] |
| Core Volume (median with IQR) | 10 (0-30) [n=177] |
| ICA Occlusion (%) | 129 (27%) |
| M1 Occlusion (%) | 309 (64%) |
| M2 Occlusion (%) | 121 (25%) |
| Left Circulation (%) | 262 (54%) |
| Stroke Treatment | |
| Intravenous Thrombolysis, n (%) | 231 (48%) |
| Time from last known well to recanalization or angiogram completion (median with IQR), min | 298 (189-610) [n=483] |
| General Anesthesia, n (%) | 190 (39%) |
| Conscious Sedation, n (%) | 294 (61%) |
| mTICI 0, n (%) | 25 (5%) |
| mTICI 1, n (%) | 10 (2%) |
| mTICI 2a, n (%) | 20 (4%) |
| mTICI 2b, n (%) | 178 (37%) |
| mTICI 3, n (%) | 252 (52%) |
| Antihypertensive Drip, n (%)* | 192 (40%) |
| >2 PRN doses of IV Anti-hypertensive Agent, n (%)* | 57 (19%) |
| Vasopressor IV drip, n (%)* | 87 (18%) |
| Outcomes, n (%) | |
| mRS 0 | 57 (12%) |
| mRS 1 | 68 (14%) |
| mRS 2 | 61 (13%) |
| mRS 3 | 77 (16%) |
| mRS 4 | 60 (12%) |
| mRS 5 | 44 (9%) |
| mRS 6 | 79 (16%) |
| Unknown | 39 (8%) |
| Any ICH | 117 (24%) |
| Symptomatic ICH | 18 (4%) |
| Early Neurologic Recovery | 202 (42%) |
Medication used within 24 hours post-endovascular stroke treatment
ASPECTS: Alberta Stroke Program Early CT Score; CTP: CT Perfusion; ICA: Internal Carotid Artery; ICH: Intracerebral Hemorrhage; INR: International Normalized Ratio; IQR: Interquartile Range; IV: Intravenous; M1: First segment of Middle Cerebral Artery; M2: Second segment of Middle Cerebral Artery; mRS: Modified Rankin Score; mTICI: Modified Thrombolysis in Cerebral Ischemia score; NIHSS: National Institute of Health Stroke Scale
Determination of the outcome-discriminating peak SBP value
Ninety-day mRS values were available for 446 (92%) of 485 patients enrolled. Among these 446 patients, 186 (42%) patients had a good outcome (mRS of 0-2). The peak SBP value during the 24 hours after EVT that best discriminated good from a bad outcome was 158 mmHg (52% sensitivity, 68% specificity, Area Under the Curve 0.61, 95% confidence interval 0.56 to 0.66, p<0.001; Receiver Operating Characteristics curve is depicted in Supplemental Figure I). A peak SBP of >158 mmHg resulted in an unadjusted odds ratio (OR) of 2.24 [1.52-3.29], p<0.01 and adjusted OR of 1.29, [0.81- 2.06], p=0.28 for having a bad outcome (Table 2). A total of 298 patients had a peak SBP >158 mmHg, of which, 90-day outcome was available for 267. Of these, 176 (66%) had mRS 3-6 as compared to 85 (47%) of 181 patients (with available 90-day outcome) with peak SBP ≤158 mmHg. Patients with peak SBP ≤158 mmHg were younger and had lower NIHSS, lower baseline glucose level, quicker time from onset to reperfusion, higher rates of successful recanalization, and a higher rate of IV thrombolysis administration (Supplemental Table I). Higher peak SBP on a continuous scale associated with bad outcome in unadjusted (OR 1.02 [1.01-1.03], p<0.01) but not in adjusted analysis (OR 1.00 [0.99-1.01], p=0.79; Table 2). After imputing of the 90-day mRS for the 8% of patients lost-to-follow-up, SBP threshold that best dichotomized primary outcome (158 mmHg) and its association with outcomes remained unchanged (Table 2).
Table 2.
Association of peak systolic blood pressure with 90-day modified Rankin score
| Outcome | OR | P-value | Adjusted OR | P-value |
|---|---|---|---|---|
| mRS 3-6 (SBP dichotomized >158mmHg) | 2.24 (1.52-3.29) | <0.01 | 1.29 (0.81-2.06) | 0.28 |
| mRS 3-6 (SBP as continuous) | 1.02 (1.01-1.03) | <0.01 | 1.00 (0.99-1.01) | 0.79 |
| mRS Shift†(SBP dichotomized) | 2.27 (1.62-3.19) | <0.01 | 1.25 (0.87-1.81) | 0.22 |
| After multiple imputation of missing mRS | ||||
| mRS 3-6 (SBP dichotomized >158mmHg) | 2.24 (1.54-3.28) | <0.01 | 1.24 (0.78-1.95) | 0.35 |
| mRS 3-6 (SBP as continuous) | 1.02 (1.01-1.02) | <0.01 | 1.00 (0.99-1.01) | 0.82 |
| mRS Shift† (SBP dichotomized) | 2.27 (1.63-3.15) | <0.01 | 1.22 (0.86-1.74) | 0.27 |
Common odds ratio from the proportional odds model is shown
mRS: Modified Rankin Score; SBP: Systolic Blood Pressure
Secondary Outcomes
Having a peak SBP>158 mmHg resulted in unadjusted common OR of 2.27 [1.62-3.19], p<0.01 (Figure 1 & Table 2) and adjusted common OR of 1.25 [0.87-1.81], p=0.22 (Supplemental Figure II & Table 2), of having unfavorable shift in 90-day mRS distribution. Peak SBP >158 mmHg was also associated with lower early neurologic recovery in unadjusted (OR 0.68 [0.47-0.99], p=0.04) but not in adjusted (OR 0.90 [0.58-1.39], p=0.63). Of the 485 patients, 117 (24%) had a post-EVT ICH, and of those, 18 (15%) had a symptomatic ICH. There was insufficient evidence of an association between peak SBP (continuous or dichotomized at the cutoff of 158 mmHg) and ICH (OR 0.91 [0.60-1.39], p=0.67 for any ICH and OR 1.78 [0.63-5.09], p=0.28 for symptomatic ICH). Secondary outcomes are tabulated in Table 3.
Figure 1. Distributions of 90-day Modified Rankin Scores.
Distribution in patients with a peak systolic blood pressure of ≤158 mmHg and >158 mmHg in the initial 24 hours after endovascular stroke therapy for an anterior large cerebral vessel occlusion.
Table 3.
Secondary Outcomes
| Outcome | OR | P-value | Adjusted OR | P-value |
|---|---|---|---|---|
| ENR (SBP dichotomized >158 mmHg) |
0.68 (0.47-0.99) | 0.04 | 0.90 (0.58-1.39) | 0.63 |
| Any ICH (SBP dichotomized >158 mmHg) |
0.91 (0.60-1.39) | 0.67 | 0.78 (0.48-1.24) | 0.29 |
| Symptomatic ICH (SBP dichotomized >158 mmHg) |
1.78 (0.63-5.09) | 0.28 | 2.02 (0.60-6.83) | 0.26 |
ENR: Early Neurologic Recovery; ICH: Intracerebral Hemorrhage; SBP Systolic Blood Pressure
Sensitivity Analysis
A total of 430 (89%) patients achieved successful recanalization (mTICI 2b-3). Patients with mTICI 2b-3 had a lower mean peak SBP (164±24 mmHg) compared to those with mTICI 0-2a (172 ±22 mmHg, p=0.02; Supplemental Figure III). There was no evidence of a differential effect of high peak SBP on bad outcome by mTICI status (interaction p=0.96; Figure 2). The Receiver Operating Curve test characteristics for association of peak SBP >158 with 90-day mRS 3-6 in mTICI 2b-3 group was sensitivity 52%, specificity 66%, and in mTICI 0-2a group was sensitivity 40%, specificity 76%. Values over this threshold resulted in an OR of 2.15 [1.43-3.23], p<0.01 in mTICI 2b-3 group and 2.07 [0.48-8.83], p=0.33 in mTICI 0-2a group for having a bad outcome (adjusted OR 1.31 [0.81-2.12], p=0.28 for mTICI 2b-3). Other outcomes according to mTICI group are listed in the Supplemental Table-II. Notably, in the unadjusted analyses, higher peak SBP on a continuous scale associated with bad outcome (OR 1.02 [1.01-1.03], p<0.01) and dichotomized at 158 mmHg associated with shift in mRS towards bad outcomes (OR 2.10 [1.47-3.00], p<0.01). Adjustment for all prespecified covariates (except mTICI) resulted in p-values that did not meet our prespecified threshold for significance.
Figure 2. Subgroup analysis for heterogeneity of effect of peak systolic blood pressure >158 mmHg on 90-day Modified Rankin Score 3-6.
The odds of having a modified Rankin Score of 3-6 for each subgroup is depicted with a square and a line spanning it representing a 95% confidence interval on the forest plot on the right.
Subgroup Analyses
No heterogeneity in effect of peak SBP >158 mmHg (Figure 2) or higher peak SBP on a continuous scale (Supplemental Figure IV) on worse outcome was observed according to subgroups of patients based on the type of anesthesia, age, baseline NIHSS, ASPECT score, IV thrombolytic administration, administration of IV anti-hypertensive agent, and history of hypertension, after correction for multiplicity. Similarly, in an institution-wise subgroup analysis, no site-specific heterogeneity of effect of peak SBP >158 mmHg on worse outcome was found (Supplemental Table III). Given the significantly different characteristics of patients with and without a peak SBP >158 mmHg, we attempted to analyze propensity-matched patients between these groups but were unsuccessful in generating balanced groups.
Exploratory Analyses
A generalized additive model was fitted to the 24-hour BP values to visualize any time-dependent trends. This fit revealed that generally there was a decline in all BP parameters within the first few hours after an EVT followed by a plateau (Supplemental Figure V). Patients with the worse outcome generally had higher SBP values than those with a good outcome (Figure 3A). SBP trend of patients with mRS 3-4 was similar to those of mRS 0-2 initially but rose towards that of mRS 5-6 patients later in the 24 hours (Figure 3B). Institution-wise Time-dependent SBP curves for each institution showed similar general trend of declining BP (Supplemental Figure V). Patients with bad outcomes had a lower DBP on average compared to those with a good outcome, without any difference in the MAP (Supplemental Figure VI).
Figure 3. Time-dependent changes in the systolic blood pressure according to the 90-day patient outcome.
Lines with a ribbon represent a generalized additive model (mean-like) fitted to all (17,850) SBP values recorded over 24 hours with 95% confidence interval.
Each percent increase in the proportion of SBP values >158 resulted in a 2% increase in the odds of mRS 3-6 (OR 1.02 [1.01-1.03], p=0.004). DBP and MAP did not associate with any primary or secondary outcomes (Supplemental Table-IV). Higher peak to mean ratio of SBP trended with bad outcome (OR 4.70 [0.91-24.26], p=0.06).
The peak 24-hour post-EVT SBP that best discriminated 90-day mRS 0-1 vs. 2-6 remained 158 mmHg (55% sensitivity, 65% specificity, Area Under the Curve 0.62, 95% confidence interval 0.57 to 0.66, p<0.001). A peak SBP of >158 mmHg resulted in an unadjusted odds ratio (OR) of 2.33, [1.53-3.55], p <0.01 and adjusted OR of 1.34, [0.81-2.22], p=0.25 for having a 90-day mRS 2-6.
DISCUSSION
In this multi-center, prospective observational study of acute ischemic stroke patients with LVO, we upheld our prespecified hypothesis that a peak SBP around 160 mmHg (determined here to be 158 mmHg) in the 24 hours post-EVT best dichotomizes good vs bad functional outcomes. Patients with a peak post-EVT SBP >158 mmHg have a higher likelihood of achieving a bad functional outcome, defined as 90-day mRS 3-6, in unadjusted analysis. Although the difference in primary outcomes based on this SBP threshold did not reach statistical significance after adjustment for the prespecified variables, the direction and the size of the effect are consistent with prior observations.5-7, 15 Secondary, subgroup, and sensitivity analyses also revealed similar direction and size for the effect of higher SBP as in prior studies. By capturing the heterogenous post-EVT SBP management practices across institutions, our study provides generalizable results which are not limited to individual institutional practices.
Although prior retrospective observational studies5-7 have found an independent association between higher peak SBP after EVT and worse outcomes, an underlying causal relationship of this association is yet to be established. Preclinical experiments suggest the potential for causality. In murine models of transient LVO, cerebral arteries show impaired autoregulation following recanalization.16, 17 This may render previously oligemic brain tissue vulnerable to hyperperfusion injury.18 In fact, in a randomized trial involving murine models, BP lowering after recanalization was associated with lower infract volumes and potentially better clinical outcomes.19 Conversely, increased BP may merely be a marker of underlying higher infarct volumes, worse baseline control of hypertension, or inadequate collateral circulation.20, 21 In patients with larger infarcts, for example, we found a non-significant interaction of effect of SBP on outcomes according to baseline NIHSS. This may suggest that patients with larger strokes have higher SBP, a manifestation of a compensatory response to rescue the oligemic penumbral tissue. Similarly, patients with worse collaterals may require a more robust pressor response to brain ischemia resulting in larger increase in BP. Although we accounted for baseline history of hypertension, the controlled vs uncontrolled nature of the baseline hypertension may play a role in determination of effect of post-EVT BP on outcomes. Those with poorly controlled hypertension demonstrate a rightward shift of their cerebral autoregulatory curve22, 23 and may require higher SBP in order to maintain normal cerebral perfusion even after recanalization.
A randomized trial to determine if the observed association in our study is causative and modifiable is imperative and urgently needed.4, 10, 24 This large, targeted, prospective cohort study, provides a data-driven and potentially actionable threshold that may guide the design of SBP targeting trials in EVT-treated stroke patients. If modifiable, the unadjusted probability of 66% bad outcome with peak SBP>158 mm Hg compared to 47% with peak SBP ≤158 mmHg, resulting in 19% absolute risk reduction and number needed to treat of 6 to prevent one more poor outcome
Expectedly, other BP parameters, including minimum SBP, were not associated with outcomes in this study. A “U” shape relationship between post-stroke BP and outcomes has been observed in studies that largely included patients without LVO.25, 26 Successful recanalization with resultant perfusion due to EVT for LVO strokes may diminish the detrimental effect of lower SBP on outcomes. Interestingly, among those with a post-EVT peak SBP >158 mmHg, values above 158 mmHg were proportionally associated with worse outcome, suggesting a “dose effect.” Wide excursions, defined by a higher peak to mean ratio, also associated bad outcomes. In other words, sustained high SBP or repeated ‘spikes’ both had a negative effect on outcomes.
In our exploratory, time-dependent BP trend analyses, we reveal interesting patterns of SBP trends in EVT-treated patients in the 24 hours post-EVT. In the setting of an LVO, there appears to be a physiological increase in BP to attempt to maintain brain perfusion. After successful recanalization with an EVT, a decline in SBP is seen in most patients. The pace and depth of decline in SBP throughout the 24 hours post-EVT appears proportional to the functional outcome of the patient.
The results of our study should be interpreted with caution, given the limitations inherent in this nonrandomized design. Confounding variables, such as recanalization status, cannot be adequately addressed, although the overall results did not differ in the subgroup analysis restricted to those with successful recanalization. Our results in the unsuccessful recanalization subgroup are limited by power (11% of all patients, n=55). Similarly, lack of sufficient evidence of association between higher post-EVT peak SBP and sICH may be due to limited number of patients (4% of all patients with sICH, n=18). Further, in this cohort, patients with a peak SBP>158 mmHg had different baseline characteristics, particularly those known to confer a worse outcome.27-31 This is underscored by our inability to generate balanced groups with propensity matching. Although our prespecified analysis plan included adjustment for majority of these variables, additional unmeasured confounding may also be present in a nonrandomized design. For example, we did not collect the collateral status and final infarct volume on all patients which may influence individual post-EVT SBP targets. Furthermore, this nonrandomized study cannot address the causal effect of higher SBP with worse outcome. Nonetheless, by performing our study prospectively in consecutive patients at multiple centers encompassing diverse clinical practices of post-EVT BP targets, we minimized the intrinsic bias of the past retrospective studies.
CONCLUSION
This study validates that a peak SBP of 158 mmHg during the first 24 hours post-EVT best dichotomizes good vs bad outcomes in EVT-treated patients. Having a peak SBP >158 mmHg may increase the likelihood of bad outcomes. We suggest that a future randomized clinical trial of goal-targeted antihypertensive treatment after EVT should consider a peak SBP target of 158 mmHg for testing of safety and efficacy.
Supplementary Material
Acknowledgments
SOURCES OF FUNDING
Supported by the Society of Vascular and Interventional Neurology, University of Cincinnati Gardner Neuroscience Institute, NIH/ NINDS (U01 NS086872 and U10 NS086512), and National Center for Advancing Translational Sciences/ Clinical and Translational Science Awards Program (UL1 TR002243).
DISCLOSURES
Dr. E. Mistry reports research grants from the Society of Vascular and Interventional Neurology, University of Cincinnati Gardner Neuroscience Institute, and Vanderbilt Faculty Research Scholar Program. Dr. Sucharew reports research grants from the University of Cincinnati Gardner Neuroscience Institute. Dr. Siegler reports salary support from NIH/NINDS StrokeNet (U10 NS086474). Dr. James reports holding shares in Remedy Pharmaceuticals, Inc. Dr. Chitale reports research grants from Medtronic and Cerenovus. Dr. Volpi reports personal fees from Amgen and personal fees from Johnson and Johnson. Dr. Kasner reports research grants and consulting fees from Medtronic. Dr. Jagadeesan reports consulting fees from Microvention Inc. Dr. Jayaraman reports honoraria from Medtronic. Dr. Khatri reports research grant from Cerenovus. Remaining authors have no disclosures.
REFERENCES
- 1.Albers GW, Marks MP, Kemp S, Christensen S, Tsai JP, Ortega-Gutierrez S, et al. Thrombectomy for stroke at 6 to 16 hours with selection by perfusion imaging. N Engl J Med. 2018;378:708–718 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Goyal M, Menon BK, van Zwam WH, Dippel DW, Mitchell PJ, Demchuk AM, et al. Endovascular thrombectomy after large-vessel ischaemic stroke: A meta-analysis of individual patient data from five randomised trials. Lancet (London, England). 2016;387:1723–1731 [DOI] [PubMed] [Google Scholar]
- 3.Nogueira RG, Jadhav AP, Haussen DC, Bonafe A, Budzik RF, Bhuva P, et al. Thrombectomy 6 to 24 hours after stroke with a mismatch between deficit and infarct. N Engl J Med. 2018;378:11–21 [DOI] [PubMed] [Google Scholar]
- 4.Powers WJ, Rabinstein AA, Ackerson T, Adeoye OM, Bambakidis NC, Becker K, et al. 2018 guidelines for the early management of patients with acute ischemic stroke: A guideline for healthcare professionals from the american heart association/american stroke association. Stroke. 2018;49:e46–e110 [DOI] [PubMed] [Google Scholar]
- 5.Goyal N, Tsivgoulis G, Pandhi A, Chang JJ, Dillard K, Ishfaq MF, et al. Blood pressure levels post mechanical thrombectomy and outcomes in large vessel occlusion strokes. Neurology. 2017;89:540–547 [DOI] [PubMed] [Google Scholar]
- 6.Maier IL, Tsogkas I, Behme D, Bahr M, Knauth M, Psychogios MN, et al. High systolic blood pressure after successful endovascular treatment affects early functional outcome in acute ischemic stroke. Cerebrovasc Dis. 2018;45:18–25 [DOI] [PubMed] [Google Scholar]
- 7.Mistry EA, Mistry AM, Nakawah MO, Khattar NK, Fortuny EM, Cruz AS, et al. Systolic blood pressure within 24 hours after thrombectomy for acute ischemic stroke correlates with outcome. J Am Heart Assoc 2017;6:e006167. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Anadani M, Orabi MY, Alawieh A, Goyal N, Alexandrov AV, Petersen N, et al. Blood pressure and outcome after mechanical thrombectomy with successful revascularization. Stroke. 2019:STROKEAHA118024687 [DOI] [PubMed] [Google Scholar]
- 9.Goyal N, Tsivgoulis G, Pandhi A, Dillard K, Alsbrook D, Chang JJ, et al. Blood pressure levels post mechanical thrombectomy and outcomes in non-recanalized large vessel occlusion patients. J Neurointerv Surg. 2018;10:925–931 [DOI] [PubMed] [Google Scholar]
- 10.Jadhav AP, Molyneaux BJ, Hill MD, Jovin TG. Care of the post-thrombectomy patient. Stroke. 2018;49:2801–2807 [DOI] [PubMed] [Google Scholar]
- 11.Mistry EA, Mayer SA, Khatri P. Blood pressure management after mechanical thrombectomy for acute ischemic stroke: A survey of the strokenet sites. Journal of Stroke and Cerebrovascular Diseases. 2018;27:2474–2478 [DOI] [PubMed] [Google Scholar]
- 12.Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research electronic data capture (redcap)--a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform. 2009;42:377–381 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Campbell BC, Mitchell PJ, Kleinig TJ, Dewey HM, Churilov L, Yassi N, et al. Endovascular therapy for ischemic stroke with perfusion-imaging selection. N Engl J Med. 2015;372:1009–1018 [DOI] [PubMed] [Google Scholar]
- 14.Heit JJ, Mlynash M, Kemp SM, Lansberg MG, Christensen S, Marks MP, et al. Rapid neurologic improvement predicts favorable outcome 90 days after thrombectomy in the defuse 3 study. Stroke. 2019:STROKEAHA119024928 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Amrhein V, Greenland S, McShane B. Scientists rise up against statistical significance. Nature. 2019;567:305–307 [DOI] [PubMed] [Google Scholar]
- 16.Cipolla MJ, Curry AB. Middle cerebral artery function after stroke: The threshold duration of reperfusion for myogenic activity. Stroke. 2002;33:2094–2099 [DOI] [PubMed] [Google Scholar]
- 17.Cipolla MJ, McCall AL, Lessov N, Porter JM. Reperfusion decreases myogenic reactivity and alters middle cerebral artery function after focal cerebral ischemia in rats. Stroke. 1997;28:176–180 [DOI] [PubMed] [Google Scholar]
- 18.Nour M, Scalzo F, Liebeskind DS. Ischemia-reperfusion injury in stroke. Interv Neurol. 2013;1:185–199 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Elewa HF, Kozak A, Johnson MH, Ergul A, Fagan SC. Blood pressure lowering after experimental cerebral ischemia provides neurovascular protection. J Hypertens. 2007;25:855–859 [DOI] [PubMed] [Google Scholar]
- 20.Fischer U, Mattle HP. Blood pressure in acute stroke: Still no answer for management. Stroke. 2017;48:1717–1719 [DOI] [PubMed] [Google Scholar]
- 21.Wartenberg KE, Mayer SA. Determining the optimal target blood pressure after thrombectomy: High or low? Neurology. 2017;89:528–529 [DOI] [PubMed] [Google Scholar]
- 22.Beason-Held LL, Moghekar A, Zonderman AB, Kraut MA, Resnick SM. Longitudinal changes in cerebral blood flow in the older hypertensive brain. Stroke. 2007;38:1766–1773 [DOI] [PubMed] [Google Scholar]
- 23.Tzeng YC, Ainslie PN. Blood pressure regulation ix: Cerebral autoregulation under blood pressure challenges. Eur J Appl Physiol. 2014;114:545–559 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Liebeskind DS, Derdeyn CP, Wechsler LR, * SXC. Stair x. Stroke. 2018;49:2241–2247 [DOI] [PubMed] [Google Scholar]
- 25.Castillo J, Leira R, Garcia MM, Serena J, Blanco M, Davalos A. Blood pressure decrease during the acute phase of ischemic stroke is associated with brain injury and poor stroke outcome. Stroke. 2004;35:520–526 [DOI] [PubMed] [Google Scholar]
- 26.Leonardi-Bee J, Bath PM, Phillips SJ, Sandercock PA, Group ISTC. Blood pressure and clinical outcomes in the international stroke trial. Stroke. 2002;33:1315–1320 [DOI] [PubMed] [Google Scholar]
- 27.Adams HP Jr., Davis PH, Leira EC, Chang KC, Bendixen BH, Clarke WR, et al. Baseline nih stroke scale score strongly predicts outcome after stroke: A report of the trial of org 10172 in acute stroke treatment (toast). Neurology. 1999;53:126–131 [DOI] [PubMed] [Google Scholar]
- 28.Rha JH, Saver JL. The impact of recanalization on ischemic stroke outcome: A meta-analysis. Stroke. 2007;38:967–973 [DOI] [PubMed] [Google Scholar]
- 29.Saver JL, Goyal M, van der Lugt A, Menon BK, Majoie CB, Dippel DW, et al. Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: A meta-analysis. JAMA. 2016;316:1279–1288 [DOI] [PubMed] [Google Scholar]
- 30.Smith WS, Lev MH, English JD, Camargo EC, Chou M, Johnston SC, et al. Significance of large vessel intracranial occlusion causing acute ischemic stroke and tia. Stroke. 2009;40:3834–3840 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Wada S, Yoshimura S, Inoue M, Matsuki T, Arihiro S, Koga M, et al. Outcome prediction in acute stroke patients by continuous glucose monitoring. Journal of the American Heart Association. 2018;7:e008744. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.