Sishir Mannava, MD
@sishmannMD
Blood pressure (BP) control after endovascular therapy (EVT) for large vessel occlusion is an oft-debated topic. Many stroke providers, although aware of the potential harm of reperfusion injury, do not know how aggressive to be with BP lowering. The risk with acute lowering of BP is also the risk of hypoperfusion and placing the penumbra at risk of worsening ischemic damage. The authors of this work conducted a post-hoc analysis of the BP TARGET trial (recently published and showing neutral results from intensive BP lowering in patients with successful reperfusion after EVT) to assess the association between change in systolic blood pressure at different time intervals with clinical and radiographic outcomes.
This was a secondary observational post-hoc analysis of the original BP TARGET trial, which enrolled patients with anterior circulation large vessel occlusion stroke and had acute ischemic stroke with successful reperfusion with EVT but had SBP > 130 mmHg at end of procedure (considered the baseline SBP). Patients in the BP TARGET trial were randomized 1:1 to intensive SBP treatment to be achieved within 1 hour of randomization (target 100-129 mmHg) or standard SBP treatment (130-185mmHg). BP was measured every 15 minutes for 2 hours after randomization, then every 30 minutes for 6 hours and then every hour for the remaining 16 hours. The authors defined ΔSBP as baseline SBP minus the mean of achieved SBP levels during 3 periods post-reperfusion: 15 to 60 minutes (ΔSBP15–60M), 1 to 6 hours (ΔSBP1–6H), and 6 to 24 hours (ΔSBP6–24H).
Interestingly, the authors chose to study poor outcome as the primary outcome measure, defined as modified Rankin score (mRS) 3-6 at 90 days (or an increase in pre-stroke mRS for patients with baseline mRS≥3). 90 day mortality, any intraparenchymal hemorrhage (IPH) at 24-36 hours, symptomatic ICH (sICH), and neurological deterioration were reported as safety outcome measures.Of 267 total patients, 130 were intensive BP lowering group and 137 were standard BP group. The authors found through logistical regression models that poor outcome occurred in 52.3 % and 53.2% in the intensive and standard groups, respectively; however, compared to patients with favorable outcome, patients with poor outcome had lower ΔSBP at all time intervals. After adjusting for potential confounders including baseline SBP, both ΔSBP1-6H and ΔSBP6-24H were associated with lower odds of poor outcome (aOR per 5 mmHg SBP reduction, 0.89; 95% 0.81 to 0.99, and aOR per 5 mmHg SBP reduction 0.82; 95% CI 0.73 to 0.92, respectively), and patients with intraparenchymal hemorrhage had lower ΔSBP at all time intervals. ΔSBP15-60M was associated with lower odds of any intraparenchymal hemorrhage (aOR per 5 mmHg SBP reduction, 0.91; 95% CI 0.83 to 0.99). ΔSBP was not associated with mortality or neurological deterioration at any time interval.
Ultimately, the authors sought to establish a clearer relationship between using ΔSBP and outcomes using restricted cubic spline regression models, and found a linear association between ΔSBP after successful reperfusion and two other factors: the risk of poor outcome and intraparenchymal hemorrhage. Interestingly, there was a potential deleterious effect of significant alteration of BP after successful reperfusion giving the U-shaped relationship with mortality. The authors concluded that per their analysis, SBP reduction was associated with lower risk of poor outcome and intraparenchymal hemorrhage. This effect of ∆SBP on poor outcome was consistent in subgroup analyses based on baseline hypertension, baseline SBP, and reperfusion status. Although these results conflict with the original BP TARGET results, the authors argue that the original trial was limited by too small a difference between achieved mean SBP between intensive and standard treatment arms (128 vs 138 mmHg, respectively). Optimal BP targets for patients after successful and unsuccessful EVT for large vessel occlusion stroke have yet to be determined, but this study adds interesting findings to the currently available data.
