After comprehensively searching multiple databases and eliminating duplicate studies, we reviewed 1400 published studies based on their titles and abstracts. Among these, we examined entire texts of 79 studies and excluded those who lacked blood pressure target data. Eventually, 17 studies10,12,13,16,21–33 were included in this meta-analysis (Fig. 1), encompassing 5825 EVT patients. eTable 2 (supplement 1) presents a summary of the key characteristics of the included studies.
Risk of Bias
The RCTs and observational studies’ quality evaluation were illustrated in eFigure 1 and eTable 3 (supplement 1). The results of Begg’s test (p = 0.50) and Egger’s test(p = 0.12) show that p > 0.05, so there is no publication bias (eFigure 2, supplement 1). Meta-regression analysis showed no signifcant correlations between different blood pressure, country and year of publication. But with statistically significant differences were observed in different study type (p < 0.01) and sample size (p = 0.04), eFigure 3,4 (supplement 1). Funnel plot was used for assessing publication bias (eFigure 5, supplement 1). The leave-one-out method demonstrated that the overall effect size remained moderate and was not significantly influenced by any individual study, indicating the robustness of the findings (eFigure 6, supplement 1).
90 days functional independence
16 studies with 5549 patients provided data on 90 days functional independence (mRS score = 0–2). A meta-analysis using a random-effects model indicated statistically significant differences in blood pressure management (OR = 1.58, 95% CI [1.17, 2.13]), with obvious heterogeneity (I2 = 83.20%, p < 0.01), Fig. 2. Given the high heterogeneity and varying specific blood pressure values across the included studies, subgroup analyses were performed. Subgroup analyses revealed statistically significant differences in SBP < 140 mmHg (OR = 1.62, 95% CI [1.17, 2.24];I2 = 74.20%, p < 0.01), with reduced heterogeneity noted, (eFigure 7, supplement 1). The lowest heterogeneity was observed when SBP was < 160 mmHg (OR = 2.23, 95% CI [1.65, 3.03]༛I2 = 0%, p = 0.85). Due to different study types, subgroup analysis revealed that observational studies favored intensive blood pressure management (OR = 1.94, 95% CI [1.66, 2.28]). In contrast, results from randomized controlled trials (RCTs) showed the opposite effect (OR = 0.66, 95% CI [0.51,0.87]). Heterogeneity decreased in both subgroup classifications (I2 = 18.20%, p = 0.27 and I2 = 29.30%, p = 0.24), (eFigure 8, supplement 1).
90-day mortality
The results of 90-day mortality data were obtained from 15 studies encompassing 5051 patients. As seen in Fig. 3, Statistically significant difference in 90-day mortality was observed across different blood pressure management (OR = 0.77, 95% CI [0.59, 1.00]; I 2 = 52.60%, p = 0.01) in a random-effect model analysis. The subgroup analysis exhibited there was no statistically significant difference in different blood pressure values, (eFigure 9, supplement 1). In subgroup analysis based on different study types, heterogeneity (observational studies, I 2 = 34.30%, p = 0.12 VS. RCTs, I 2 = 0%, p = 0.85) was reduced. Observational studies indicated that more intensive blood pressure management was beneficial for the 90-day survival rate of patients (OR = 0.63, 95% CI [0.48, 0.82]), (eFigure 10, supplement 1).
7-day mortality
The results of 7-day mortality data were obtained from 4 studies encompassing 2084 patients. No statistically significant difference in 7-day mortality was observed across different blood pressure management (OR = 0.88, 95% CI [0.49, 1.58]; I 2 = 67.10%, p = 0.03) in a random-effect model analysis, Fig. 4. The subgroup with RCTs’ study type (OR = 1.36, 95% CI [0.87, 2.13]; I 2 = 0%, p = 0.68) exhibited the lowest heterogeneity, (eFigure 11, supplement 1).
Symptomatic intracranial hemorrhage
The results of sICH data were obtained from 13 studies encompassing 4284 patients. As seen in Fig. 5, Statistically significant difference in sICH was observed across different blood pressure management (OR = 0.74, 95% CI [0.58, 0.95]; I 2 = 33.70%, p = 0.11) in a fixed-effect model analysis. The subgroup with SBP < 140 mmHg (OR = 0.69, 95% CI [0.50, 0.97]; I 2 = 27.50%, p = 0.20) exhibited that more intensive blood pressure management was beneficial for the sICH, (eFigure 12, supplement 1). In subgroup analysis based on different study types, heterogeneity (observational studies, I 2 = 24.00%, p = 0.22 VS. RCTs, I 2 = 0%, p = 0.78) was reduced, (eFigure 13, supplement 1).
Complications
The results of complications data were obtained from 6 studies encompassing 1602 patients. As seen in Fig. 6, no statistically significant difference in complications was observed across different blood pressure management (OR = 1.32, 95% CI [0.50, 3.50]; I 2 = 70.80%, p < 0.01) in a random-effect model analysis. The subgroup with RCTs’ study type (OR = 3.42, 95% CI [1.43, 8.18]; I 2 = 0%, p = 0.35) exhibited that less intensive blood pressure management was beneficial for the complications, (eFigure 14, supplement 1).
Hemicraniectomy
The results of hemicraniectomy data were obtained from 4 studies encompassing 1415 patients. As seen in Fig. 7, statistically significant difference in hemicraniectomy was observed across different blood pressure management (OR = 0.42, 95% CI [0.22, 0.81]; I 2 = 0%, p = 0.47) in a fixed-effect model analysis. The subgroup with SBP < 140 mmHg (OR = 0.34, 95% CI [0.17, 0.70]; I 2 = 0%, p = 0.77) exhibited that more intensive blood pressure management was beneficial for the hemicraniectomy, (eFigure 15, supplement 1).