3.1 Included Studies and Patient Characteristics

A total of 12 RCTs or cohort studies [4,5,6, 8,9,10,11,12,13,14,15,16] were included in this study, with a total of 9271 patients. A total of ten studies with documented antihypertensive strategy were included to compare the effect of intensive or conventional antihypertensive treatment in this meta-analysis. Another two studies were used to compare the mortality risk of antihypertensive drugs, in which the antihypertensive strategy was not distinguished by intensive or conventional treatments. The features of the included studies are presented in Supplementary Table 1. The Bayesian network diagram for the efficacy comparison of included antihypertensive drugs is shown in Supplementary Fig. 4.

3.2 Quality Evaluation of the Included Literature

Literature quality evaluation with the Newcastle–Ottawa Scale evaluation criteria for cohort study and with the Cochrane bias risk assessment tool for RCTs was conducted by two researchers. The evaluations are shown in Supplementary Fig. 5.

3.3 Effectiveness Analysis of Intensive and Conventional Antihypertensive Treatments

As shown in Fig. 2, for 90-day functional independence, a total of 6436 patients, including 3354 patients in the intensive antihypertensive treatment group and 3082 in the conventional antihypertensive treatment group from three RCTs and four cohort studies, were analyzed. The comparison between intensive antihypertensive treatment and conventional strategy (RR = 0.95, 95% CI = 0.89–1.02, P = 0.14) indicated that there was no significant difference in improving 90-day functional independence in patients with intracerebral hemorrhage.

Fig. 2

Forest diagram of effectiveness. CI confidence interval, M-H Mantek Haenszel, Intensive intensive antihypertensive treatment, Conventional conventional antihypertensive treatment

For 90-day mortality, a total of 6557 patients, including 3416 patients in the intensive antihypertensive group and 3141 in the conventional antihypertensive group from four RCTs and four cohort studies, were analyzed. There was no significant difference between intensive antihypertensive treatment and conventional strategy (RR = 0.97, 95% CI = 0.83–1.13, P = 0.67).

For 24-h hematoma enlargement, 3607 patients, including 1955 patients in the intensive antihypertensive group and 1652 in the conventional antihypertensive group from three RCTs and four cohort studies, were analyzed. Intensive antihypertensive treatment significantly reduced the occurrence of 24-h hematoma enlargement in patients with intracerebral hemorrhage compared with conventional strategy (RR = 0.76, 95% CI = 0.67–0.87, P < 0.0001). The definition of hematoma enlargement remains controversial. Generally, it is defined by an absolute increase by 6 ml or 12.5 ml in the hematoma volume or a relative increase by 33% in literature [17, 18]. Since the absolute increase of hematoma volume by 12.5 ml or the relative increase by 33% is commonly adopted for the definition of hematoma enlargement in the studies, and the large volume of ICH is correlated with poor prognosis [19], this definition of hematoma enlargement was chosen in our study. For the comprehensive analysis, the absolute increase of hematoma volume by 6 ml was also analyzed, as shown in Supplementary Fig. 6, indicating that intensive antihypertensive therapy also significantly reduced hematoma enlargement.

Considering 24-h neurological deterioration, a total of 4469 patients, consisting of 2210 patients in the intensive antihypertensive group and 2259 in the conventional antihypertensive group from four RCTs and three cohort studies, were analyzed. There was no significant difference between intensive antihypertensive treatment and conventional treatment (RR = 1.05, 95% CI = 0.90–1.22, P = 0.52).

Overall, in terms of effectiveness, intensive antihypertensive treatment significantly reduced the risk of 24-h hematoma enlargement in acute intracerebral hemorrhage, but it had no effect on 90-day functional independence, 90-day mortality, or 24-h neurological deterioration compared with conventional antihypertensive treatment.

3.4 Safety Analysis of Different Antihypertensive Treatments

In Fig. 3, for 90-day renal insufficiency, there were 1659 patients analyzed, including 808 patients in the intensive antihypertensive group and 851 in the conventional antihypertensive group, from two RCTs. Compared with the conventional antihypertensive treatment, intensive antihypertensive treatment was associated with an increased probability of 90-day renal insufficiency (RR = 2.31, 95% CI = 1.05–5.05, P = 0.04).

Fig. 3

Forest diagram of safety. CI confidence interval, M-H Mantek Haenszel, Intensive intensive antihypertensive treatment, Conventional conventional antihypertensive treatment

For 90-day renal failure, a total of 2005 patients were analyzed, including 982 patients in the intensive antihypertensive group and 1023 in the conventional antihypertensive group, from two RCTs and one cohort study. The comparison between intensive antihypertensive treatment and conventional treatment indicated that intensive antihypertensive treatment was more likely to lead to 90-day renal failure (RR = 2.42, 95% CI = 1.20–4.86, P = 0.01).

For 90-day acute coronary events, a total of 4175 patients were analyzed, consisting of 2073 patients in the intensive antihypertensive group and 2102 in the conventional antihypertensive group from three RCTs. There was no significant difference between intensive antihypertensive treatment and conventional strategy (RR = 1.33, 95% CI = 0.48–3.68, P = 0.59).

For 90-day cerebral rebleeding, 4488 patients, including 2207 patients in the intensive antihypertensive group and 2281 in the conventional antihypertensive group from two RCTs, were analyzed. The data showed that intensive antihypertensive treatment was significantly less likely to cause 90-day cerebral rebleeding than conventional antihypertensive treatment (RR = 0.71, 95% CI = 0.52–0.96, P = 0.03).

There were 4834 patients analyzed for 90-day hypotension, including 2381 patients in the intensive antihypertensive group and 2453 in the conventional antihypertensive group, from three RCTs and one cohort study. There was no significant difference between intensive antihypertensive treatment and conventional antihypertensive treatment (RR = 1.37, 95% CI = 0.76–2.47, P = 0.30).

For 90-day cerebral infarction, a total of 4488 patients, including 2207 patients in the intensive antihypertensive group and 2281 in the conventional antihypertensive group from two RCTs and two cohort studies, was analyzed. There was no significant difference in the occurrence of 90-day cerebral infarction between intensive antihypertensive treatment and conventional treatment (RR = 0.78, 95% CI = 0.53–1.14, P = 0.19).

Overall, intensive antihypertensive treatment significantly increased the incidence of 90-day renal insufficiency and renal failure, although there was no statistical difference in the occurrence of 90-day acute coronary events, hypotension, or cerebral infarction compared with conventional antihypertensive treatment.

3.5 Subgroup Analysis of Different Intracerebral Hematoma Volume

Since most RCTs studies assessing the efficacy of antihypertensive treatment on acute intracerebral hemorrhage enrolled the patients with an average hematoma volume of fewer than 15 ml, the effect of antihypertensive treatments with hematoma volume greater than 15 ml deserved further exploration. Therefore, subgroup analysis was performed to confirm whether the efficacy and safety of antihypertensive treatments depended on the blood volume of intracerebral hemorrhage.

For the analysis of 24-h hematoma enlargement, RevMan software was used to combine the effect size in the literature, and it was found that heterogeneity of literature was high (I2 > 50%) in intracerebral hemorrhage with blood volume greater than 15 ml. Then, sensitivity analysis was carried out for all enrolled studies using the one-by-one culling method. After the sources of heterogeneity were identified and culled, recombined effect size was performed and the subgroup analysis was conducted again. The results of subgroup analysis are shown in Figs. 4 and 5.

Fig. 4

Forest diagram of efficacy subgroup analysis; A. intracranial hemorrhage < 15 ml; B. intracranial hemorrhage ≥ 15 ml. CI confidence interval, M-H Mantek Haenszel, Intensive intensive antihypertensive treatment, Conventional conventional antihypertensive treatment

Fig. 5

Forest diagram of safety subgroup analysis; A. intracranial hemorrhage < 15 ml; B. intracranial hemorrhage ≥ 15 ml. CI confidence interval, M-H Mantek Haenszel, Intensive intensive antihypertensive treatment, Conventional conventional antihypertensive treatment

As shown in Fig. 4A, for patients with intracerebral hemorrhage with fewer than 15 ml of blood volume, there was no significant difference in 90-day functional independence between intensive antihypertensive treatment and conventional treatment (RR = 0.98, 95% CI = 0.91–1.05, P = 0.55). In terms of 90-day mortality, there was also no significant difference (RR = 0.98, 95% CI = 0.84–1.15, P = 0.83). For 24-h neurological deterioration, intensive antihypertensive treatment was not statistically significant compared with conventional strategy (RR = 0.94, 95% CI = 0.81–1.12, P = 0.49). However, in terms of 24-h hematoma enlargement, intensive antihypertensive treatment had a significant protective effect compared with conventional strategy (RR = 0.77, 95% CI = 0.67–0.89, P = 0.0003).

In Fig. 4B, for patients with intracerebral hemorrhage greater than 15 ml of blood volume, the functional independence of the patients in the intensive antihypertensive treatment group was significantly superior to those in the conventional antihypertensive treatment group (RR = 0.78, 95% CI = 0.65–0.94, P = 0.01). In terms of 90-day mortality, there was no significant difference between intensive antihypertensive treatment and conventional treatment (RR = 0.77, 95% CI = 0.42–1.42, P = 0.40). For 24-h neurological deterioration, intensive antihypertensive treatment was similar with conventional treatment (RR = 1.90, 95% CI = 0.64–5.61, P = 0.25). In terms of 24-h hematoma enlargement, the two antihypertensive treatments had no significant difference (RR = 1.37, 95% CI = 0.85–2.20, P = 0.20).

In terms of safety, subgroup analysis was conducted on a total of 5 indicators, including 90-day renal insufficiency, renal failure, intracranial rebleeding, hypotension, and cerebral infarction. As shown in Fig. 5A, for the subgroup including patients with less than 15 ml of hematoma volume, in terms of 90-day renal insufficiency, intensive antihypertensive treatment showed an unfavorable trend compared with conventional antihypertensive treatment (RR = 2.00, 95% CI = 0.50–7.95, P = 0.33). There was no significant difference between intensive antihypertensive treatment and conventional strategy in 90-day intracranial rebleeding (RR = 1.01, 95% CI = 0.33–3.14, P = 0.98), hypotension (RR = 1.21, 95% CI = 0.61–2.39, P = 0.59), and cerebral infarction (RR = 0.94, 95% CI = 0.57–1.55, P = 0.81). However, for 90-day renal failure, intensive antihypertensive treatment was more likely to result in renal failure than conventional treatment (RR = 2.33, 95% CI = 1.07–5.04, P = 0.03).

For the subgroup enrolling patients with greater than 15 ml of blood volume in Fig. 5B, there was no difference between intensive antihypertensive treatment and conventional treatment in terms of 90-day renal insufficiency (RR = 2.47, 95% CI = 0.95–6.42, P = 0.06) or renal failure (RR = 2.85, 95% CI = 0.56–14.58, P = 0.21), although there was an adverse trend for intensive antihypertensive treatment. The difference was not statistically significant for the comparison of intensive antihypertensive treatment with conventional treatment regarding 90-day hypotension (RR = 1.99, 95% CI = 0.59–6.75, P = 0.27) or cerebral infarction (RR = 0.59, 95% CI = 0.32–1.08, P = 0.09). However, compared with conventional antihypertensive treatment, intensive antihypertensive treatment was more likely to improve 90-day intracranial rebleeding (RR = 0.68, 95% CI = 0.49–0.94, P = 0.02).

For subgroup analysis, when cerebral hematoma volume was less than 15 ml, intensive antihypertensive treatment showed only an advantage of 24-h hematoma enlargement, but not the improvement of 90-day functional independence, mortality, or 24-h neurological deterioration. Furthermore, intensive antihypertensive treatment significantly increased the incidence of 90-day renal failure. However, for the subgroup of cerebral hematoma volume greater than 15 ml, intensive antihypertensive treatment had obvious advantages in 90-day functional independence and intracranial rebleeding and did not cause adverse effects on renal function.

3.6 Mortality Risk of Antihypertensive Drugs3.6.1 Comparison of Mortality Risk of Antihypertensive Drugs

A Bayesian network meta-analysis was performed for the mortality risk of antihypertensive drugs. As shown in Fig. 6, nicardipine was compared with labetalol (RR = 1.08, 95% CI = 0.50–2.30), nitroprusside (RR = 0.62, 95% = 0.33–1.18), and nicardipine combined with labetalol (RR = 1.03, 95% = 0.49–2.18). In addition, labetalol was compared with nitroprusside (RR = 0.58, 95%CI = 0.22–1.56) and nicardipine combined with labetalol (RR = 0.96, 95% CI = 0.33–2.77). Further, nitroprusside was compared with nicardipine combined with labetalol (RR = 1.65, 95% CI = 0.62–4.38).

Fig. 6

Head-to-head comparison diagram of mortality risk; the diagram demonstrates the comparison of the column with the row treatment modality. For example, nicardipine was compared with labetalol (RR = 1.08, 95% CI = 0.50–2.30), and labetalol was compared with nicardipine (RR = 0.93, 95% CI = 0.44–1.99). The same color represents the mutual comparison of the same two antihypertensive drug regimens

3.6.2 Rank Probability and Surface Under the Cumulative Ranking Curve (SUCRA)

For the four antihypertensive drugs regimes included in this study, the mortality risk probability, as shown in Fig. 7, was successively nicardipine (0.01), labetalol (0.09), nitroprusside (0.77), and nicardipine combined with labetalol (0.11), indicating that nitroprusside had the highest mortality risk. The rank of SUCRA was successively nitroprusside (0.89), nicardipine (0.39), nicardipine combined with labetalol (0.38), and labetalol (0.33). In conclusion, among the four antihypertensive drug regimens, nitroprusside had the highest risk of death, while labetalol was the best regimen.

Fig. 7

Probability diagram of mortality risk. The color from dark to light represents the death risk from low to high. SUCRA surface under the cumulative ranking curve