Effect of active/intensive versus placebo/guideline blood pressure lowering interventions on outcomes after acute intracerebral haemorrhage (ICH)
| Outcome | Randomly allocated treatment group | Adjusted OR* (95% CI) | P value | |
| Active/intensive | Placebo/guideline | |||
| Primary: unfavourable shift in modified Rankin Scale (mRS) scores | /3062 | /3039 | 0.97 (0.88 to 1.06) | 0.50 |
| 0 | 206 (6.7) | 219 (7.2) | ||
| 1 | 566 (18.5) | 525 (17.3) | ||
| 2 | 555 (18.1) | 553 (18.2) | ||
| 3 | 521 (17.0) | 514 (16.9) | ||
| 4 | 582 (19.0) | 578 (19.0) | ||
| 5 | 221 (7.2) | 245 (8.1) | ||
| 6 | 411 (13.4) | 405 (13.3) | ||
| Secondary | /3062 | /3039 | ||
| Dependency or death (mRS 3–6) | 1735 (56.7) | 1742 (57.3) | 0.95 (0.84 to 1.08) | 0.42 |
| Severe dependency or death (mRS 4–6) | 1214 (39.7) | 1228 (40.4) | 0.95 (0.84 to 1.08) | 0.41 |
| Death | 411/3111 (13.2) | 405/3087 (13.1) | 1.01 (0.85 to 1.20) | 0.91 |
| Safety outcomes | ||||
| Any SAE | 781/3000 (26.0) | 727/2982 (24.4) | 1.13 (0.99 to 1.29) | 0.07 |
| Neurological deterioration† | 304/2911 (10.4) | 300/2913 (10.3) | 1.02 (0.86 to 1.22) | 0.81 |
| Severe hypotension† | 25/2831 (0.9) | 15/2840 (0.5) | 1.73 (0.89 to 3.37) | 0.11 |
| Cardiac SAE | 69/2978 (2.3) | 70/2980 (2.3) | 1.01 (0.72 to 1.42) | 0.96 |
| Renal SAE | 34/2978 (1.1) | 28/2980 (0.9) | 1.26 (0.75 to 2.09) | 0.38 |
| Haematoma growth at 24 hours‡ | ||||
| Mean growth (95% CI), mL | 3.2 (2.5 to 3.9) | 4.3 (3.4 to 5.2) | Absolute difference: −1.10 (−2.22 to 0.01) | 0.05 |
| Absolute growth ≥6 mL | 212/1280 (16.6) | 252/1230 (20.5) | 0.75 (0.60 to 0.92)§ | 0.007 |
| Relative growth ≥33% | 296/1280 (23.1) | 326/1230 (26.5) | 0.82 (0.68 to 0.99)§ | 0.03 |
*Model adjusted for age, sex, National Institutes for Health Stroke Scale score, time from onset to randomisation and trial (random effect); all patients followed up for 90 days except those in the Scandinavian Candesartan Acute Stroke Trial who were followed for 6 months.
†Treatment related, as defined by each trial.
‡Trials contributing data for analyses of haematoma growth at 24 hours: INTERACT1&2, ICH-ADAPT, ATACH-II, FAST-MAG.
§Model adjusted for age, sex, baseline haematoma volume and time from symptom onset to randomisation.
CI, confidence interval; mRS, modified Rankin Scale; OR, odds ratio; SAE, serious adverse event.