Subarachnoid Extension of Intracerebral Hemorrhage and 90-Day Outcomes in INTERACT2
Background and Purpose—The prognostic significance of subarachnoid extension of intracerebral hemorrhage was determined in the INTEnsive blood pressure Reduction in Acute Cerebral hemorrhage Trial (INTERACT2) study.
Methods—INTERACT2 was an open randomized controlled trial of early intensive compared with guideline-recommended blood pressure lowering in patients with elevated systolic blood pressure within 6 hours of intracerebral hemorrhage. Independent predictors of death or major disability (scores 3–6 on the modified Rankin Scale) at 90 days were analyzed in logistic regression models.
Results—Of 2582 participants, 192 (7%) had subarachnoid extension, which was associated with larger hematoma volumes (P<0.0001) and higher National Institute of Health Stroke Scale score (P<0.0001). Subarachnoid extension predicted death or major disability at 90 days (71% versus 53%; unadjusted odds ratio, 2.25; 95% confidence interval, 1.63–3.10; P<0.0001). The association remained significant after adjusting for age, region, lipid-lowering therapy, systolic blood pressure, glucose, location of hematoma, intraventricular extension, and randomized treatment (odds ratio, 2.17; 95% confidence interval, 1.50–3.14; P<0.0001), but not after further adjustment for baseline hematoma volume (P=0.62).
Conclusions—Subarachnoid extension of intracerebral hemorrhage is associated with poor prognosis, which is determined by a larger volume of the underlying intraparenchymal hematoma.
Acute intracerebral hemorrhage (ICH) is the least treatable form of stroke affecting more than a million people in the world each year.1 Effective management requires better understanding of pathophysiology and prognostic factors. Recently, a hospital-based observational study found that subarachnoid extension of ICH was significantly associated with increased 14-day mortality and 28-day disability in 234 patients with acute ICH.2 The authors proposed that subarachnoid extension could produce cortical dysfunction and injury from mechanisms akin to those observed in aneurysmal subarachnoid hemorrhage (SAH). In comparison with subarachnoid hemorrhage, however, only a small amount of blood usually reaches the subarachnoid space in ICH.3 The clinical significance of subarachnoid extension and clinical outcome in ICH was examined in the INTERACT2 study,4 where rigorous evaluations were undertaken in a large number of participants with 90 days of follow-up.
The design and main results of the INTERACT2 trial are outlined elsewhere.4 Briefly, 2829 patients with spontaneous ICH within 6 hours of onset and elevated systolic blood pressure were assigned to receive intensive (target systolic level <140 mm Hg within 1 hour) or guideline-recommended (<180 mm Hg) blood pressure–lowering treatment. The study was approved by the ethics committees for each site and informed consent was obtained from all patients or relevant surrogates.
Brain imaging was performed according to standardized techniques. Digital computed tomographic images were collected in uncompressed DICOM format and analyzed centrally for the presence of bleeding into the subarachnoid space, location and volume of hematoma, and any intraventricular extension. Subarachnoid extension of ICH was classified into sulcal and generalized patterns, the latter was further classified according to the modified Fisher scale.5
Outcomes were death or major disability, defined as a modified Rankin Scale (mRS) score6 of 3 to 6, death, major disability (mRS, 3–5), and an ordinal analysis of mRS scores at 7, 28, and 90 days.
Overall, 2582 participants with information on baseline computed tomography and 90-day mRS were included in analyses. There were 192 (7%) patients with subarachnoid extension of ICH at baseline, who had significantly higher National Institute of Health Stroke Scale scores and larger hematoma volumes (all P<0.0001; Table I in the online-only Data Supplement).
The Table shows that subarachnoid extension was associated with increased risk of death or major disability at 90 days (71% versus 53%; P<0.0001). Likewise, patients with subarachnoid extension had higher risks of poorer function on ordinal analysis of mRS scores (P<0.0001; Figure). The association between subarachnoid extension and death or major disability remained significant in multivariable-adjusted model 1 (P<0.0001; Table). However, there was no longer any independent significance after further adjustment for baseline hematoma volume (model 2; P=0.62). Additional adjustment of seizures during follow-up did not change the findings (Table II in the online-only Data Supplement).
Table III in the online-only Data Supplement shows baseline characteristics by distribution of subarachnoid extension of ICH. There were no clear differences in baseline hematoma volume between sulcal (median, 28.9 mL) and generalized patterns (28.5 mL; P=0.49). Patients with generalized pattern had significantly higher 90-day mortality than those with sulcal subarachnoid extension (odds ratio, 2.54; 95% confidence interval, 1.32–4.90; P=0.005; Table IV in the online-only Data Supplement) but not after adjustment for other covariates including baseline hematoma volume (model 2; P=0.12).
Among 964 patients in the computed tomographic substudy, 952 with information on 90-day mRS were included in further analysis. There were 73 (8%) patients with baseline and 36 (4%) with delayed subarachnoid extension. Increased risks of death or major disability at 90 days were observed for both baseline (odds ratio, 1.98; 95% confidence interval, 1.04–3.80; P=0.008) and delayed subarachnoid extension (odds ratio, 4.01; 95% confidence interval, 1.27–12.61; P=0.02) even after adjustment for other covariates (model 1), but not after further adjustment for baseline hematoma volume (model 2; P=0.96 for baseline and 0.19 for delayed subarachnoid extension).
In INTERACT2, which included >2500 patients, subarachnoid extension of ICH was associated with increased risks of death or major disability, but not after adjustment for baseline hematoma volume. These findings suggest that hematoma volume is the primary determinant of poor prognosis in patients with subarachnoid extension of ICH.
Our results differ from those of a previous hospital-based observational study of acute ICH, where a strong independent association was evident between subarachnoid extension and 14-day mortality and 28-day disability.2 However, that study reported a much higher rate of subarachnoid extension (40%), possibly because of differences in hematoma location and use of antithrombotics and was based on relatively small numbers (234) of patients. The greater sample size in INTERACT2 allowed us to undertake more precise adjustment for prognostic variables and as the data came from multiple centers, the results have strong external validity.
Our findings suggest that subarachnoid extension in ICH is often restricted to a few sulci of the ipsilateral convexity. Given the strong association between the amount of subarachnoid hemorrhage and the risk of delayed cerebral ischemia in patients with ruptured intracranial aneurysms,7,8 it is unlikely that the small amounts of sulcal hemorrhage often observed in ICH would be a major determinant of outcome. Although a generalized pattern of subarachnoid extension was associated with greater mortality in univariable analysis, this association was no longer significant after adjustment for other covariates.
Our study is limited in being subsidiary to the overall design of the trial and for the number of patients with each type of subarachnoid extension being insufficient to allow detailed multivariable analyses.
In summary, subarachnoid extension is associated with more severe clinical grade at presentation and poorer 90-day prognosis in acute ICH. The prognostic significance of subarachnoid extension of ICH on clinical outcome can be explained by its association with larger hematoma volumes.
Sources of Funding
The INTERACT2 study was supported by Program (571281) and Project (512402 and 1004170) grants from the National Health and Medical Research Council of Australia. The study was designed, conducted, analyzed, and interpreted by the investigators independent of sponsors.
*For a full list of investigators, see Ref. 4.
The online-only Data Supplement is available with this article at http://stroke.ahajournals.org/lookup/suppl/doi:10.1161/STROKEAHA.113.003524/-/DC1.
- Received September 12, 2013.
- Accepted September 19, 2013.
- © 2013 American Heart Association, Inc.
- Maas MB,
- Nemeth AJ,
- Rosenberg NF,
- Kosteva AR,
- Guth JC,
- Liotta EM,
- et al
- de Rooij NK,
- Greving JP,
- Rinkel GJ,
- Frijns CJ