Lien vers les tableaux de données probantes et la liste des références (en anglais)
Patients with suspected intracerebral hemorrhage (ICH) should undergo a non-contrast CT or MRI immediately to confirm the diagnosis. Both forms of imaging have been shown to accurately detect acute intracranial hemorrhage (Chalela et al. 2007, Fiebach et al. 2004). Given that an underlying macrovascular cause is responsible for 15%-25% of non-traumatic ICHs, further imaging studies should be conducted using CT angiography, MR angiography or digital subtraction angiography to detect possible arteriovenous malformations, aneurysms or cases of cerebral venous sinus thrombosis. In the DIAGRAM study, Van Ash et al. (2015) estimated the diagnostic yield and accuracy of CTA performed in the acute phase after non-contrast CT, and with the addition of MRI/MRA and then digital subtraction angiography combined (DSA), if the results of the CTA scans were negative. In a cohort of 298 patients, an underlying vascular cause was identified in 69 patients (23%), using the reference standard of best available evidence from all diagnostic procedures. The diagnostic yield of CTA was 17%, 18% with the addition of MRI/MRA and 23% with the addition of DSA. The positive predictive value (PPV) of CTA was 72% (95% CI 60% to 82%). The addition of MRI/MRA increased PPV to 77%, (95% CI 65% to 86%), while addition of DSA increased it to 100% (95% CI 80%-100%). A single cavernoma was not identified using any of the imaging techniques. The accuracy of CTA to identify vascular lesions compared with DSA reported in other studies has been higher. In another DIAGRAM publication, younger age, lobar or posterior fossa location of ICH, absence of neuroimaging markers of cerebral small vessel disease, and a positive or inconclusive CTA were independent predictors for an ultimate macrovascular cause for the ICH being identified within 1 year of follow-up (Hilkens et al. 2017). Josephson et al. (2014) examined the diagnostic test accuracy of CTA and MRA versus intra-arterial digital subtraction angiography (IADSA) for the detection of intracranial vascular malformations. Eight studies compared CTA with IADSA and 3 studies compared MRA with IADSA. The sensitivity and specificity of both strategies was excellent (CTA: sensitivity 0.95, specificity 0.99; MRA: 0.98 and 0.99). Wong et al. (2011) reported the sensitivity, specificity and accuracy of CTA to be 100%, 98.6% and 99.1%, respectively, in a prospective sample of 109 patients, while Delgado Almandoz et al. (2009) reported the respective sensitivity and specificity as 96.1% and 98.5%.
Blood Pressure Management
While the optimum blood pressure targets for patients who have experienced a spontaneous ICH are not known, systolic blood pressure (SBP) greater than 180 mm Hg is thought to increase the risks of rebleeding and hematoma expansion. While this finding suggests that steps to lower blood pressure aggressively would be beneficial, the results from several large controlled trials on the topic are not conclusive. Qureshi et al. (2016) reported in the ATACH-2 trial that intensive blood pressure management, with an SBP target of 110-139 mm Hg did not reduce the risk of death or disability at 90 days (adjusted OR=1.04, 95% CI 0.85-1.27, p=0.72), or hematoma expansion within 24 hours (adj OR=0.78, 95% CI 0.58-1.03, p=0.08), compared with standard treatment (target of 140-179 mm Hg) in 1,000 patients admitted acutely with an ICH, while recent results from a subgroup analysis of the trial (Leasure et al. 2019) suggested that patients with deep intracranial hemorrhage may benefit from intensive treatment. Within this subgroup, the risk of hematoma expansion (defined as an increase of ≥33%) was significantly lower for patients in the intensive group (adj OR=0.61, 95% CI 0.42-0.88, p=0.009. The effect of treatment was modified by deep ICH location (p for interaction=0.02), whereby patients with a basal ganglia hemorrhage benefited from intensive BP reduction and those with thalamic hemorrhages did not. In the INTERACT-2 trial, (Anderson et al. 2013), patients in the intensive treatment arm also had SBP target of <140 mm Hg. At 90 days, 52.0% of patients in the intensive group had experienced a poor outcome (mRS score 3-5) compared with 55.6% of patients in the standard treatment group (OR=0.87, 95% CI 0.75-1.01, p=0.06). There was no significant difference between groups in 90-day mortality (11.9% vs. 12.0%, OR=0.99, 95% CI 0.79-1.25, p=0.96). There was, however, a significant shift towards the distribution of mRS scores favouring less disability among patients in the intensive group (OR=0.87, 95% CI 0.77-1.00, p=0.04). In contrast, data from the INTERACT 1 study showed that early intensive blood pressure lowering reduced hematoma growth (Anderson et al. 2008). Recent evidence from the EnRICH trial (Meeks et al. 2019) indicates blood pressure variability in the hyperacute and acute periods may play a more important role in outcome, whereby high variability was associated with poorer outcomes.
Although not currently recommended for use in spontaneous ICH, another potential treatment that may help to optimize hemostasis and minimize hematoma expansion is recombinant activated factor VII (rFVIIa). In a recent trial that included 69 patients with primary spontaneous acute ICH who were spot-sign positive and randomized to receive rFVIIa (80 μg/kg or placebo), there were no significant differences between groups in the change (increase) in median parenchymal ICH volume from baseline to 24 hours (2.5 vs. 2.6 mL, p=0.89), or in median total hemorrhagic volume (3.2 mL vs. 4.8 mL, p=0.91) (Gladstone et al. 2019). Results of the FAST II (Mayer et al. 2005) and FAST III (Mayer et al. 2008) trials, suggested that treatment with rFVIIa could help to blunt the increase in ICH volume at 24 hours post treatment; however, the trials conflicted with respect to functional outcome. The FAST III trial did not report a significant difference in the proportion of patients with death or severe disability at 90 days, while FAST II reported a lower proportion in active treatment group patients. The authors of a recent Cochrane review (Al-Shahi Salman et al. 2018) stated that they could not draw firm conclusions of the benefit of blood clotting factors in the treatment of ICH, but noted ongoing research in subgroups (e.g. younger patients, earlier time windows). Other hemostatic therapies are under investigation. The benefits of the antifibrinolytic agent tranexamic acid in major trauma have increased interest in its potential benefits in spontaneous ICH. In the TICH-2 trial, the use of tranexamic acid (1 g bolus, followed by 1 g infused over 8 hours) was shown to be safe, seemed to reduce hematoma expansion and reduced early deaths, but ultimately did not improve functional outcomes at 90 days in spontaneous ICH patients treated within 8 hours of symptom onset (Sprigg et al. 2018).
Management of Anticoagulation
For patients who had been managed with warfarin prior to ICH, the results of the INCH trial (Steiner et al. 2016) indicate that treatment with prothrombin complex concentrate (PCC) is superior to intravenous fresh frozen plasma (FFP). The trial was halted early due to safety concerns, after significantly more patients in the PCC group achieved anticoagulation reversal (INR ≤1.2) within 3 hours after treatment (67% vs. 9%, OR=30.6, 95% CI 4.7-197.9, p=0.0003). There are other options when treating patients taking non-vitamin K oral anticoagulants. Treatment with idarucizumab, has been shown to be effective in reversing anticoagulation for patients requiring surgery or other invasive procedures, who had been previously receiving treatment with the direct oral anticoagulation agent, dabigatran (Pollack et al. 2015). The ANNEXA-4 trial (Connolly et al. 2019) included patients who had sustained acute major bleeding occurring while taking a factor Xa inhibitor. The primary site of bleeding was intracranial in 64% of 352 patients enrolled. Following treatment with andexanet, there was a median reduction of 92% in anti–factor Xa activity among the patients who had been taking apixaban or rivaroxaban, while 82% of all patients who could be evaluated had excellent or good hemostasis 12 hours after infusion. The ongoing ANNEXA-I trial is assessing the clinical efficacy of random assignment to andexanet alfa compared with standard treatment (including PCC) in factor Xa inhibitor-related ICH.
The role of surgical intervention for the evacuation of supratentorial ICH remains uncertain. While these procedures can stop bleeding, prevent rebleeding, and prevent secondary brain damage by removing the mass effect, trial results have been disappointing. In the Surgical Trial in Intracerebral Hemorrhage (STICH) trial, 1,033 patients with CT evidence of a spontaneous ICH that had occurred within 72 hours were randomized to early (within 24 hours) surgery for evacuation of the hematoma or to initial conservative treatment (Mendelow et al. 2005). There was no difference in the percentage of patients with a favourable outcome, which was defined based on initial prognosis. 26% of patient in the early surgical group vs. 24% of patients in the medical management group had a favourable outcome (OR=0.89, 95% CI 0.66-1.19, p=0.414, absolute benefit=2.3, 95% CI -3.2 to 7.7). There was speculation that the null findings may have been attributed, in part, to the inclusion of patients with intraventricular hemorrhages with poorer prognosis and the late timing of intervention. Therefore, in the Surgical Trial in Lobar Intracerebral Haemorrhage (STICH II) trial (Mendelow et al. 2013), 601 patients were randomized to early craniotomy (within 12 hours) to evacuate hematoma or treated conservatively, following spontaneous superficial ICH affecting the lobar region, within 1 cm of the cortex and without ventricular extension, the subgroup of patients thought to be most likely to benefit. While there were no differences between groups in the proportion of patients who experienced a good outcome at 6 months (41% surgical group vs. 38% medical management group; OR=0.86, 95% CI 0.62-1.20, p=0.367) or who had died (18%. surgical vs. 24% medical management, OR=0.71, 95% CI 0.48-1.06, p=0.095), patients with poor prognosis were more likely to have a favourable outcome (OR=0.49, 95% CI 0.26-0.92, p=0.04). In contrast, patients with a good prognosis were no more likely to benefit from early surgery (OR=1.12, 95% 0.75-1.68, p=0.57). The results of a patient-level meta-analysis, which included the results from 8 RCTs indicated that the odds of unfavourable outcome at 3-6 months were significantly reduced among persons aged 50-69 years, in those who received surgery within 8 hours of the event, in those with baseline hematoma volumes of 20-50 mLs and with baseline Glasgow Coma Scale (GCS) score was between 9 and 12 (Gregson et al. 2012).
Minimally invasive surgery with the addition of thrombolysis has been used to treat patients with ICH and intraventricular hemorrhages, with mixed results. In the CLEAR III trial (Hanley et al. 2017), 500 patients, with spontaneous ICH ≤30 cc and an intraventricular hemorrhage (IVH) obstructing third and/or fourth ventricles, were included. Patients were randomized to irrigation of the ventricles with a maximum dose of 12.0 mg alteplase or saline placebo via a routine extraventricular drain. Treatment with alteplase did not improve the likelihood of a good functional outcome. The proportion of patients achieving an mRS score of ≤3 at 6 months was non-significantly higher in the alteplase group (48% vs. 45%, RR=1.06, 95% CI 0.88-1.28, p=0.554), although the odds of death at 6 months were significantly reduced in the alteplase group (OR=0.50, 95% CI 0.31-0.80, p=0.004). Treatment with alteplase via the MISTIE technique significantly reduced hematoma size compared with standard care in 506 patients with supratentorial ICH of ≥30 mL, although there was no significant difference between groups in the proportion of patients who achieved a good functional outcome (mRS 0-3) at one year (45% vs. 41%)(MISTIE III,Hanley et al. 2019). One-year and 180-day mortality were both significantly lower in the MISTIE group, but not 30-day mortality.
Sex and Gender considerations
Data on sex specific differences in ICH is limited. Future research directions should include sex or gender specific analysis, regardless of ICH cause and should consider biological age (specifically across the women’s lifespan), clinical presentations, hematoma location/volume, expansion/risk for expansion, imaging, therapy, functional outcomes and patient-reported outcome measures.
Prior studies have demonstrated sex-disparities in ischemic stroke, but there is still a knowledge-gap regarding the role of sex or gender on the ICH risk, clinical presentation, management and /or outcomes.
Large studies, such as the ERICH study and the MGH hospital-based ICH cohort study have observed sex-related differences in primary ICH location. Lobar ICH is more common in females, while deep ICH was more frequent in males.
In the acute setting, women do not receive less aggressive care, including surgery or palliative care, than men after controlling for the substantial comorbidity differences. However, some studies found that women are more likely to receive early DNR orders after ICH than men.