Complications of intracerebral haemorrhage
The Lancet Neurology - Volume 11, Issue 1 (January 2012) -
|
Complications of intracerebral haemorrhage
Joyce S Balami, MRCPa
Alastair M Buchan, Prof, FMedSci
a Acute Stroke Programme, Department of Medicine and Clinical Geratology, Oxford University Hospitals NHS Trust, Oxford, UK
b Acute Vascular Imaging Centre, University of Oxford, Oxford University Hospitals NHS Trust, Oxford, UK
| * Correspondence to: Prof Alastair M Buchan, Biomedical Research Centre, University of Oxford, Oxford University Hospitals NHS Trust, John Radcliffe, Oxford, OX3 9DU, UK |
| E-mail address: alastair.buchan@medsci.ox.ac.uk |
Introduction
Spontaneous or primary intracerebral haemorrhage (ICH) is a major cause of morbidity and mortality worldwide. It is the second most common form of stroke, accounting for 10–30% of first-ever strokes. [1] , [2] , [3] The 30 day mortality for ICH has been reported to be 30–55%, [1] , [2] , [3] , [4] , [5] , [6] with half the deaths occurring in the acute phase, especially in the first 48 h. [1] , [2] , [5] The complications of ICH are among the major predictors of early mortality and poor outcome.
ICH complications include haematoma expansion (HE), perihaematomal oedema (PHE), intraventricular extension of haemorrhage (IVH) with hydrocephalus, seizures, venous thromboembolic events (VTE), hyperglycaemia, increased blood pressure (BP), fever, and infections. Complications such as HE, [7] , [8] IVH with obstructive hydrocephalus,[9] , [10] , [11] and hyperglycaemia [4] , [12] , [13] are major predictors of increased early mortality and adverse outcome during the hyperacute phase of ICH. Similarly, HE, hydrocephalus, and PHE have been associated with early neurological deterioration (END) and poor outcome.[14] A prospective observational study reported END in about 33% of patients with ICH within 48 h of onset, with an associated 30 day mortality of 47% in those with END.[14]
The complications of ischaemic stroke (IS) and their management have been reviewed extensively, [15] , [16] with little discussion of the complications of ICH. Despite its seriousness, the treatment options for ICH are restricted and few evidence-based data exist to guide the management of ICH complications. By contrast with IS, for which therapeutic advances have been made to improve clinical outcome, the management of ICH is generally supportive, but with poor prognosis because no specific treatments have been shown to improve outcome after ICH. For example, the management of HE is of unproven value because all measures aimed at restricting HE have so far not improved outcome in controlled trials. Similarly, PHE management is mainly supportive and measures aimed at decreasing intracranial pressure (ICP) are also of unproven value. Also, insufficient evidence exists with regard to the efficacy of surgical treatment for ICH, and whether or not surgical approaches are beneficial remains controversial.
Additional therapeutic dilemmas can arise over the safest and most effective approach to VTE prevention in patients with ICH because prophylaxis with anticoagulants can increase the risk of HE or further bleeding. Perhaps a more difficult therapeutic dilemma is how to manage patients with clinical thromboembolic complications after ICH, balancing the risk of subsequent life-threatening thromboembolism if untreated against the risk of recurrence of ICH. A related issue is whether or when to resume anticoagulation after ICH in patients with cardiac disease associated with high embolism risk, such as those who need mechanical valve prostheses or those with atrial fibrillation.
In this review, we focus on the early complications of ICH, discussing emerging therapies and relevant preventative and management strategies based on available evidence and guidelines. We draw attention to the scarcity of evidence to guide the management of many important and common complications of ICH.
Haematoma expansion
HE, defined as an increase in volume of 33–50% or an absolute change in haematoma volume of 12·5–20 mL (on repeat CT), is a common early and severe complication of ICH. [8] , [17] , [18] , [19] , [20] Although HE is one of the main pathophysiological phases of ICH, it can also be a serious complication subsequent to the acute phase—up to 40% of the haematoma grows in the first few hours post ictus. Various terms such as haematoma extension, expansion, progression, growth, enlargement, and rebleeding have been used to describe this subsequent increase in haematoma volume after ICH.
The precise mechanism of early HE during the acute phase is poorly understood. It is proposed to be a heterogeneous process that includes dysregulation of haemostasis via inflammatory cascade activation and matrix metalloproteinase (MMP) overexpression, breakdown of the blood–brain barrier, a sudden increase in ICP leading to local tissue distortion and disruption, and vascular engorgement due to reduced venous outflow.[8] Early HE can also result from an increased plasma concentration of cellular fibronectin (c-FN) and the inflammatory mediator interleukin-6 (IL-6).[21] Pathological studies have shown multiple microscopic and macroscopic bleeding points around the border of haemorrhages, arising from ruptured arterioles or venules that result from the stretching of surrounding vessels after clot expansion.[22]
Several studies indicate that early HE occurs in 18–38% of patients scanned within 3 h of ICH onset [7] , [18] , [19]and more than 70% develop at least some extent of HE within 24 h of symptom onset,[23] even in the absence of known coagulopathy, suggesting an active bleeding process in the hyperacute phase of ICH.[22] In warfarin-associated ICH, 27–54% of patients develop early HE and a delayed expansion because of protracted bleeding, [24] ,[25] , [26] which is associated with up to 70% increase in mortality. [24] , [27]
A prospective study showed 38% of patients had an increase in haematoma volume (>33% increase compared with admission CT) within 3 h of symptom onset. HE was evident even within 1 h of the baseline scan in two-thirds of those patients, whereas an additional 12% developed growth within the next 21 h, which suggests continued active bleeding.[7] Subsequent prospective studies have confirmed this finding, noting that 23–32% of patients had an increase in haematoma volume (>33% from baseline or 12·5 mL) in the first 24 h. [8] , [20]
CT angiographic studies that show contrast extravasation (the so-called spot sign) into the haematoma [28] , [29] ,[30] have provided additional evidence of progressive bleeding several hours after the onset of ICH. The spot sign is an important predictor of haematoma growth and might be useful in the prediction of HE with high specificity [28] , [29] ,[30] and as a predictor of mortality. [28] , [30] , [31] In fact, the spot-sign score (figure 1), which is used to grade the number of spot signs and their maximum dimensions and attenuation, is the strongest predictor of HE and is an independent predictor of in-hospital mortality and poor outcome in people with ICH.[31]
Other important predictors of HE include large haematoma volume on presentation, [29] , [32] early presentation (especially within 3 h of onset), [17] , [19] , [29] , [32] heterogeneity of haematoma density on admission CT,[33] and prior use of warfarin. [17] , [24] , [25] Blood biomarkers such as increased IL-6, MMP-9, c-FN, and tumour necrosis factor, reduced platelet activity,[21] reduced fibrinogen concentrations,[19] and increased serum creatinine,[32] have been suggested to predict HE in patients with ICH. Conflicting results were seen for increased D-dimers as a predictor of HE. [34] , [35] Some studies have shown an association between systolic BP (SBP) and HE, [36] , [37]but others have not. [7] , [19] , [32] , [38] Similarly, some studies show an association between prior use of antiplatelet drugs and HE, [32] , [34] , [39] although others do not. [40] , [41] Other risk factors include hyperglycaemia, [4] , [12] , [36] previous cerebral infarction, liver disease,[36] a decreased level of consciousness, and heavy alcohol intake.[19]
HE is often associated with END and is an independent predictor of poor outcome and increased mortality. [1] , [8] ,[23] , [42] Findings from a meta-analysis of 218 patients with ICH who had CT scans within 3 h of onset and follow-up scans within 24 h[23] showed that for every 10% increase in ICH growth there was a 5% increased risk of death, a 16% increased risk of worsening outcome as measured with the modified Rankin score (mRS), and an 18% increased likelihood of being dependent or of a poor outcome on the Barthel index.
Management
Interventions to restrict HE include haemostatic therapy, cautious lowering of high BP, quick reversal of prior anticoagulation, and surgical evacuation (table 1). Clinical trials [8] , [43] targeting HE in ICH and a meta-analysis[65]have shown that the use of recombinant factor VII (rFVIIa) limits the extent of HE in patients with non-coagulopathic ICH. However, there was an increase in thromboembolic risk with no clear clinical benefit in unselected patients. The SPOTLIGHT (Spot Sign Selection of Intracerebral Hemorrhage to Guide Hemostatic Therapy; ClinicalTrials.gov
identifier NCT01359202) and STOP-IT (Spot Sign for Predicting and Treating ICH Growth Study; NCT00810888) trials that are underway use the CT angiography spot sign to stratify patients most at risk of HE who might benefit from therapy with rFVIIa. However, rFVIIa is not recommended at present for routine use to restrict HE.[45]
identifier NCT01359202) and STOP-IT (Spot Sign for Predicting and Treating ICH Growth Study; NCT00810888) trials that are underway use the CT angiography spot sign to stratify patients most at risk of HE who might benefit from therapy with rFVIIa. However, rFVIIa is not recommended at present for routine use to restrict HE.[45]| Description | Level of evidence[*] | |
|---|---|---|
| Medical management | ||
| Haemostatic therapy with rFVIIa | In a phase 2 trial in patients with ICH, rFVIIa (NovoSeven; Novo Nordisk Health Care, Bagsvaerd, Denmark) led to a reduction in HE (p=0·01) and improvement in neurological outcomes and mortality,[8] but the phase 3 study (FAST trial), despite showing a statistically significant (p=0·009) reduction in haematoma growth in rFVIIa-treated patients, did not show any functional or survival benefit;[43] however, post-hoc analysis of the FAST data suggests a potential benefit of rFVIIa in younger patients (<70 2="" a="" as="" factors="" for="" given="" h="" haematoma="" href="http://www.mdconsult.com/das/article/body/402007529-5/jorg=journal&source=&sp=24859238&sid=0/N/944818/1.html?issn=1474-4422#r11702642044" if="" ivh="" known="" large="" ml="" name="xref_r11702642044" of="" onset="" or="" outcome="" poor="" style="color: #0560a6; margin: 0px; padding: 0px; text-decoration: initial;" substantial="" such="" symptoms="" the="" volume="" within="" without="" years="">[44] | |
AHA/ASA guidelines do not recommend rFVIIa for routine use in restricting HE in patients with ICH[45]
AHA/ASA=American Heart Association/American Stroke Association. ATACH=Antihypertensive Treatment of Acute Cerebral Hemorrhage. FAST trial=Factor Seven for Acute Haemorrhagic Stroke trial. HE=haematoma expansion. ICH=intracerebral haemorrhage. INTERACT=Intensive Blood Pressure Reduction in Acute Cerebral Haemorrhage Trial. IVH=intraventricular extension of haemorrhage. MIS=minimally invasive surgery. MISPTT=minimally invasive stereotactic puncture and thrombolysis therapy for acute intracerebral haemorrhage. mRS=modified Rankin score. NIHSS=National Institutes of Health Stroke Scale. rFVIIa=recombinant factor VIIa. STICH=Surgical Trial in Intracerebral Haemorrhage.
|
| * | The level of evidence is according to the Oxford Centre for Evidence-based Medicine (Level 1A being the highest level of evidence).[64] |
Evidence from the INTERACT (Intensive Blood Pressure Reduction in Acute Cerebral Haemorrhage Trial) and ATACH (Antihypertensive Treatment of Acute Cerebral Hemorrhage) trials show that SBP reduction might restrict HE in the hyperacute phase of ICH. [20] , [46] , [47] Both INTERACT 2 (NCT00716079) and ATACH 2 (NCT01176565) are phase 2 trials underway to further investigate the benefit of early BP reduction on HE and clinical outcome in patients with ICH.
) and ATACH 2 (NCT01176565) are phase 2 trials underway to further investigate the benefit of early BP reduction on HE and clinical outcome in patients with ICH.
The role of surgical treatment for ICH is controversial. Surgical procedures with varying amounts of supportive evidence include conventional craniotomy, [48] , [49] minimally invasive surgery (MIS), [52] , [53] , [54] , [55] , [56] ,[57] , [58] , [59] , [60] , [61] , [62] , [63] , [66] and decompressive craniectomy. [50] , [51] Results from the STICH (Surgical Trial in Intracerebral Haemorrhage) trial[49] showed no overall benefit of early surgical clot evacuation compared with initial conservative treatment in patients with ICH. However, a subgroup analysis showed a potential benefit for surgery in lobar ICH within 1 cm of the cortical surface.[49] Pending the results of the STICH phase 2 trial (NCT00716079), craniotomy is currently recommended in selected patients with lobar clots of more than 30 mL and within 1 cm of the surface. [2] , [45]
), craniotomy is currently recommended in selected patients with lobar clots of more than 30 mL and within 1 cm of the surface. [2] , [45]
MIS is a promising option with many advantages over conventional craniotomy, such as shorter surgery time, reduced tissue damage, and the fact that the procedure can be done with local anaesthesia. The several methods of MIS include stereotactic guidance with aspiration and thrombolysis with alteplase [52] , [53] , [66] or urokinase, [54], [55] , [56] , [57] , [58] and image-guided stereotactic endoscopic aspiration. [59] , [60] , [61] , [62] , [63] The preliminary analysis of MISTIE (Minimally Invasive Surgery Plus rtPA for Intracerebral Hemorrhage Evacuation;NCT00224770), an ongoing evacuation trial combining stereotactic clot aspiration with different doses of alteplase, suggests that MIS plus alteplase shows greater clot resolution than does conventional medical management.[66]
), an ongoing evacuation trial combining stereotactic clot aspiration with different doses of alteplase, suggests that MIS plus alteplase shows greater clot resolution than does conventional medical management.[66]
Although decompressive hemicraniectomy is a life-saving procedure for malignant middle cerebral artery infarction, no randomised controlled trial has been done in patients with ICH. The evidence for the potential beneficial effect of decompressive craniectomy comes from small case series. [50] , [51] In the absence of further data, both MIS and decompressive craniectomy are not recommended for routine use in patients with ICH. [2] , [45]
In anticoagulant-associated ICH (AAICH) the goal of treatment is to rapidly normalise the international normalised ratio (INR) and correct clotting factors immediately (table 2). AAICH should be reversed immediately with vitamin K and fresh frozen plasma or prothrombin complex concentrate.[45] The INCH (INR Normalization in Coumadin Associated Intracerebral Haemorrhage; NCT00928915) trial comparing the use of fresh frozen plasma with prothrombin complex concentrate in patients with AAICH is underway. Further studies are needed to confirm the efficacy of platelet replacement or other interventions aimed at preventing or treating antiplatelet-related ICH because the usefulness of platelet transfusion in this group of patients is unclear.[45] The decision as to whether or when to restart anticoagulation therapy after ICH will depend on the risk of subsequent arterial or venous thromboembolism, the risk of recurrent ICH, and the clinical state of the patient.
) trial comparing the use of fresh frozen plasma with prothrombin complex concentrate in patients with AAICH is underway. Further studies are needed to confirm the efficacy of platelet replacement or other interventions aimed at preventing or treating antiplatelet-related ICH because the usefulness of platelet transfusion in this group of patients is unclear.[45] The decision as to whether or when to restart anticoagulation therapy after ICH will depend on the risk of subsequent arterial or venous thromboembolism, the risk of recurrent ICH, and the clinical state of the patient.| Description | Level of evidence[*] | |
|---|---|---|
| Reversal of anticoagulation | The AHA/ASA recommend immediate reversal of anticoagulation; [2] , [45] recommendations for reversal of anticoagulation in patients with ICH are as follows: | .. |
| Patients with a severe coagulation factor deficiency or severe thrombocytopenia should receive appropriate factor replacement therapy or platelets, respectively[45] | Level 1C | |
| Patients with ICH whose INR is increased because of oral anticoagulants should have their warfarin treatment discontinued and receive therapy to replace vitamin K-dependent factors and to correct their INR[45] | Level 1C | |
| Vitamin K (5–10 mg intravenously) remains an adjunct to initial therapy for OAC-associated haemorrhage because normalisation of INR can take up to 24 h[45] | Level 1C | |
| FFP (10–50 U/kg) is commonly used as an adjunct to vitamin K; it acts within a few h, but is associated with greater volume expansion, which might precipitate heart failure, and requires much longer infusion times[2] | Level 2b B | |
| Prothrombin complex concentrates (10–50 U/kg), which act within a few min and have not shown improved outcome compared with FFP, might lead to fewer complications than FFP and can be considered as an alternative to FFP[45] | Level 2a B | |
| rFVIIa (40–80 μg/kg), which acts within a few min, does not replace all clotting factors, and although the INR might be lowered, clotting might not be restored in vivo; rFVIIa is therefore not routinely recommended as a sole agent for OAC reversal in ICH[45] | Level 3C | |
| The usefulness of platelet transfusions in ICH patients with a history of antiplatelet use is unclear and is considered investigational[45] | Level 2b B | |
| Resumption of anticoagulation therapy after ICH | The decision as to whether to resume anticoagulation depends on analysis of the risk of recurrent haemorrhage balanced against the risk of thromboembolism because either complication can be associated with poor outcome and high mortality | .. |
| The decision as to whether to resume anticoagulation should be based on underlying risk factors for recurrence: lobar location of the initial ICH, older age (>65 years), ongoing anticoagulation, presence of the apolipoprotein ɛ2 or ɛ4 alleles, and greater number of microbleeds on MRI[45] | Level 2a B | |
| For nonvalvular AF, long-term anticoagulation should be avoided after spontaneous lobar ICH because of the high risk of recurrence, but antiplatelet agents might be considered; antiplatelet treatment is probably safer than anticoagulation because it carries a substantially lower risk of bleeding [2] , [40] | Level 2a B | |
| Anticoagulation after non-lobar ICH might be considered depending on the indication[2] | Level 2a B | |
| When to resume anticoagulation | The optimum timing of the resumption of anticoagulation is a crucial issue with conflicting evidence: | |
| A large retrospective study of 2869 patients with AF, mechanical heart valves, and additional risk factors for stroke with warfarin-related ICH suggests resumption of warfarin after about 10–30 weeks[67] | Level 2B | |
| By contrast with reference 67, another systematic review of 492 patients concluded that anticoagulation might be resumed after 72 h[68] | Level 2A | |
| A systematic review of six retrospective studies of 120 patients with mechanical heart valves and ICH concluded that resumption of warfarin within 2 weeks is safe[69] | Level 2A | |
| However, the AHA/ASA suggest that in patients with a very high risk of thromboembolism for whom restarting warfarin is considered, warfarin can be restarted 7–10 days after ICH onset;[2] the European Stroke Initiative recommends that warfarin can be restarted after 10–14 days[70] | Level 2b B | |
| Alternatives to warfarin | Factor Xa and direct thrombin inhibitors are alternatives to warfarin, both of which might reduce the risk of thromboembolism with fewer bleeding complications[71] | .. |
| The direct thrombin inhibitor dabigatran (Pradaxa) has been shown to prevent ischaemic stroke to a similar extent as does warfarin, with reduced bleeding complications[72] | Level 1C |
AF=atrial fibrillation. AHA/ASA=American Heart Association/American Stroke Association. ATACH=Antihypertensive Treatment of Acute Cerebral Hemorrhage. FFP=fresh frozen plasma. ICH=intracerebral haemorrhage. INCH=International Normalized Ratio (INR) Normalization in Coumadin Associated Intracerebral Haemorrhage. OAC=oral anticoagulant. rFVIIa=recombinant factor VIIa.
|
| * | The level of evidence is according to the Oxford Centre for Evidence-based Medicine (Level 1A being the highest level of evidence).[64] |
Although guidelines from the American Heart Association and American Stroke Association (AHA/ASA)[2] suggest restarting warfarin 7–10 days after ICH onset in patients with a very high risk of thromboembolism, the European Stroke Initiative (EUSI) recommends starting warfarin after 10–14 days.[70]
Perihaematomal oedema
PHE, which is present in most patients with ICH, can be associated with increased mass effect and END [14] , [73]and is a predictor of poor functional outcome and mortality. [1] , [73] , [74] , [75] PHE develops early in the hyperacute phase (increasing in volume by 75% in the first 24 h),[75] evolves over many days, and increases strongly during the first week [74] before it reaches its maximum during the second week after bleeding. [1] , [14] ,[76]
Although the mechanisms of oedema formation after ICH are not fully understood, several potential mechanisms have been postulated for the different stages of PHE formation after ICH. Whereas early PHE is caused by the vasogenic effect of pro-osmotic substances (protein, electrolytes) from the clot, starting immediately after bleeding and peaking at 4–5 days,[76] delayed PHE arises from a combination of vasogenic and cytotoxic effects and lasts for 2–4 weeks.[77] Within the hyperacute first phase (a few hours post ictus) the development of hydrostatic pressure during haematoma formation and clot retraction leads to leakage of serum proteins from the clot into the surrounding tissue, resulting in vasogenic oedema.[78] A second phase (which begins a few days post ictus) results from activation of the coagulation cascade and thrombin production.[79] The third, delayed phase (which begins days to weeks post ictus) is related to erythrocyte lysis and haemoglobin-mediated toxic effects caused by the iron-catalysed production of reactive oxygen species. [80] , [81]
Thrombin-induced activation of the inflammatory cascade and overexpression of MMPs is another potential mechanism that leads to blood–brain barrier breakdown and PHE formation after ICH. [79] , [82] MMPs probably act by enhancing extracellular matrix proteolysis, damaging the basal lamina, and degrading c-Fn, a glycoprotein that is essential for haemostasis.[83]
Radiological evidence of PHE formation after ICH has been provided by various studies. [73] , [74] , [75] , [84] , [85] ,[86] In the INTERACT study of patients with CT-confirmed ICH who were assessed within 6 h of onset, PHE volume increased within 72 h of the initial CT.[84] In an MRI study, PHE volume increased most rapidly in the first 2 days after symptom onset and peaked towards the end of the second week.[86]
Other factors that have been proposed to affect PHE volume include hyperglycaemia, coagulation factors, and use of statins. [86] , [87] Likewise, increased serum concentrations of MMP-9 [88] , [89] and persistently increased SBP[90] are associated with an increased PHE volume.
Although the possible presence of an ischaemic penumbra around the area of the ICH—leading to secondary neuronal injury and cytotoxic oedema—was previously a concern, evidence against a perihaemorrhagic penumbra has since been provided by MRI and perfusion CT studies, which have linked perihaematoma hypoperfusion to reduced metabolic demand rather than tissue ischaemia. [91] , [92]
The oedema volume can exceed that of the original haematoma, leading to substantial additional mass effect with tissue shifts, and might contribute to further neuronal injury and poor outcome after ICH. [14] , [73] , [74] , [80] Also, rapidly developing PHE could lead to increased ICP or obstructive hydrocephalus and subsequent herniation.[80] The increased ICP resulting from surrounding PHE can contribute to brain injury and END, and to reduced cerebral perfusion pressure.[45]
Evidence for an effect of PHE on clinical outcome and mortality after ICH is unclear: although some observational studies have recorded an association between PHE and poor outcome, [73] , [74] , [80] another study detected no clear association.[75] Also, in another study, absolute oedema volume growth was correlated with a decrease in neurological status at 48 h after ICH, but not with 3 month functional outcome.[86]
In the INTERACT trial, both absolute and relative growth in PHE volume were associated with mortality or dependency at 90 days after adjustment for age, sex, and randomised treatment, but not when further adjusted for baseline haematoma volume.[84]
Management
The goal of therapy for PHE that occurs as a complication of ICH is to prevent secondary brain insults, reduce ICP, maintain blood supply and oxygen delivery, and optimise cerebral metabolism. The treatment options for PHE and increased ICP complicating ICH are mostly supportive (table 3). Elevation of the head to 20–30° and avoidance of pain and fever could minimise any rise in ICP. Medical measures such as hyperventilation and the use of analgesia, sedatives, and osmotic diuretics are designed to lower ICP before placement of an ICP monitor or any definitive neurosurgical intervention such as craniotomy or ventriculostomy.[2] Placement of an ICP monitor is recommended, especially in patients with a Glasgow Coma Scale (GCS) score of less than 8 and those with transtentorial herniation.[45]
Table 3 -- Management of perihaematomal oedema and increased intracranial pressure after intracerebral haemorrhage
| Description | Level of evidence[*] | ||
|---|---|---|---|
| General considerations | Little evidence exists for the management of ICP in patients with ICH; data on the management principles for increased ICP in patients with ICH are based on guidelines for traumatic brain injury, which recommend maintenance of CPP at 50–70 mm Hg[93] | .. | |
| Reduction of increased ICP and maintenance at <20 50="" as="" at="" become="" cerebral="" cpp="" font="" have="" hg="" hypoperfusion="" life-threatening="" maintenance="" mm="" of="" potentially="" prevent="" size="-1" style="margin: 0px; padding: 0px;" targets="" therapeutic="" to="" well=""> [45] | |||
AHA/ASA=American Heart Association/American Stroke Association. CPP=cerebral perfusion pressure. ICH=intracerebral haemorrhage. ICP=intracranial pressure. IVH=intraventricular extension of haemorrhage. PHE=perihaematomal oedema.
|
| * | The level of evidence is according to the Oxford Centre for Evidence-based Medicine (Level 1A being the highest level of evidence).[64] |
Intraventricular extension of haemorrhage and hydrocephalus
IVH is common after ICH, occurring in 30–50% of patients, and is a major additional predictor of poor prognosis (figure 2). [9] , [10] , [11] , [95] , [96] , [97] A relation exists between the location and volume of ICH and the presence of IVH. [11] , [95] A subanalysis of the activated rFVIIa phase 2 trial showed a relation between thalamic ICH and increased frequency of IVH.[11] This was attributable to the anatomical closeness of the thalamus to the third ventricle and the predisposition for blood to spread medially. [11] , [95]
Hallevi and colleagues[95] calculated a decompression range for each location of ICH, the range of ICH volume within which IVH is more likely to occur during the acute phase of ICH as the haematoma enlarges and below which ICH is unlikely to be complicated by IVH. They noted a narrow decompression range for thalamic and pontine haemorrhages, a wide range for lobar ICH, and a non-measurable decompression range for caudate haemorrhage.[95]
Several studies have shown that IVH in patients with ICH is an independent predictor of poor functional outcome and higher mortality, [9] , [10] , [11] , [43] , [49] , [75] , [95] , [98] with a reported overall mortality rate of 50–75%. [75] ,[99] , [100] , [101] The 30 day mortality rate was shown to be 43% in ICH patients with IVH compared with 9% in those without.[9] In the rFVIIa trial, although 38% of 375 patients with ICH had IVH at baseline, 45% had developed IVH by 24 h after presentation. A good functional outcome (mRS 0–3) was achieved by 43% of the patients without IVH at presentation compared with only 20% of those with IVH at baseline. Similarly, out of the group that had IVH growth in the first 24 h, only 7% had a good functional outcome (mRS 0–3) at 90 days.[11] A mean arterial pressure (MAP) of greater than 120 mm Hg at baseline, a large baseline ICH volume, and the presence of IVH at presentation were identified as risk factors for IVH growth.[11] Both the presence of IVH at any time and IVH growth increased the likelihood of death or severe disability at 90 days.[11]
In the subanalysis of the STICH trial, only 15% of the 377 patients with IVH had good (normal, good, or moderate recovery) Glasgow Outcome Scale functional outcomes, compared with 31% of the 375 ICH patients without IVH.[10]
Proposed mechanisms for the deleterious effects of IVH on mortality and morbidity include damage to periventricular brain structures, especially the brainstem, complications of acute obstructive hydrocephalus, and IVH-induced inflammatory response, possibly due to blood and its breakdown products. [10] , [102] , [103]
The volume of IVH affects morbidity and mortality at 30 days. [11] , [98] , [104] A review of 47 patients with ICH by Young and colleagues[98] identified a lethal volume of 20 mL, above which patients had a poor outcome. Similarly, early expansion of IVH worsens clinical outcome and increases mortality to 50–75%.[9]
Although IVH volume in itself is associated with poor outcome, an even stronger association exists between the total (ICH plus IVH) volume and adverse outcomes. [1] , [10] , [95] , [98] , [104] Another study identified a 40 mL total volume as a cutoff value, above which patients were 41 times more likely to have a poor prognosis, and 50 mL as a poor outcome threshold, above which 100% of patients would have an unfavourable outcome.[104]
Clinical features of hydrocephalus
Extension of haemorrhage into the ventricles can impede normal CSF flow and, with direct mass effects of ventricular blood, lead to acute obstructive hydrocephalus (AOH). AOH can be a life-threatening disorder, especially if the third and fourth ventricles are affected. [9] , [10] , [11] Up to 50% of patients with IVH secondary to ICH can develop hydrocephalus caused by obstruction of the third and fourth ventricles by ventricular clots.[105]
Many studies have shown involvement of the third and fourth ventricles in patients with IVH after ICH to be predictive of adverse outcome and high mortality. [9] , [10] , [11] , [105] In a subanalysis of CT images from the STICH trial, the presence of hydrocephalus reduced the likelihood of positive outcome from 15·1% to 11·5%.[10]
Acute hydrocephalus is more common in patients with high IVH volume (Graeb score ≥6) than it is in patients with low to moderate IVH volume (Graeb score ≤6).[101] Hydrocephalus occurs more with thalamic than with putaminal haemorrhages and is almost absent in the case of lobar haemorrhages, probably because small thalamic haemorrhages can easily compress the cerebral aqueduct leading to obstruction to normal CSF flow and subsequent hydrocephalus, whereas small ganglionic haemorrhages rarely have any effect on ventricular size. [98] , [105]Communicating hydrocephalus can develop with impairment of the Pacchioni granulations by ventricular haemorrhage. [97] , [106] , [107] Increased ICP can result from large volume ICH, especially in the presence of IVH with hydrocephalus, leading to further clinical deterioration and poor outcome. [9] , [10] , [11]
Management
The aim of treatment in ICH patients with IVH and hydrocephalus is to evacuate the intraventricular haematoma, thus relieving the obstruction to CSF flow, reversing ventricular dilatation, and restoring normal intracerebral pressure (table 4). Insertion of external ventricular drainage (EVD) can be a life-saving procedure, relieving acute hydrocephalus and subsequent herniation in ICH patients with severe IVH and AOH. [97] , [100] , [108] EVD is recommended for the treatment of hydrocephalus in patients with decreased consciousness.[45]
| Description | Level of evidence[*] | |
|---|---|---|
| External ventricular drainage | EVD insertion has been suggested as a life-saving procedure in patients with ICH with severe IVH and AOH through reduction of ICP and subsequent herniation; [97] , [100] , [108] however, the complications of EVD include occlusion by blood clots leading to inadequate CSF drainage, [97] , [100] infections leading to frequent EVD exchange, [96] , [107] , [109] and CH caused by impairment of the Pacchioni granulations by IVH[107] | Level 2b C |
| AHA/ASA guidelines recommend that: | ||
| Patients with a GCS score of 8, those with clinical evidence of transtentorial herniation, or those with substantial IVH or hydrocephalus should be considered for ICP monitoring and treatment; treatment should aim to maintain a CPP of 50–70 mm Hg, depending on the status of cerebral autoregulation[45] | Level 2b C | |
| Ventricular drainage as treatment for hydrocephalus should be considered in patients with a decreased level of consciousness[45] | Level 2a B | |
| Intraventricular fibrinolysis | Compared with ventriculostomy alone, IVF with either urokinase [106] , [109] or alteplase [110], [111] , [112] has been shown to promote early and effective clearance of blood in the ventricles, maintain EVD functionality, and reduce the need for permanent shunts as well as improve clinical outcome and reduce mortality; the benefits of IVF might be offset by complications such as secondary haemorrhage and EVD infection [111] , [113] , [114] | Level 2b B |
| IVF can reduce the mortality rate from a range of 60–90% to only 5%;[112] two systematic reviews of clinical studies recorded a 30–50% reduction in mortality after IVF [115] , [116] | Level 2a B | |
| Improved short-term outcome in patients with EVD and IVF can improve 30 day[106] and 90 day outcome,[111] but not long-term outcome after 12 months[110] | Level 2B | |
| In the CLEAR IVH phase 2 trial, designed to assess the safety of open-label doses of intraventricular alteplase in 52 patients with IVH, one dose of 1 mg alteplase and a dosing interval of 8 h was shown to be the most appropriate dosing for IVH,[117] suggesting that low-dose alteplase can be safely given to treat stable intraventricular clots and can increase the lysis rates; symptomatic bleeding occurred in 4% and bacterial ventriculitis in 2% of patients with a 30 day mortality rate of 17%[117] | Level 3B | |
| AHA/ASA guidelines do not recommended the routine use of IVF in clinical practice[45] | Level 2bB | |
| Endoscopic surgical evacuation | Evidence for the potential beneficial effect of neuroendoscopic surgical removal of IVH comes from observational studies: [118] , [119] , [120] , [121] in one study, 24 of 25 patients who had ESE of IVH and obstructive hydrocephalus had resolution of hydrocephalus;[120] similarly, in another study of 17 patients with IVH and hydrocephalus treated with ESE, all patients had successful resolution of hydrocephalus with good outcomes;[121] and in a non-randomised comparison study, patients treated with ESE of IVH had a higher rate of good recovery at 2 months than did those treated with EVD[119] | Level 3B |
| Lumbar drainage | LD is an alternative option for extracorporeal CSF drainage in patients with CH and has been shown to be a simple, safe, and less invasive procedure; it avoids the need for EVD exchange and can substantially reduce the incidence of permanent hydrocephalus and the need for shunt surgery; it also has a lower complication profile than does EVD [97] , [107] , [122] | .. |
| In a prospective case series of three patients, the combination of IVF and LD was shown to be a simple and promising alternative for the treatment of CH after ICH and IVH[122] | Level 1C | |
| In a retrospective analysis of 16 patients with persisting CH after secondary IVH who received an EVD and concurrent LD compared with 39 historical patients treated with EVD alone, LD replaced the need for repeated EVD exchanges, extending the duration of extracorporeal CSF drainage (16 days EVD vs 21 days EVD plus LD) and eventually reducing the need for a permanent ventriculoperitoneal shunt (18·75% vs 33%; p<0 a="" href="http://www.mdconsult.com/das/article/body/402007529-5/jorg=journal&source=&sp=24859238&sid=0/N/944818/1.html?issn=1474-4422#r11702642107" in="" ld-treated="" name="xref_r11702642107" patients="" style="color: #0560a6; margin: 0px; padding: 0px; text-decoration: initial;">[107] |
AHA/ASA=American Heart Association/American Stroke Association. AOH=acute obstructive hydrocephalus. CH=communicating hydrocephalus. CLEAR IVH=Clot Lysis: Evaluating Accelerated Resolution of Intraventricular Hemorrhage. CPP=cerebral perfusion pressure. ESE=endoscopic surgical evacuation. EVD=external ventricular drainage. GCS=Glasgow Coma Scale. ICH=intracerebral haemorrhage. ICP=intracranial pressure. IVF=intraventricular fibrinolysis. IVH=intraventricular extension of haemorrhage. LD=lumbar drainage.
|
| * | The level of evidence is according to the Oxford Centre for Evidence-based Medicine (Level 1A being the highest level of evidence).[64] |
Because EVD is often occluded by blood clots, leading to inadequate CSF drainage, thrombolytics can be instilled into the ventricles to maintain EVD functionality and promote fast clearance of the ventricles. [106] , [109] , [110] ,[111] , [112] , [113] , [114] , [115] , [116] , [117] Findings from the CLEAR IVH (Clot Lysis: Evaluating Accelerated Resolution of Intraventricular Hemorrhage) phase 2 trial[117] suggest that low-dose alteplase can be safe and might increase the lysis rate. However, pending the results of the ongoing CLEAR phase 3 trial (NCT00784134), routine clinical use of intraventricular fibrinolysis is not recommended.[45] Other treatment options include endoscopic surgical evacuation of the haemorrhage [118] , [119] , [120] and the less-invasive lumbar drainage. [97] , [107] , [122]
Seizures and epilepsy), routine clinical use of intraventricular fibrinolysis is not recommended.[45] Other treatment options include endoscopic surgical evacuation of the haemorrhage [118] , [119] , [120] and the less-invasive lumbar drainage. [97] , [107] , [122]
Seizures are a frequent complication of ICH and can even be the presenting symptom.[123] Seizures most commonly occur at the onset of ICH, but can be delayed. About 50–70% of seizures occur within the first 24 h, and 90% in the first 3 days, [124] , [125] , [126] , [127] with an overall 30 day risk of seizures of about 8%.[123]
Early seizures are defined as those occurring within 2 weeks of initial ictus, late seizures occurring thereafter. [124] ,[125] Whereas early seizures are thought to be caused by structural disruption and cellular biochemical dysfunction, late seizures are attributed to gliosis and the development of meningocerebral cicatrices. [124] , [125] Early seizures can be predictive of epilepsy development.[128]
The incidence of seizures after ICH varies widely depending on study design, diagnostic criteria, duration of follow-up, and the population studied. Similarly, the true incidence of seizures might be underestimated, because subclinical seizures can be detected with use of continuous EEG (cEEG) monitoring only. Overall incidence after ICH is 4·2–20% [123] , [124] , [127] , [129] , [130] for clinical seizures and 29–31% for subclinical seizures. [125] , [126] , [131]In a series of 63 patients with ICH in an intensive care unit (ICU) who had cEEG monitoring within 72 h of admission, non-convulsive seizures were detected in 28%, four times the incidence of observable seizures. The presence of seizures was independently associated with increased midline brain shift on 48–72 h follow-up head CT scans, with neurological deterioration.[126]
In a retrospective review[125] of 102 patients with ICH with cEEG monitoring, 19% had convulsive seizures, 13% had electrographic seizures, and 5% had both, with 94% of the seizures detected in the first 72 h. Subclinical seizures were associated with expanding haemorrhages, especially if they expanded by more than 30% in the first 24 h or if they reached the cortex, and were also associated with a poor outcome. Electrographic seizures were twice as common (31% vs 14%) in patients with expanding haemorrhages.[125]
The reported frequency of status epilepticus is 0·3–21·4%. [130] , [132] , [133] In an earlier study of 1402 patients with ICH, status epilepticus occurred in 11 of 65 patients with seizures and was the initial presentation of ICH in six of these 65 patients.[130] Epilepsy (recurrent seizures) has been reported to develop in 2·5–4% of patients, [124] ,[130] but Passero and colleagues[123] reported the risk of late seizures or epilepsy in survivors of ICH to be 5–27%. Recurrent seizures occurred in four of 14 patients in one study, despite the provision of antiepileptic treatment.[133]
Although the cause of seizures in patients with ICH is unclear, the combination of sudden development of a space-occupying lesion with mass effect, focal ischaemia, and blood breakdown products has been postulated to account for seizures early in ICH.[124] Early-onset seizures are thought to be directly related to the insult of ICH to the brain.[127]
Identified predisposing factors include haemorrhagic size, the presence of hydrocephalus, intracranial midline shift, low GCS, and severe neurological deficit.[124] Lobar location is an independent predictor of early seizures.[123] Non-occipital lobar haemorrhages, [125] , [126] , [127] , [133] as well as subcortical haemorrhages,[126] are commonly associated with seizures.
Conflicting results on the association of seizures after ICH with clinical outcome and mortality have been reported. In one study,[134] the in-hospital mortality rate was 37·9% in patients with both acute ICH and IS with seizures within 48 h of symptom onset compared with 14·4% for patients without seizures. Another study[128] of ICH and IS showed an increased risk of mortality within 30 days if seizures developed within the first 24 h after stroke (32·1% vs 13·3%). Although mortality rates were higher in patients with seizures, seizures were not an independent predictor of mortality at 30 days or of poor outcome post stroke after adjusting for other factors, although the risk in patients with ICH was not independently determined.[128]
Findings from one study[126] showed that post-ICH seizures are associated with worsening neurological function as measured with the NIHSS (National Institutes of Health Stroke Scale), and recorded a trend towards increased poor outcome, with a mortality rate of 27·8% compared with 15% in those without seizures.[126] However, other studies have recorded no substantial difference in mortality between patients with or without seizures after stroke,[124] and, surprisingly, there was an association with better outcomes after stroke as measured with Scandinavian stroke scales.[135]
Management
No randomised controlled trials (RCTs) have been done to guide decisions on seizure prophylaxis or treatment in patients with ICH. Similarly, no definitive evidence or clear guidelines exist for the choice of treatment or duration of treatment for patients with one or more seizures or status epilepticus (table 5). However, available guidelines recommend that patients with clinical seizures should be treated with antiepileptic drugs (AEDs).[45] The choice of initial AED is dependent on the individual characteristics of each patient, including medical comorbidities, concurrent drugs, and contraindications. The benefit of seizure prophylaxis after ICH is controversial. Whereas previous studies have advocated the use of prophylaxis in most patients, [123] , [136] two observational studies have since shown prophylactic therapy with AEDs to be associated with poor outcome. [136] , [137] Whereas previous guidelines [2] ,[64] have recommended a 30 day course of prophylactic AED in patients with lobar haemorrhage or those who have had seizures, the most up-to-date guideline recommends against prophylactic use.[45]
| Description | Level of evidence[*] | |
|---|---|---|
| General considerations | No randomised controlled trials have been done to guide decisions on seizure prophylaxis or treatment in patients with ICH; however, AHA/ASA guidelines[45] recommend the following: | .. |
| Patients with clinical seizures should be treated with AEDs | Level 1A | |
| Patients with a change in mental status who have electrographic seizures on EEG should be treated with AEDs | Level 1C | |
| EEG monitoring is indicated in ICH patients with depressed mental status that is out of proportion with the extent of brain injury | Level 2a B | |
| Antiepileptic drugs | The choice of initial AED depends on individual circumstance and contraindications; the AEDs to be considered for the acute treatment of post-haemorrhagic stroke seizures include intravenous lorazepam (0·05–0·10 mg/kg) followed by a loading dose of phenytoin or fosphenytoin (15–20 mg/kg), valproic acid (15–45 mg/kg), levetiracetam (500–1500 mg), or phenobarbital (15–20 mg/kg) | .. |
| Prophylaxis | The benefit of seizure prophylaxis after ICH is controversial: some investigators have advocated the use of prophylaxis in most patients[126] and previous guidelines [2] , [70] have recommended a 30 day course of prophylactic AEDs in patients with lobar haemorrhage or those who have had seizures, on the basis of risk reduction of seizures in patients with lobar haemorrhage reported in observational studies[123] | .. |
| However, two observational studies have since shown prophylactic therapy with AEDs to be associated with poor outcome: [136] , [137] in a prospective study of 98 patients with ICH, prophylactic phenytoin was associated with higher fever and worse outcome at 14 days or at discharge, and worse functional outcome at 14 day, 28 day, and 3 month follow-up;[136]similarly, in another trial, an association was recorded between AED use and worse 3 month functional outcome[137] | Level 3B | |
| In the absence of data for patients with ICH showing a benefit of seizure prophylaxis, the most up-to-date AHA/ASA guidelines recommend against prophylactic use of AEDs[45] | Level 3B |
AEDs=antiepileptic drugs. AHA/ASA=American Heart Association/American Stroke Association. ICH=intracerebral haemorrhage.
|
| * | The level of evidence is according to the Oxford Centre for Evidence-based Medicine (Level 1A being the highest level of evidence).[64] |
VTE are a common and potentially fatal complication in patients with ICH that can manifest as either deep-vein thrombosis (DVT) or pulmonary embolism (PE). [138] , [139] The reported rate of clinically symptomatic VTE is 3–7%. [138] , [140] , [141] Overall, studies have estimated the short-term post-ICH risk of PE to be 1–2% and of DVT to be 1–4%.[138] Subclinical DVT is more common than is clinically apparent DVT, with asymptomatic DVT rates occurring in up to 17% of patients with ICH.[142] In a small prospective study of 52 patients with acute ICH, DVT was detected in 40% of patients within 2 weeks, with one case of PE recorded.[143] The high rate was attributed to the extent of paralysis due to severe stroke. In the rFVIIa trial,[8] about 2% of patients in both the placebo and treatment groups developed DVT, and about 2% developed PE.[8]
DVT is a risk factor for PE. PE generally arises from venous thrombi that develop in a paralysed lower leg or pelvis.[144] However, about 30% of patients diagnosed with acute PE show no evidence of DVT in their lower legs, implying that a negative venous duplex ultrasound does not exclude the diagnosis of acute PE.[145]
Independent risk factors associated with the development of VTE in patients with ICH include severe stroke, lengthy immobilisation, advanced age, and increased prothrombotic activity. [143] , [146] Discontinuation of antithrombotic agents after ICH could also accelerate the formation of DVT.[143]
Although a high D-dimer value is a useful predictor of DVT formation in patients with ICH, this value could also be increased because of the presence of infection or hyperfibrinolysis caused by the cessation of anticoagulation.[143]
In a retrospective study of symptomatic VTE in patients with ICH, the most commonly identified risk factors were age (>40 years), immobility due to paresis or restrictions for mechanical ventilation, presumed infection, and the presence of an indwelling central venous catheter.[141]
A systematic review detected racial disparity in the incidence of VTE after ICH, with a higher rate of DVT in black patients than in white patients after adjusting for differences in the risk factors for ICH.[147] A sex difference also exists, with women being at greater risk of VTE. [139] , [148]
Management
A clinical dilemma can arise as to the best possible approach to VTE prevention in patients with ICH, because anticoagulants can increase the risk of HE or rebleeding. Although this decision is often made on the basis of a risk–benefit analysis in the context of the individual patient, prophylaxis for VTE in patients with ICH has been addressed in some guidelines. [2] , [149] Options for reducing the risk of VTE after ICH include intermittent pneumatic compression, low-dose subcutaneous low-molecular-weight heparin or unfractionated heparin (table 6). [2] , [45]
Table 6 -- Prevention and clinical management of venous thromboembolic events after intracerebral haemorrhage
| Description | Level of evidence[*] | |
|---|---|---|
| Prevention | ||
| Non-pharmacological approaches | On the basis of findings from a randomised controlled trial that showed intermittent pneumatic compression combined with elastic stockings to be better than elastic stockings alone in the reduction of the rate of asymptomatic DVT after ICH (4·7% vs 15·9%)[142] and the finding that graduated compression stockings alone are ineffective in preventing DVT,[150] AHA/ASA guidelines recommend that patients with ICH should have intermittent pneumatic compression for prevention of venous thromboembolism in addition to elastic stockings [2] , [45] | Level 1B Level 1B |
| Pharmacological approaches | Randomised trials, [144] , [151] a non-randomised study,[152] and a retrospective study[140] have shown pharmacological prophylaxis with LMWH for DVT and PE prevention 24–48 h after ICH to be safe with no increased risk of haematoma expansion or further bleeding, although LMWH has not been shown to be better than elastic stockings | Level 2b B |
| On the basis of these studies, AHA/ASA guidelines suggest that after cessation of bleeding, low-dose subcutaneous LMWH or unfractionated heparin can be considered for prevention of venous thromboembolism in patients who are immobile after 1–4 days from onset of ICH[45] | Level 2b B | |
| Treatment | ||
| IVC filters | Insertion of an IVC filter is an option for patients who cannot receive therapeutic anticoagulation or for those who have to wait for several weeks before starting anticoagulation; [153] , [154] , [155] , [156] IVC filters prevent PE by trapping most of the large emboli originating from the deep veins of the pelvis and feet,[157] but the benefits can by counterbalanced by complications such as venous thromboembolism and caval occlusion, with no difference in mortality [153] , [154] , [155] , [156] | .. |
| A retrospective review of 371 mixed stroke patients (105 [28%] with ICH) who received an IVC filter recorded a 16% incidence of symptomatic post-filter DVT, a 0·8% incidence of post-filter fatal PE, and a 5·1% incidence of caval occlusion; other filter-related complications included fracture of filter, penetration of caval wall, and filter migration[156] | Level 2b C | |
| No randomised controlled trial has compared vena cava filters with anticoagulation in patients with ICH or ischaemic stroke; nonetheless, in the PREPIC randomised controlled trial of 400 patients with documented proximal DVT or PE who received concurrent anticoagulation, IVC filters reduced the risk of PE, even after 8 years, although there was an associated long-term risk of recurrent DVT but no statistically significant reduction in mortality, probably because of the older study population, and most deaths were due to cancer or cardiovascular diseases[158] | Level 1B | |
| Similarly, an earlier randomised study comparing the effectiveness of a combination of anticoagulation and IVC filters versus anticoagulation alone in patients with proximal DVT showed a statistically significant decrease in the incidence of PE in the filter group (1·1%vs 4·8%); there was, however, a subsequent excess of recurrent DVT, without any difference in mortality[153] | Level 1B | |
| The AHA/ASA and international guidelines recommend that patients with ICH who develop an acute proximal DVT, particularly those with clinical or subclinical PE, should be considered for acute placement of a vena cava filter [2] , [149] | Level 2b C | |
| Anticoagulation | The AHA/ASA guidelines recommend that the decision to start antithrombotic therapy several weeks after the placement of a vena cava filter must be made on the basis of a risk–benefit analysis of the potential for recurrent bleeding (amyloid-related ICH has higher risk of recurrent ICH than does hypertension-related ICH), as well as associated disorders with increased arterial thrombotic risk (eg, atrial fibrillation), and the overall health and mobility of the patient[2] | Level 2b B |
| Surgical embolectomy | The evidence for the benefit of surgical embolectomy as a life-saving procedure for PE as a complication of ICH comes from case series: in one series of three consecutive cases of pulmonary embolectomy in patients with ICH with massive PE, the interval between the onset of intracranial bleeding and emergency surgical embolectomy was 7–16 days, and all three patients survived without any neurological exacerbation;[159] however, the problem associated with surgical embolectomy is the possible exacerbation of intracranial bleeding from systemic heparinisation | Level 3C |
AHA/ASA=American Heart Association/American Stroke Association. DVT=deep-vein thrombosis. ICH=intracerebral haemorrhage. IVC=inferior vena cava. LMWH=low-molecular-weight heparin. PE=pulmonary embolism. PREPIC=Prévention du Risque d'Embolie Pulmonaire par Interruption Cave.
|
| * | The level of evidence is according to the Oxford Centre for Evidence-based Medicine (Level 1A being the highest level of evidence).[64] |
An even greater clinical therapeutic dilemma is the management of patients with ICH who subsequently develop VTE. If untreated, the risk of recurrent fatal PE is increased and, if treated, patients are at increased risk of bleeding. Anticoagulation has been estimated to double the risk of recurrent ICH compared with the overall recurrence risk of ICH, and the mortality rate associated with recurrent ICH can be as high as 50%.[160] The risk of recurrent ICH can depend on factors such as the patient's age and the location of ICH. Patients with lobar haemorrhages are at higher risk of rebleeding, probably due to suspected amyloid angiopathy.[161]
Insertion of a vena cava filter is recommended in patients with ICH who develop an acute proximal venous thrombosis. The decision to start antithrombotic treatment several weeks after the vena cava filter should be made on the basis of a risk–benefit analysis of the potential for rebleeding as well as the patient's comorbidities and mobility.[2]
Fever
Fever occurs in up to 40% of patients after ICH, [162] , [163] and is independently associated with a poor outcome and increased mortality. [163] , [164] , [165] , [166] , [167] The incidence of fever (temperature >38·3°C) is high in basal ganglionic and lobar ICH, especially in patients with IVH. [45] , [162] , [163] The cause of high temperature after stroke is not always apparent, although increased body temperature might be a direct consequence of brain damage caused by stroke or accompanying infections. [164] , [168] Increased body temperature could result from damage to the thermostatic centre after hypothalamic stroke[164]—patients with IVH are thought to have more of a neurogenic or central fever.[166]
A high body temperature after ICH is associated with HE, cerebral oedema, increased ICP, and END. [45] , [169]Fever after ICH is associated with longer ICU and hospital stays, poor functional outcome, and increased mortality.[167] A prospective study of 390 mixed ischaemic and intracerebral haemorrhagic stroke patients (9% with ICH) admitted within 6 h of stroke onset showed that the relative risk of poor outcome increased by 2·2 times and that mortality increased by a factor of 1·8 for each 1°C increase in baseline body temperature.[166]
In a subanalysis of data from the PAIS (Paracetamol [Acetaminophen] In Stroke) trial[168] of 1332 IS and ICH patients admitted within 12 h of stroke onset, 10% of 163 ICH patients had admission body temperatures of more than 37·5°C. An early rise in body temperature rather than high body temperature on admission was the greatest risk factor for adverse outcome.[164]
In a study of 251 cases of ICH, initial body temperature was not an independent prognostic factor, but an increase in body temperature during the first 72 h, which occurred in 91% of patients, was associated with poor clinical outcome. Also, for those patients surviving the first 72 h after hospital admission, a longer duration of fever was associated with a worse outcome.[163]
In a retrospective study of 330 patients with acute IS and ICH, of the 37·6% of 330 patients who had fever, 22·7% had a documented infection and 14·8% had fever without a documented infection.[170] Reith and colleagues[166]detected infection in a fifth of the 25% of mixed IS and ICH pyrexial patients (body temperature >37·5°C on admission within 6 h of stroke onset). Pulmonary and urinary infections are the main causes of infectious fever in patients with stroke. [162] , [163] , [168] , [171] In the PAIS study, most infectious fevers were due to pneumonia and urinary tract infections (UTIs).[168] In the review with only ICH patients, pulmonary infection was diagnosed within the first 72 h in 84 patients (43% of 196 patients) and UTI was diagnosed within the same time in 69 patients (35% of 196 patients).[163]
Management
ICH patients with a high temperature should be physically examined and investigated to establish the cause of fever and possibly the source of infection. Generally, treatment with antipyretics and cooling blankets is used[169] for patients with a sustained fever of more than 38·3°C (panel). Although new adhesive surface-cooling systems and endovascular heat-exchange catheters might prove more effective, [171] , [173] they have not been systematically investigated in patients with ICH. No evidence is available from RCTs linking fever treatment with improved clinical outcome or reduced mortality. A Cochrane review showed no statistically significant effect of pharmacological or physical temperature-lowering therapy in reducing the risk of death or dependency.[172] In the PAIS RCT,[168] no difference was seen between the placebo group and the group treated with paracetamol in patients with IS and ICH within 12 h of symptom onset. However a post-hoc analysis of patients with a baseline body temperature of 37–39°C who were treated with paracetamol showed improved outcome. Sources of infectious fever should be treated with appropriate antibiotics, and antipyretics should be given to lower temperature in ICH patients with fever.[2]
| Clinical management of fever and infection after intracerebral haemorrhage |
The goal should be to maintain normothermia and treat infections in patients with ICH. Patients with ICH and sustained fever in excess of 38·3°C should be treated with antipyretics and cooling blankets [163] , [169] (Level 1C). However, there is no evidence from randomised controlled trials to link fever treatment with improved clinical outcome or reduced mortality. A Cochrane review of five pharmacological temperature-lowering trials and three physical cooling trials with 423 participants recorded no statistically significant effect of pharmacological or physical temperature-lowering therapy in reducing the risk of dependency (odds ratio 0·9, 95% CI 0·6–1·4) or death (0·9, 0·5–1·5)[172] (Level 1C). Equally, an updated meta-analysis of a Cochrane review with six pharmacological temperature-lowering therapies did not show any statistically significant difference between the treated and placebo groups in the proportion of patients who were alive and independent (modified Rankin score ≤2) at final follow-up (1·1, 0·9–1·3)[168] (Level 1C). In PAIS, a randomised controlled trial, treatment with high-dose acetaminophen (6 g daily) in 1400 patients with ischaemic stroke and ICH within 12 h of symptom onset was associated with a 0·26°C (95% CI 0·18–0·31) reduction in mean body temperature measured 24 h after admission; the trial did not provide sufficient evidence to lend support to routine use of high-dose paracetamol in patients with acute stroke. However, in a post-hoc analysis of patients with a baseline body temperature of 37–39°C treated with paracetamol, a 9% absolute increase was recorded in the number of patients with improved outcome, with a number needed to treat of 11; serious adverse events occurred in 8% of the paracetamol group versus 10% of the placebo group (Level 1B). Although no evidence from randomised controlled trials is available to lend support to the routine use of physical or pharmacological strategies to reduce temperature in patients with acute stroke, the AHA/ASA guidelines recommend that the sources of fever should be treated with appropriate antibiotics and that antipyretic drugs should be given to lower temperature in febrile patients with ICH (Level 1C).[2]
AHA/ASA=American Heart Association/American Stroke Association. ICH=intracerebral haemorrhage. PAIS=Paracetamol (Acetaminophen) In Stroke. The level of evidence is according to the Oxford Centre for Evidence-based Medicine (Level 1A being the highest level of evidence).[64]
|
A high proportion of patients (about 60%) might develop hyperglycaemia even in the absence of a previous history of diabetes after ICH.[174] Increased blood glucose in the acute setting of ICH is probably a response to stress and severity of ICH[13] and can persist for up to 72 h after ICH.[174]
Many studies have shown that increased serum glucose on admission is associated with larger haematoma size, HE, PHE, cell death, and increased risk of poor outcome, [12] , [32] , [174] , [175] , [176] and that it is a potent predictor of 30 day mortality in both diabetic and non-diabetic patients with ICH[13] as well as an independent predictor of early mortality and worse functional outcome in non-diabetic patients with ICH. [175] , [176]
In the post-hoc analysis of the ATACH trial of 60 patients, the risk of poor outcome (mRS 4–6) in patients with persistent or increasing serum glucose concentrations was more than twice that of patients with a reduction in serum glucose concentrations (relative risk [RR]=2·64, 95% CI 1·03–6·75). Also, the RRs were 2·59 (1·27–5·30) for HE of more than 33%, and 1·25 (0·73–2·13) for a relative oedema expansion of more than 40%.[177]
Management
Poor outcomes associated with hyperglycaemia could be avoided or minimised through monitoring glucose and maintaining normoglycaemia (table 7). Control of hyperglycaemia in the acute setting of ICH decreases the likelihood of HE, PHE, ICP, and seizures, and improve outcomes, [12] , [177] although the optimum glucose target and duration of glucose control in patients with ICH are unclear. However, studies have suggested targeting a glucose concentration of 150 mg/dL in the acute setting of ICH, [12] , [174] and the avoidance of hyperglycaemia during the first 72 h after symptom onset, possibly because admission hyperglycaemia can persist for at least 72 h post ictus.[174] , [177]
| Description | Level of evidence[*] | |
|---|---|---|
| General considerations | Data from the post-hoc analysis of the ATACH trial has linked a decrease in serum glucose concentration to a reduction in haematoma expansion and improved clinical outcome[177] | Level 1B |
| Optimum glucose target | The optimum glucose target is unclear and various glucose concentrations from 127 mg/dL to 180 mg/dL have been suggested; some prospective studies have suggested a target glucose concentration of 150 mg/dL in the acute setting of ICH [12] , [174] | Level 1B |
| In the Oxfordshire Community Stroke Project, in 645 patients (86%) with ischaemic stroke and 105 (14%) with haemorrhagic stroke, a plasma glucose concentration higher than 144 mg/dL (>8 mmol/L) was associated with increased mortality, particularly in the first month after stroke[178] | Level 2B | |
| In the Acute Brain Bleeding Analysis, a prospective study of 1387 patients with ICH, an admission glucose concentration of >167 mg/dL was shown to be an independent risk factor for early mortality[176] | Level 1B | |
| Optimum duration of glucose control | The optimum duration of glucose control is unclear, although studies have suggested the avoidance of hyperglycaemia during the first 72 h of symptom onset, possibly because admission hyperglycaemia can persist for up to 72 h post ictus [174] , [177] | .. |
| In a post-hoc analysis of the ATACH trial, patients who had serum glucose reduction during the first 72 h after ICH had a lower relative risk of death or disability at 3 months[177] | Level 1B | |
| In a prospective observational study of 295 patients, individuals with a serum glucose concentration maintained within 60–150 mg/dL during the first 72 h after ICH did better than those with a persistently high serum glucose concentration[174] | Level 1B | |
| In GIST-UK, which had a shorter duration than did ATACH, 24 h of intensive glucose control did not reduce mortality or improve functional outcome in the treatment group,[179] although the trial had a small number of patients with ICH (114 of the total 933) randomly allocated to treatment within 24 h of stroke onset; no separate analysis for patients with ICH was reported | Level 2B | |
| AHA/ASA guidelines suggest that serum glucose concentrations of >140 mg/dL (>7·8 mmol/L) should be treated with insulin [2] , [45] | Level 2a C |
AHA/ASA=American Heart Association/American Stroke Association. ATACH=Antihypertensive Treatment of Acute Cerebral Hemorrhage. GIST-UK=UK Glucose Insulin in Stroke Trial. ICH=intracerebral haemorrhage.
|
| * | The level of evidence is according to the Oxford Centre for Evidence-based Medicine (Level 1A being the highest level of evidence).[64] |
Increased blood pressure
Increased BP (≥140/90 mm Hg) is common in the acute phase of ICH, occurring in more than 70% of patients on presentation. [180] , [181] Increased BP in the acute setting of ICH occurs even in the absence of a previous history of hypertension and is independently associated with poor outcome. [180] , [182]
The mechanism for the acute increase of BP after ICH is unknown. However, it is proposed to be a multifactorial process that includes activation of the neuroendocrine systems (sympathetic nervous system, glucocorticoid system, or the renin–angiotensin axis), increased cardiac output, and stress response to conditions such as increased ICP, headache, and urinary retention. [45] , [180] , [182]
Many studies have shown an association between increased BP in the acute phase of ICH and HE, PHE, and rebleeding. [19] , [36] , [37] , [90] Increased BP in the acute setting of ICH is associated with worse outcome and increased mortality. [180] , [182] , [183] In a large study of 3930 mixed stroke patients (1760 with ICH), the in-hospital case-fatality rate was 5·9% for patients with ICH. Also, both SBP and diastolic BP were associated with odds of death or disability in such patients.[183]
However, the effect of BP levels on mortality seems to conform to a U-shaped distribution after ICH, as poorer outcomes have also been seen at very low SBP (<120 and="" at="" hg="" high="" mm="" very="">220 mm Hg) SBP values.[184]
Management
The poor outcome associated with increased BP after ICH could be minimised with BP monitoring and treatment aimed at maximisation of cerebral perfusion while minimising ongoing bleeding (table 8). The rationale for lowering BP after ICH is because SBP reduction can reduce the rate of HE and possibly improve clinical outcome.
| Description | Level of evidence[*] | |
|---|---|---|
| Blood pressure control | The INTERACT pilot trial that compared intensive (target SBP 140 mm Hg) with guideline-based (target SBP 180 mm Hg) BP control showed the feasibility and safety of early intensive BP reduction within the first 6 h of onset of supratentorial ICH; however, no statistically significant difference was recorded in functional outcome at 3 months between the two groups (median mRS score was 2 in both groups; p=0·66)[20] | Level 1B |
| In the ATACH phase 1 trial, patients with a SBP of >170 mm Hg were treated with intravenous nicardipine within 6 h of onset of lobar supratentorial ICH to reduce BP in three escalating target BP tiers: tier 1 (170–200 mm Hg; n=18), tier 2 (140–170 mm Hg; n=20), and tier 3 (110–140 mm Hg; n=22); BP reduction was safe and feasible in all three tiers[47] | Level 1B | |
| AHA/ASA guidelines recommend that in patients presenting with a SBP of 150–220 mm Hg, acute lowering of SBP to 140 mm Hg is probably safe[45] | Level 2a B | |
| Target blood pressure | The following target BP should be considered in various situations: [2] , [45] | |
| If SBP is >200 mm Hg or MAP is >150 mm Hg, aggressive reduction of BP could be considered with continuous intravenous infusion, with frequent BP monitoring every 5 min | Level 2b C | |
| If SBP is >180 mm Hg or MAP is >130 mm Hg and there is the possibility of increased ICP, then consider monitoring ICP and reducing BP using intermittent or continuous intravenous drugs while maintaining a cerebral perfusion pressure of 60 mm Hg | ||
| If SBP is >180 mm Hg or MAP is >130 mm Hg and there is no evidence of increased ICP, then consider a slight reduction of BP (eg, MAP of 110 mm Hg or target BP of 160/90 mm Hg), using intermittent or continuous intravenous drugs to control BP, and clinically re-examine the patient every 15 min | ||
| If SBP exceeds 180 mm Hg or MAP exceeds 130 mm Hg, management should involve either bolus or continuous infusion of drugs to a target MAP of 110 mm Hg or less or a goal BP of 160/90 mm Hg | Level 2b C | |
| Antihypertensive drugs recommended in AHA/ASA guidelines | Suitable antihypertensive drugs include labetalol, esmolol, or nicardipine (drugs with a cerebral vasodilation effect should be avoided)[2] Labetalol: 5–20 mg (intermittent boluses) every 15 min or 2 mg/min in continuous infusion (drip) Esmolol: 250 mcg/kg (loading dose), then 25–300 mcg/kg per min (maintenance) Nicardipine: 5–15 mg/h Hydralazine: 5–20 mg (intermittent boluses) Nitroprusside: 0·1–10 mcg/kg per min | Level 2bC |
AHA/ASA=American Heart Association/American Stroke Association. ATACH=Antihypertensive Treatment of Acute Cerebral Haemorrhage. BP=blood pressure. ICH=intracerebral haemorrhage. ICP=intracranial pressure. INTERACT=Intensive Blood Pressure Reduction in Acute Cerebral Haemorrhage Trial. MAP=mean arterial pressure. mRS=modified Rankin score. SBP=systolic blood pressure.
|
| * | The level of evidence is according to the Oxford Centre for Evidence-based Medicine (Level 1A being the highest level of evidence).[64] |
The INTERACT trial showed that a reduction of SBP to 140 mm Hg is safe and might reduce the risk of HE in patients treated within 6 h of onset of ICH, but had no effect on outcome. [20] , [46] In ATACH, aggressive SBP reduction to 110–140 mm Hg in the first 24 h was well tolerated, with a low risk of HE, neurological deterioration, and in-hospital mortality.[47]
The results of both INTERACT 2 and ATACH 2 are awaited, and might further answer the question of whether there are other benefits of early SBP reduction in patients with ICH, and provide clearer evidence for the target BP and the choice of drugs. Pending these results, the AHA/ASA guidelines recommend BP lowering if SBP is 150–220 mm Hg or MAP is higher than 150 mm Hg, and that acute lowering of SBP to 140 mm Hg is probably safe.[45] Meanwhile, the EUSI recommends a target of 160/100 mm Hg or a MAP of 125 mm Hg in patients with a history of hypertension and 150/90 mm Hg or a MAP of 110 mm Hg in those without a history of hypertension.[64]
Conclusions
Complications occurring during the acute phase of ICH add further detrimental effects to an already potentially fatal disorder and substantially affect the clinical outcome. In view of the small number of therapeutic options available for ICH, a need exists to provide the best evidence-based supportive care. Patients should be managed in a neuroscience ICU or a similar setting during the acute phase of ICH, with continuous monitoring of vital signs and frequent neurological assessment. Improved surveillance is needed for the early detection of complications, which should be managed on the basis of current recommendations and guidelines, especially when little evidence exists or when trials are still in progress. Once patients are stabilised, early rehabilitation is needed to prevent or reduce the risk of further complications. The implementation of appropriate preventative and therapeutic interventions could minimise the morbidity and mortality associated with this devastating type of stroke. A clear need exists for further research into the prevention and treatment of ICH complications to improve the level of evidence available, because most guidelines and recommendations are based on empirical data.
No hay comentarios:
Publicar un comentario