Scenario: You are evaluating a 58-year-old male with hypertension and cirrhosis who presents with a sharp, right-sided, severe, non-radiating, constant headache that started suddenly two hours ago while he was walking to get something to eat. Shortly after the headache started, he noticed his vision became blurry and he developed nausea, both of which resolved upon arrival to your ED. He denies taking blood thinners such as aspirin/clopidogrel, trauma, fever, chest pain, vomiting, diarrhea, abdominal pain, and dizziness. As you are speaking to him, you notice that it takes multiple prompts for him to answer some of your questions. His BP is 186/110, but his other vital signs are within normal limits. His neurological exam is unremarkable, but he keeps perseverating on the severity of his headache, despite multiple assurances that he will receive pain medication. You draw his labs, order a head CT then move on to the next patient. When you check back on his imaging you see:
You hear your medical school professors in your head telling you that cerebellar hemorrhages need surgery. You immediately bring the patient over to the critical care area of the ED, page neurosurgery, and transfer care of the patient to your colleagues. Towards the end of your shift, you circle back to the critical care area to follow up on your patient and find that the neurosurgeons said no acute interventions were indicated and recommended admitting him to the medical intensive care unit. Incredulous, you start to wonder, was I taught incorrectly?
Here we will review the evidence behind neurosurgical interventions for spontaneous intracerebral hemorrhage (sICH). This is an area of continued controversy and research.
The European Stroke Organization (ESO) Guidelines, published in 2014, claim there is no evidence to support routine surgical intervention after supratentorial ICH, but with the caveat that early surgery may be valuable in patients with a GCS of 9-12 (1). Let’s see what led them to these recommendations and see if newer evidence can clarify who might benefit from surgery. In 2005, Mendelow et al published the International Surgical Trial in Intracerebral Hemorrhage (STICH), a randomized, controlled trial (RCT) comparing early surgery (n = 468) – within 24 hours of randomization – with medical therapy (n = 497) using an intention to treat analysis and a favorable extended Glasgow outcome scale (GOSE) at 6 months as the primary outcome (mortality was a secondary outcome). To date, this is the largest RCT comparing early surgery to conservative management for sICH. They excluded patients with likely aneurysm, arteriovenous malformation, trauma, tumors, cerebellar hemorrhage, intraventricular hemorrhage (IVH), brainstem involvement, pre-existing disabilities, or severe comorbidities. Most other sICH studies have similar exclusion criteria. 77% of patients who underwent surgery had a craniotomy, with the rest receiving similar proportions of burr holes, endoscopy, stereotactic, or other procedures. In the control group, 140 patients ended up receiving surgery. They found no difference between groups for the primary outcome or mortality rate. It should be noted, that meeting the primary outcome differed based upon initial prognosis. Patients with a poor initial prognosis were classified as meeting the primary outcome with GOSE scores of 4-8, while those with a good initial prognosis could only reach the primary outcome with a score of 5-8. While the groups were similar in overall prognosis at baseline, there were 26 fewer patients with a good prognosis in the early surgery group, which may have inflated the effects of early surgical intervention. A post-hoc subgroup analysis found that those patients with lobar ICH within 1 cm to the cortex may benefit from early surgery, and those with a GCS <9 had poor outcomes regardless of intervention (2).
A meta-analysis of individual patient data from RCTs from 1985 – 2012, including the STICH trial, found significantly improved outcomes with surgery when performed within 8 hours of ictus, or onset of symptoms, with hematoma volumes of 20-50 mL, initial Glasgow Coma Scale (GCS) of 9-12, and patients 50 to 69 years old. They found a trend towards better outcomes for superficial hematomas without intraventricular hemorrhage (IVH). It is important to note that most of these findings were driven in large part by a study in 2009 by Wang et al that included 377 cases with an uncharacteristically high number (45%) of patients with favorable outcomes (4). This multicenter study from China only employed a minimally invasive craniopuncture technique for ICH in a select group of patients with basal ganglia hemorrhage, motor deficits on exam, GCS of at least 9, 40-75 years of age, with hematoma volumes of 25-30 mL (4).
In 2013, the STICH II trial randomized 297 patients to early surgery (within 12 hours of randomization) and 286 patients to conservative treatment in patients with 10-100 mL lobar ICH less than or equal to 1 cm from the cortical surface of the brain. All patients were conscious at randomization. This follow up study also found no significant difference between groups (5). In contrast to STICH, 99% of the intervention group received craniotomy, one patient received craniectomy, three received minimally invasive surgical techniques, and 62 of the control group patients wound up receiving surgery (2,5). 98% of patients were did receive surgery within 12 hours of randomization, however, the mean time from ictus to surgery was 26.7 hours. Nearly two-thirds of patients in this trial had an initial GCS of at least 13 and based on the previous meta-analysis, these are patients who were unlikely to benefit from surgery.
The Minimally Invasive Surgery plus Recombinant Tissue Plasminogen Activator (r-TPA) Trial (MISTIE II) was a RCT that compared 79 patients who received minimally invasive hematoma evacuation with r-TPA to 39 patients who received medical treatment. This trial was set up to establish the safety of hematoma evacuation with r-TPA and to demonstrate reductions in perihematoma edema (PHE). The exact clinical significance of PHE is unclear, but it is associated with poor outcomes. As expected, the intervention group had significantly decreased volume of hematoma and PHE (6). Of note, 10 patients who underwent surgery did not receive r-TPA, and there was no significant difference in hematoma or PHE volumes. MISTIE III, which is currently enrolling, will compare the two groups with clinical outcomes. The aforementioned trials that compared traditional neurosurgical techniques to medical management found no significant difference between groups for their primary outcomes. Trials evaluating newer, minimally invasive techniques, while not as well designed, appear to show a benefit for these techniques over medical therapy.
A recent retrospective review by Li et al of 99 patients with supratentorial lobar sICH volume of at least 30 mL, GCS of 4-14, and surgery within 24 hours of ictus found endoscopic surgery and stereotactic aspiration to have significant advantages over traditional craniotomy. Craniotomy was associated with the longest ICU and hospital length of stay, the highest expenses, the most digestive tract ulcers, and the most perihematomal edema. More importantly, the craniotomy group also had a significantly lower Glasgow Outcome Scale (GOS) score at six months (7), although there was no difference in mortality. Due to the retrospective design, these results must be taken with a grain of salt.
The authors of STICH II also performed a meta-analysis of studies of supratentorial ICH in any location, incorporating their findings, and found a significant advantage to surgery, but with significant heterogeneity between trials. They did not find a significant difference for surgeries in the subgroup of lobar ICH without IVH (5). The data seems to suggest a benefit to early surgery, but the precise timing is unclear; studies suggest benefit if surgery is performed within 8 hours (3) or 21 hours (5). There is possible harm if surgery is done within 4 hours of ictus (1,7).
Bottom line: Due to heterogeneity and conflicting results between trials, it is difficult to draw firm conclusions. Overall, it appears that the European Stroke Organization (ESO) guidelines are correct in that there are no clear benefits to surgical intervention for supratentorial ICH when looking at all patients. However, when you incorporate newer trials and look more critically at older trials, a few trends start to emerge:
1) Timing – early is likely better, ideally between 4-21 hours of symptom onset
2) Technique – minimally invasive surgery may be preferable to traditional craniotomy
3) Hematoma volume – look for that middle ground since neither large nor small hematomas seem to benefit from surgery
4) GCS at presentation – scores of 9-12 seem to be the sweet spot
Patients with ICH with IVH are a special group because IVH is an independent risk factor for poor outcomes. Patients with IVH have a 51% risk of death compared to 20% without IVH (7). Theoretically, removal of the hemorrhage should decrease the risk of hydrocephalus since blood in the ventricles can clot and cause obstruction. Historically, external ventricular drains (EVD) have often been placed for this indication, but they are often complicated by clot formation within the drain itself or other mechanical factors that interfere with proper drainage. To date, no RCTs directly compare EVD placement vs. medical therapy alone for acute ICH. The ESO and the American Heart Association/American Stroke Association (AHA/ASA) guidelines state it is “reasonable” to insert an EVD if the patient demonstrates clinical or radiological signs of hydrocephalus, but these recommendations are based on limited, low-quality evidence (1,8).
Since the ESO and AHA/ASA guidelines were released, the CLEAR III trial has been completed. This was a RCT that compared EVD placement with alteplase to EVD placement with normal saline in patients with sICH with hemorrhage volume <30 mL causing obstruction of the 3rd or 4th ventricle. Theoretically, clearing intracatheter clots with alteplase should improve drainage and decrease hydrocephalus. The trial found that while EVD use with alteplase was relatively safe, it did not confer any clear clinically oriented benefits. There were fewer serious adverse events in the alteplase group apart from a nonsignificant increase in hemorrhage. There was a significant decrease in mortality at six months with alteplase, but also a significantly increased proportion of patients with a modified Rankin Scale (mRS) of 5.
A large limitation of CLEAR III is that the treatments were not well concealed and the two groups received significantly different care. Because alteplase was much more effective at clearing hematoma than normal saline, the control group ended up receiving significantly more doses of normal saline than the alteplase group in order to clear the EVD. As a result, the control group had the EVD in place for a significantly longer period of time, effectively unmasking the interventions for each patient. Prolonged EVD placement also provides a nidus for infection. The normal saline group subsequently had nearly double the rate of ventriculitis (9), an infection of the ventricular system. When associated with an EVD, it is usually caused by skin flora and requires antibiotics intravenously or through the EVD itself (10). This increase in EVD-related infections likely contributed to the mortality difference (11). These confounders limit interpretation of the data. If the two groups had EVDs placed for the same duration, the results may well have shown no mortality difference at all.
The ESO and AHA guidelines are right on the mark when it comes to IVH. Unfortunately, the CLEAR III trial had some serious flaws, and as a result, they do not add anything to the existing body of evidence. It should also be emphasized that despite the mortality difference in the CLEAR III trial, surviving patients did not have favorable neurological outcomes. If an intervention can save a life, but the remaining life is of poor quality requiring constant care, would the patient even want to be saved?
Bottom line: There is no evidence of benefit from EVD placement for IVH. More studies are needed evaluating EVD with alteplase, but it appears to be safe and may result in shortened EVD placement time and subsequently reduced iatrogenic infections.
The classic teaching in Emergency Medicine is that cerebellar hemorrhage needs neurosurgical intervention due to the limited space in the posterior fossa where increased ICP can rapidly progress to hydrocephalus, herniation, and death. A meta-analysis of outcomes in cerebellar hemorrhage by Witsch et al in 2013 found a pooled mortality rate of 30.9% with all interventions (12). There have been no well-designed RCTs for infratentorial hemorrhage, likely due to ethical considerations for such a morbid condition. Most of the evidence for infratentorial hemorrhage is based on retrospective data. Based upon a few older studies of cerebellar hemorrhage, the AHA/ASA guidelines make a Class I recommendation for hematoma evacuation if the patient develops neurological deterioration, brainstem compression, or hydrocephalus. One of the largest of these studies was a multicenter study in Italy that suggested an initially low GCS and hematoma size >3 cm are poor prognostic indicators (8,13). That study also suggested that medical therapy had better outcomes than surgical treatment. However, this study had poorly described methods, was not randomized, did not define indications for surgery and surgical techniques, and lacked demographic data (13). In a retrospective study of 49 cases of spontaneous cerebellar ICH, initial treatment of hydrocephalus with EVD frequently required delayed hematoma evacuation (14), leading to the AHA/ASA recommendation that EVD alone should not be performed (8). The ESO guidelines make no recommendations for cerebellar ICH, citing a lack of evidence from RCTs (1).
A recent retrospective study by Luney et al of 104 patients with acute cerebellar hemorrhage supported the above recommendations, demonstrating significantly improved mortality with suboccipital decompressive craniectomy (17.4%) over medical therapy (45.7%) or EVD placement with medical therapy (47.8%). This study unfortunately did not include any data on favorable neurological outcomes (15). A prospective study of patients at least 65 years of age with cerebellar hemorrhage with IVH and hematoma diameter of at least 3 cm evaluated the effectiveness and safety of EVD with intraventricular fibrinolysis with urokinase compared to conservative management or clot evacuation. The study authors initially offered all patients surgical decompression (n = 47). If this was refused then EVD + urokinase was offered (n = 43), and if that was refused then patients received medical therapy (n = 28). All patients underwent surgery within 8 hours of ictus and there were no significant differences between baseline characteristics of the groups. Compared to conservative treatment, both EVD + urokinase and suboccipital decompression had significantly more patients with favorable neurological outcomes, defined as a mRS of 0-3 at six months, and significantly improved mortality. There was no significant difference between EVD + urokinase and surgical decompression for these outcomes (16). Trials evaluating minimally invasive techniques for cerebellar hemorrhage have been marred by small numbers of patients, lack of clinically oriented outcomes, and no comparison groups.
Bottom line: Based on existing literature, the AHA/ASA recommendations are adequate, and newer studies do not add much. For cerebellar hemorrhage, providers must maintain a low threshold for neurosurgical intervention, especially for patients with large hematoma volumes. Any signs of neurological deterioration, brainstem compression, or hydrocephalus warrant intervention. Decompressive craniectomy is standard of care. More data is required to evaluate whether EVD with fibrinolysis or minimally invasive techniques may offer benefit.
Due to declining mental status, the patient received a EVD and unfortunately passed away in the ICU shortly thereafter.
There is weak evidence that neurosurgical interventions offer benefit in sICH. We may wring our hands when our consultants say no acute interventions are required, but their inertia is not unfounded. Patients who suffer sICH are at high risk of death and poor functional status, and craniotomy is a morbid procedure. However, there may be a subset of patients who will benefit from early neurosurgical intervention. Newer evidence appears to point toward improved outcomes with minimally invasive techniques. There are several ongoing trials that hopefully will shed light on this murky topic, including MISTIE III and the ENRICH trial that should be wrapping up sometime in the next few years. In the ED, we rely on the expertise of our consultants to help guide management. However, we will undoubtedly come across cases where inaction by a consultant can be deleterious to our patients. We need to know our consultants’ literature so we can intelligently advocate for our patients. In our case, the patient should have received a decompressive craniectomy and not an EVD alone. Now, armed with the evidence, we can better push for our patients to receive appropriate neurosurgical care to optimize outcomes.
- Spontaneous supratentorial ICH: consider neurosurgical intervention, preferably minimally invasive for patients with:
- GCS 9-12
- Hematoma volumes < 50 mL
- Within 21 hours of ictus
- sICH with IVH and evidence of hydrocephalus: EVD placement is reasonable, but no clear benefit
- Cerebellar hemorrhage: craniotomy +/- EVD is indicated for:
- Neurological deterioration
- Brainstem compression
- Hematoma diameter of at least 3 cm (strongly considered)
Peer Reviewers: Raul Hernandez, MD and Kyle Kelson, MD
Faculty Advisor: Ian deSouza, MD
- Steiner T, Al-Shahi Salman R, Beer R, Christensen H, Cordonnier C, Csiba L, Forsting M, Harnof S, Klijn CJ, Krieger D, Mendelow AD, Molina C, Montaner J, Overgaard K, Petersson J, Roine RO, Schmutzhard E, Schwerdtfeger K, Stapf C, Tatlisumak T, Thomas BM, Toni D, Unterberg A, Wagner M; European Stroke Organisation. European Stroke Organisation (ESO) guidelines for the management of spontaneous intracerebral hemorrhage. Int J Stroke. 2014 Oct;9(7):840-55. doi: 10.1111/ijs.12309. Epub 2014 Aug 24.
- Mendelow AD, Gregson BA, Fernandes HM et al. Early surgery versus initial conservative treatment in patients with spontaneous supratentorial intracerebral haematomas in the International Surgical Trial in Intracerebral Haemorrhage (STICH): a randomized trial. Lancet 2005; 365:387-97.
- Gregson BA, Broderick JP, Auer LM, et al. Individual patient data subgroup meta-analysis of surgery for spontaneous supratentorial intracerebral hemorrhage. Stroke 2012; 43: 1496–504.
- Wang WZ, Jiang B, Liu HM, et al. Minimally invasive craniopuncture therapy vs conservative treatment for spontaneous intracerebral hemorrhage: results from a randomized clinical trial in China. Int J Stroke 2009; 4: 11–16.
- Mendelow AD, Gregson BA, Rowan EN, Murray GD, Gholkar A, Mitchell PM. Early surgery versus initial conservative treatment in patients with spontaneous supratentorial lobar intracerebral haematomas (STICH II): A randomised trial. Lancet 2013; 382: 397–408.
- Mould WA, Carhuapoma JR, Muschelli J, et al . Minimally invasive surgery plus recombinant tissue-type plasminogen activator for intracerebral hemorrhage evacuation decreases perihematomal edema. Stroke 2013; 44: 627–34.
- Li Y, Yang R, Li Z, Yang Y, Tian B, Zhang X, Wang B, Lu D, Guo S, Man M, Yang Y, Luo T, Gao G, Li L. Surgical evacuation of spontaneous supratentorial lobar intracerebral hemorrhage: comparison of safety and efficacy of stereotactic aspiration, endoscopic surgery, and craniotomy. World Neurosurgery (2017). doi: 10.1016/j.wneu.2017.05.134.
- Hemphill JC, 3rd, Greenberg SM, CS Anderson K, et al. American Heart Association Stroke, C. Council on, N. Stroke, C. Council on Clinical, Guidelines for the Management of Spontaneous Intracerebral Hemorrhage: a guideline for healthcare professionals from the American Heart Association/American Stroke Association. Stroke2015;46:2032–60.
- Hanley DF, Lane K, McBee N, Ziai W, Tuhrim S, Lees KR, Dawson J, Gandhi D, Ullman N, Mould WA, Mayo SW, Mendelow AD, Gregson B, Butcher K, Vespa P, Wright DW, Kase CS, Carhuapoma JR, Keyl PM, Diener-West M, Muschelli J, Betz JF, Thompson CB, Sugar EA, Yenokyan G, Janis S, John S, Harnof S, Lopez GA, Aldrich EF, Harrigan MR, Ansari S, Jallo J, Caron JL, LeDoux D, Adeoye O, Zuccarello M, Adams HP Jr, Rosenblum M, Thompson RE, Awad IA; CLEAR III Investigators: Thrombolytic removal of intraventricular haemorrhage in treatment of severe stroke: results of the randomised, multicentre, multiregion, placebo-controlled CLEAR III trial. Lancet. 2017 Feb 11;389(10069):603-611.
- Beer R, Pfausler B, Schmutzhard E. Management of nosocomial external ventricular drain-related ventriculomeningitis. Neurocrit Care. 2009;10(3):363-7.
- Muralidharan R. External ventricular drains: Management and complications. Surg Neurol Int. 2015;6(Suppl 6):S271–4.
- Witsch J, Neugebauer H, Zweckberger K, Juttler E. Primary cerebellar haemorrhage: complications, treatment and outcome. Clinical Neurol Neurosurg. 2013 July;115(7):863-9.
- Da Pian R, Bazzan A, Pasqualin A. Surgical versus medical treatment of spontaneous posterior fossa haematomas: A cooperative study on 205 cases. Neurological Research. 1984;6(3):145-151.
- van Loon J, Van Calenbergh F, Goffin J, Plets C. Controversies in the management of spontaneous cerebellar haemorrhage: a consecutive series of 49 cases and review of the literature. Acta Neurochir (Wien). 1993;122:187–193.
- Luney MS, English SW, Longworth A, et al. Acute Posterior Cranial Fossa Hemorrhage—Is Surgical Decompression Better than Expectant Medical Management? Neurocritical Care. 2016;25(3):365-370. doi:10.1007/s12028-015-0217-7.
- Zhang J, Wang L, Xiong Z, Han Q, Du Q, Sun S, Wang Y, You C, Chen J. A treatment option for severe cerebellar hemorrhage with ventricular extension in elderly patients: intraventricular fibrinolysis. J Neurol. 2014 Feb;261(2):324-9.