Study published in Journal of Stroke and Cerebrovascular Diseases published Nov 2020 looking at the link between blood pressure and outcome in ICH patients using the data from the VISTA trials (Virtual International Stroke Trials Archive)

Retrospective analysis of VISTA-ICH trial. n=384. Results below.

 Odds of unfavorable outcome for blood pressure categories. Left: unadjusted ORs. Right: adjusted ORs.

Study concluded that elevated BP after ICH is associated with poor outcome and supports the practice of targeting SBP of 140 mm Hg.

Limitations of the study:

  1. observational nature
  2. small sample size
  3. study population skewed towards less disease severity

Note that ENLS (published 2017) changed guidelines:

…it may be reasonable to target a systolic blood pressure between 140 and 180 mmHg with the specific threshold determined based on patient comorbidities and level of chronic hypertension.

While AHA/ASA (published 2015) guidelines originally stated that: (prior to recent ICH trials being published)

For ICH patients presenting with SBP between 150 and 220 mmHg and without contraindication to acute BP treatment, acute lowering of SBP to 140 mmHg is safe (Class I; Level of Evidence A) and can be effective for improving functional outcome


Claude Hemphill, J., & Lam, A. (2017). Emergency Neurological Life Support: Intracerebral Hemorrhage. Neurocritical Care27(S1), 89-101. doi: 10.1007/s12028-017-0453-0

Francoeur, C., & Mayer, S. (2020). Acute Blood Pressure and Outcome After Intracerebral Hemorrhage: The VISTA-ICH Cohort. Journal Of Stroke And Cerebrovascular Diseases30(1), 105456. doi: 10.1016/j.jstrokecerebrovasdis.2020.105456

Hemphill, J., Greenberg, S., Anderson, C., Becker, K., Bendok, B., & Cushman, M. et al. (2015). Guidelines for the Management of Spontaneous Intracerebral Hemorrhage. Stroke46(7), 2032-2060. doi: 10.1161/str.0000000000000069

Images of Vasospasm

This is a review of an interesting study from 1997 (published in Stroke) that illustrates how vasospastic arteries in a rat subarachnoid hemorrhage model looks like under the scanning electron microscope.

The researchers injected hemolysate (lysed autologous blood) into the cisterna magna of male Sprague-Dawley rats.  After ten minutes, a polymer resin casting medium was injected intravenously.  Once the resin has casted, the tissue and bones were corroded using NaOH solution, until only the vascular cast remains.  The casts were visualized under scanning electron microscope, and the following images were derived:


Basilar artery of:  A.  saline-injected control rats;  B. hemolysate-injected rats;  arrowheads = PICA;  note the narrowing and corrugation seen in the basilar artery of the hemolysate-injected rats.

Capture.JPG Other major arteries were also observed to be in vasospasm.  (note corrugation in these arteries)


Vasospasm (demonstrated as corrugation in the casted vessels) is seen in the major arteries (A) as well as the small arteries (B).


High magnification showing “corrugation” of the basilar artery.  The arrows point to nuclear indentations which correspond the the endothelial cell nucleus.  (see below0


The researchers also performed conventional SEM.  A and B are normal vessels while C and D are hemolysate-induced vasospastic blood vessels.   A and C are low magnification, B and D are high magnification.

The normal vessels (A) show that the inner surface is very smooth and the vessel wall is thin, and (B) the endothelial nuclei are clearly observed, projecting into the inner surface at regular intervals of 10-20 um.

The vasospastic vessels (C) shows that the smooth muscle layer is thicker, corrugation is observed and (D) many humps are sandwiched and flattened between hills formed by the endothelial cells.

Cast model shows corrugation, characteristic folds of endothelial cells at regular intervals and indentations of endothelial cell nuclei at each peak of those folds.  These indentations correspond to the humps seen in conventional SEM analysis.  The mechanical force of corrugation compressed the endothelial cells, flattened their nuclei and likely disturbed their function.  These physical alterations cause narrowing of the vessels, disturbs local blood flow, and may disturb blood coagulation and adhesion of WBC and platelets to the endothelium.  This may be a mechanism that explains thrombus inflammation and inflammatory response in these diseased vessels.

Their research also showed that arteries exposed to greater amount of hemolysate exhibit more severe vasospasm.


Ono, S. et al. “Three-Dimensional Analysis Of Vasospastic Major Cerebral Arteries In Rats With The Corrosion Cast Technique”. Stroke 28.8 (1997): 1631-1638.

Data Collection Sheet for AVMs

I created this form based on the article from Stroke Journal referenced below.  Next step is to create a public Excel template or REDCap database template that each neurocritical care unit can use for their own AVM database. Figure below shows screen capture from first page (form is 2 pages long).  PDF and DOCX versions of the form area available (see bottom for links to files.)


Data Collection Sheet for Brain AVM (pdf)

Data Collection Sheet for Brain AVM (docx)



  1. BAVM (brain AVM) – preferred over cerebral AVM (CAVM), abnormal tangle of vessels that results in arteriovenous shunting (nonutritive blood flow) demonstrated by 4-vessel cerebral contrast angiography.
  2. Date of presentation – date on which patient experienced signs or symptoms that led to the medical evaluation
  3. Index date – medical encounter through which the date of presentation was learned
  4. diagnosis date – date of definitive diagnosis of BAVM (i.e. date of contrast angiography)
  5. language cortex – the left hemisphere unless additional clinical data suggets otherwise
  6. eloquence – locations as per Spetzler-Martin score, plus the thalamus/hypothalamus/basal ganlgia;  ??nondominant parietal lobe may be considered eloquent as visuospatial deficits may be under-recognized but disabling
  7. diffuse nidus – has peninsula or islands of intervening brain
  8. old BAVM hemorrhage – all instances of CT or MR evidence of bleeding that is not temporally related to imaging for current signs and symptoms
  9. superficial drainage – all drainage from BAVM is through cortical venous system
  10. deep drainage – if any of all of drainage is through deep veins
  11. deep veins – internal cerebral v., basal v., precentral cerebral v.; in the posterior fossa, only cerebellar hemispheric veins that drain directly into straight sinus, torcula or transverse sinus are considered superficial
  12. venous stenosis – narrowing of any draining vein outflow pathway in 2 angiographic views; %stenosis – narrowest diameter of vein (in mm) divided by largest diameter of vein prox to stenosis (in mm)
  13. venous ectasia – any change in venous caliber in the venous runoff or drainage from BAVM with a >2-fold caliber change in any draining venous channel.
  14. sinus thrombosis – defect in dural venous sinus, excludes arachnoid granulations
  15. feeding vessel – arterial structure that angiographically demonstrates a contribution of flow to the AV shunt
  16. feeding arteries – parent arteries that give rise to vessels that supply flow to BAVM
  17. penetrators / perforators – vessels that are normally end arteries
  18. flow-related aneurysm – aneurysm lies on a pathway that carries non-nutritive blood flow supplying BAVM
  19. nidal aneurysm – aneurysm contiguous with the vascular mass
  20. proximal – located on the vessel or branch points of circle of Willis or proximal to it
  21. distal – refers to more distal locations beyond the circle of Willis


“Reporting Terminology For Brain Arteriovenous Malformation Clinical And Radiographic Features For Use In Clinical Trials”. Stroke 32.6 (2001): 1430-1442. Web.

Landmark Stroke Trials

Within 3 hours – within 4.5 hours – beyond 4.5 hours

Within 3 hours but thrombolysis contraindicated

Within 4.5 hours but thrombolysis contraindicated

Proximal artery vs distal artery occlusions


Catheter-based approaches

IA approaches

Ischemic penumbra model:

Cerebral artery occlusion – hypoperfused braintissue at risk for infarction salvageable by restoration of blood flow (ischemic penumbra) reversible – irreversible infarction;  –  brain titssue with irreversible damage (ischemic core); – decreased perfusion but no infarction risk regardless of treatment (benign oligemia)

Reperfusion leads to better outcomes

Time to irreversible infarction?  What predicts?

Every minute artery occluded – 2M neurons die

10 hours = neuronal loss occurring with 26 years of normal aging

Coil retrievers – wraps around clot and pulls it back

Stent retrievers – expands site of occlusion by stent, traps and extracts thrombus

Aspiration devices – sucks thrombus

Challenge: distinguish penumbra from core infarct from benign oligemia


National Institute of Neurological Disorders and Stroke (NINDS) 1990s[i] – 1995 sponsored 2 RCTs of IV rtPA vs placebo:  624 patients with ischemic stroke within 3 hours – 16% inc in favorable outcome (mRS0-1) at 3 monthes (42.6% vs 26.6% p<0.01 NNT 6), inc risk for Sxic brain hemorrhage (6.4% vs 0.6% p<0.001) – FDA approval

European Cooperative Acute Stroke Study III (ECASS-3)[ii] – 2008 821 patients with stroke <80y/o present within 3-4.5hours, (mRS 0-1 52.4% vs 45.2% p=0.04 NNT 14)

*dichotomous analysis – compelling; if analyzed for shift towards improved outcomes across full range – even more strongly associated with benefit (NNT 3 in 0-3h, NNT 7 in 3-4.5h)

*IV rtPA established as standard therapy for AIS within 3 hours; still not approved by FDA for use I 3-4.5 although recommended for moderately severe Sx <80y and without C/I

Stroke diagnostic tests:  CT, MRI, CTA, MRA, echo, telemetry, outpatient cardiac monitoring, HbA1C, lipid panel;  in select patients (inflammatory markers, hypercoag work-up, US lower ex, LP, blood cultures)

Meta-analysis of 2775 patients undergoing IV rtPA – odds of good outcome dependent on time – 0-90 mins OR 2.55, 91-180mins OR 1.64; 181-270 minutes OR 1.34, no benefit beyond 4.5 hours; confirmed similar risk of symptomatic brain hemorrhage seen in NINDS trial (5.2 vs 1.0% OR 5.37)


Outcomes are better overall

However: after tPA aone – only 10-15% ICA occlusions and 25-50% prox MCA occlusions recanalize; on 35-40% achieve functional independence

*prox artery occlusions (1/3 of all AIS) may be resistant to IV rtPA alone;  goal: improve recanalization rates with other methods

Catheter based treatment

Early trials (first generation approaches) failed to show clinical benefit despite successful recanalization

Chemical Thrombolysis

Prolyse in Acute Cerebral Thromboembolism (PROACT) II trial[iii] – 1999 180 patients within 6h + angio confirmed MCA occlusions, IA recombinant prourokinase (r-proUK) with heparin (2000 U bolus + 500 U/H x4h) vs heparin alone – (mRS 0-2 at 90 days:  39.4% vs 25.4% OR 2.13 p=0.04 NNT7) (higher risk of symptomatic brain hemorrhage (10.2% vs 1.9% p=0.06

*benefits marginal, offset by increased risk of harm, not approved for IA thrombolysis in AIS

Japanese study using urokinase vs best medical care supported findings of PROACT II (mRS 0-1 42.1 vs 22.8% p=0.045, sxic hemorrhage 8.8 vs 1.8% p=0.21)

Mechanical Thrombectomy

Mechanical thrombectomy devices approved by FDA based on technical efficacy and safety reports from large multicenter case registries – acceptable complication rates:  7-19% experience device and procedure-related complications (device fracture, vessel perforation and hemorrhage, nontarget artery embolization

Coil retriever / aspiration devices – approved by FDA based on single-group studies showing improved revascularization for prox artery occlusions, results were compared to historical control (from PROACT II)

Interventional Management of Stroke (IMS) III trial[iv] – 2013 standard dose IV rtPA vs low dose IV rtPA and IA rtPA or mechanical thrombectomy;  only 1% stent retrievers; no preprocedure vascular imaging selection in 46.6% (so 21% did not have proximal artery occlusion in IA treatment group); 656 patients over 6 years, stopped for futility (mRS0-2 at 90d 40.8% vs 38.7%; mortality 19.1% vs 21.6% p=0.52) or symptomatic brain hemorrhage (6.2% vs 5.9% p=0.83)

Intra-arterial vs Systemic Thrombolysis for AIS (SYNTHESIS EXP) study[v] – 2010 2 groups of 181 patients; IV rtPA vs mech thrombectomy or IA therapy within 4.5h; of IA patients, 60% treated with rtPA infusion and microguidewire thrombus fragmentation, 31% with thrombectomy devices, 13% with stent retrievers; no benefit observed, no disability at 90d (30.4% vs 34.8% p=0.037) no safety differences symptomatic ICH (5.5% vs 5.5% p=0.99) and mortality 7.7% vs 6.1% p=0.53.

2 studies[vi][vii]: both 2012 newer stent retriever devices vs earlier coil retriever – improve recanalization, reduced mortality, better functional outcomes; established superiority but no direct comparison with control group with IV rtPA

4 RCTS of stent retrievers vs medical treatment

Multicenter RCT of Endovascular Treatment for AIS (MR CLEAN)[viii] phase 3 – 2015 mech thrombectomy within 6h vs standard treatment improved improved 90d clinical outcomes; 90.6% received IV rtPA within 4.5h; 16 stroke centers in Holland; stroke patients with confirmed prox artery occlusions – randomized to std (267) or standard + IA (pred stent retriever) treatment; mRS0-2 32.6% vs 19.1% p<0.01 NNT8; no hemorrhagic safety concerns! (symptomatic ICH 7.7% vs 6.4%; 30d mortality 18.4 vs 18.9); increased risk of new ischemic stroke within 90d (5.6 vs 0.4 p<0.01) likely procedure-related embolization

Endovascular Treatment for Small Core and Proximal Occlusion Ischemic Stroke (ESCAPE) trial[ix] – 2015 n=316; within 12h, 22 global sites; prox artery occlusions identified with CT angio randomized to best medical therapy (IV rtPA in 78%) vs IA (86% stent retrievers) – stopped early for efficacy (mRS0-2 53% vs 29.3% p<0.01 NNT4) – improved workflow and enhanced patient selection emphasized – onset to reperfusion time of 241 minutes; only 15.5% treated beyond 6h

Extending the Time for Thrombolysis in Emergency Neurological Deficits – Intra arterial (EXTEND-IA)[x] – 2015 Australian, used CT perfusion imaging, randomized patients with favorable mismatch patterns to IV rtPA vs IVrtPA + stent retriever within 4.5h – stopped early for technical efficacy after 35 patients in each group; mRS 0-2 71.4% vs 40% p<0.01 NNT3

SOLITAIRE with intention for Thrombectomy as Primary Endovascular Treatment (SWIFET PRIME)[xi] – 2015 stopped early after 196 patients; rtPA able to undergo cath within 6h with ant circulation occlusion – thrombectomy superior (mRS0-2 60.2 vs 35.5% p<0.01 NNT4)

Imaging-based patient selection

Diffusion and Perfusion Imaging Evaluation for Understanding Stroke Evolution Study 2 (DEFUSE-2)[xii] – 2012 uncontrolled, prospective cohort study, n=99, acute stroke; favorable penumbral pattern (penumbra to infarct core ratios >1.8) better 90d outcomes with successful reperfusion vs no reperfusion (mRS0-2 56.5 vs 31.3% p=0.04); no benefit in patients without a favorable penumbra pattern (mRS0-2 25% vs 22% p>0.99)

MR RESCUE study[xiii] – 2013 phase 2b open-label RCT with blinded outcome, 118 patients with AIS – no benefit of mech thrombectomy in favorable penumbral patterns (20.6 vs 26.5% p=0.78)

*DEFUSE-2 used only MRI-based selection; MR RESCUE included CT perfusion in 20% of analyzed patients

*DEFUSE-2 defined penumbra to core ratio as >1.8 with max core infarct vol 70ml; MR RESCUE used smaller penumbra to core ratio of >1.4 and larger max core infarct vol of 90ml

*led to greater rates of futile reperfusion in MR RESCUE

*EXTEND-IA used similar algorithm as MR RESCUE but at earlier time points (<4.5h)


  1. Pragmatic and simple – CT angio to identify prox art occlusions; enroll based on time (<6h) as in MR CLEAN
  2. Assess early infarct signs (core infarct) with noncontrast CT and time window as in ESCAPE (<12h) and in SWIFT PRIME (<6h)
  3. CT angio assessment of collaterals as in ESCAPE
  4. Penumbra imaging with CT or MRI perfusion imaging with angio to confirm occluded artery within 4.5h as in EXTEND-IA and some in SWIFT PRIME

*which is superior? Not clear but all 3 shown to select patients who benefit from adjunctive IA therapy

Need more accurate and reliable measurement of brain ischemia

Penumbral imaging may take up to 30 minutes – will this negate any efficacy advantage?

Endovascular Trial Comparisons

Interventional Management of Stroke (IMS III), SYNTHESIS EXP, MR RESCUE – tested first-gen strategies for IA treatment of prox occlusions

*differences cf more recent trials: rates of reperfusion, time to reperfusion, selection

*SYNTHESIS EXP used clot fragmentation with IA rtPA in 60%


*rates of substantial reperfusion (TICI grades 2b or 3) lower in IMS III (40%) and MR RESCUE (27%) cf stent retriever trials (58-88%)

*time to reperfusion lower in recent trials – 4h in ESCAP, 4.1h in EXTEND-IA, 4.2h in SWIFT PRIME, 5.4 in IMSIII

*MR CLEAN ESCAPE EXTEND IA SWIFT PRIME required confirmation of prox artery occlusion on baseline CT angio – more homogenous cohort, selects more likely to benefit, decreases rate of futile reperfusion

AMBULYSIS – ambulances staffed by stroke experts fitted with CT scanners – thrombolysis in ambulance

Current AHA/ASA guidelines – IV rtPA administer to all eligible patients as quickly as possible (door-to-needle time <60 mins) in the 0-3h window (Class 1-A), in the 3-4.5 window (class I-B) and even if considering other adjunctive therapies (Class 1-A).  Reduce and avoid delays to reperfusion (Class 1-A);  IA thrombolyssi with rtPA in carefully selected patients with MCA occlusion  within 6h onset (Class 1-B), based on MELT and PROACT II).  Recommend stent retrievesr over earlier generation coil retrievers (Class I-A).  Weak recommendations for clinical efficacy of mechanical thrombectomy (Class IIa-B) – does not include the 4 new trials in 2015.

MCA Embolism Local Fibrinolytic Intervention Trial (MELT)

[i] Tissue plasminogen activator for acute ischemic stroke. N Engl J Med. 1995;333(24):1581-1587.

[ii] Hacke  W, Kaste  M, Bluhmki  E,  et al.  Thrombolysis with alteplase 3 to 4.5 hours after acute ischemic stroke. N Engl J Med. 2008;359(13):1317-1329.

[iii] Furlan  A, Higashida  R, Wechsler  L,  et al.  Intra-arterial prourokinase for acute ischemic stroke. JAMA. 1999;282(21):2003-2011.

[iv] Broderick  JP, Palesch  YY, Demchuk  AM,  et al.  Endovascular therapy after intravenous t-PA vs t-PA alone for stroke. N Engl J Med. 2013;368(10):893-903.

[v] Ciccone  A, Valvassori  L, Ponzio  M,  et al.  Intra-arterial or intravenous thrombolysis for acute ischemic stroke? J Neurointerv Surg. 2010;2(1):74-79.

[vi] Nogueira  RG, Lutsep  HL, Gupta  R,  et al.  Trevo vs Merci retrievers for thrombectomy revascularisation of large vessel occlusions in acute ischaemic stroke (TREVO 2). Lancet. 2012;380(9849):1231-1240.

[vii] Saver  JL, Jahan  R, Levy  EI,  et al.  Solitaire flow restoration device vs the Merci retriever in patients with acute ischaemic stroke (SWIFT). Lancet. 2012;380(9849):1241-1249.

[viii] Berkhemer  OA, Fransen  PS, Beumer  D,  et al.  A randomized trial of intra-arterial treatment for acute ischemic stroke. N Engl J Med. 2015;372(1):11-20.

[ix] Goyal  M, Demchuk  AM, Menon  BK,  et al.  Randomized assessment of rapid endovascular treatment of ischemic stroke. N Engl J Med. 2015;372(11):1019-1030.

[x] Campbell  BC, Mitchell  PJ, Kleinig  TJ,  et al.  Endovascular therapy for ischemic stroke with perfusion-imaging selection. N Engl J Med. 2015;372(11):1009-1018.

[xi] Saver  JL, Goyal  M, Bonafe  A,  et al. SOLITAIRE FR with the Intention for Thrombectomy as Primary Endovascular Treatment for Acute Ischemic Stroke. Paper presented at: International Stroke Conference; February 2015; Nashville, TN.

[xii] Lansberg  MG, Straka  M, Kemp  S,  et al.  MRI profile and response to endovascular reperfusion after stroke (DEFUSE 2). Lancet Neurol. 2012;11(10):860-867.

[xiii] Kidwell  CS, Jahan  R, Gornbein  J,  et al.  A trial of imaging selection and endovascular treatment for ischemic stroke. N Engl J Med. 2013;368(10):914-923.

Picture2 m_jrv150006f1 m_jrv150006t1 m_jrv150006f2 m_jrv150006f3 Picture1:
Acute Stroke Intervention A Systematic Review – Prabhakaran, S. et al.  JAMA. 2015;313(14):1451-1462. doi:10.1001/jama.2015.3058.