Presentation is loading. Please wait.

Presentation is loading. Please wait.

Updates in Stroke Natalie Hendon MD Virginia Mason Medical Center Seattle, WA 2/21/2015.

Similar presentations


Presentation on theme: "Updates in Stroke Natalie Hendon MD Virginia Mason Medical Center Seattle, WA 2/21/2015."— Presentation transcript:

1 Updates in Stroke Natalie Hendon MD Virginia Mason Medical Center Seattle, WA 2/21/2015

2 Learning Objectives Review stroke burden Discuss early dual antiplatelet therapy in minor CVA or TIA Use of SSRIs as potential neuromodulators for improved motor function in stroke patients at 3 months Optimal treatment of stroke or TIA due to intracranial stenosis Identify components of aggressive medical management in secondary stroke prevention New directions in emergent treatment of acute ischemic stroke

3 Outline CHANCE trial – Early short-term dual antiplatelet therapy FLAME trial – Fluoxetine for motor recovery after ischemic stroke Intracranial stenosis – optimal treatment – SAMMPRIS trial and WASID trial Updates in secondary prevention of stroke – Aggressive medical management – Statin usage Intra-arterial treatment for acute ischemic stroke – MR CLEAN – ESCAPE – EXTEND-IA

4 Epidemiology and Burden of Stroke 800,000 strokes annually in the US – 600K are new CVA, 200K are recurrent CVA Stroke is the leading cause of chronic disability in the US One of the leading causes of death, with 130K- 140K stroke deaths annually in the US Heterogenous disease – primary hemorrhage (10%), subarachnoid hemorrhage (3%), ischemia (87%), venous thrombosis, and more – Varying risk factors

5 Ischemic Stroke and TIA Patients with ischemic stroke and TIA are at high risk for recurrent events, especially in the first few days – Risk of CVA after TIA at 2 days is % – Risk of CVA after TIA at 90 days is 9-17% ABCD2 score to identify patients at higher risk of stroke – Patient age, BP, DM, type of symptoms, duration of symptoms – High risk scores (>5 points) have 12% risk of recurrent ischemic event in 1 week

6

7 CHANCE Clopidogrel in High-Risk Patients with Acute Nondisabling Cereberovascular Events  Rationale: Stroke is common during the first few weeks after a TIA or minor ischemic stroke. Dual antiplatelet therapy may provide additional benefit.  Methods:  Randomized, double blind, placebo controlled trial at 114 centers in China  5170 patients within 24 hrs after onset of high-risk TIA or minor ischemic stroke  Randomized to combination therapy with clopidogrel and aspirin OR to placebo plus aspirin  Primary outcome was stroke during 90 days

8 CHANCE: Inclusion Criteria  Age 40 or older  Diagnosis of acute minor ischemic CVA  NIHSS 3 or less at time of randomization  OR  Diagnosis of TIA  Focal brain ischemia with resolution of symptoms within 24 hrs, plus moderate to high risk of stroke recurrence (>4 on ABCD)  Ability to start study drug within 24 hrs after symptom onset

9 CHANCE: Exclusion Criteria  Hemorrhage, vascular malformation, tumor abscess, other major nonischemic brain disease  Isolated sensory symptoms  Isolated visual changes  Isolated dizziness or vertigo without evidence of acute infarct on baseline imaging  Score of >2 on modified Rankin scale prior to event  NIHSS of 4 or more  Clear indication for anticoagulation  Presumed cardioembolic source  Contraindication to clopidogrel or aspirin  History of intracranial hemorrhage  Others, standard tPA exclusion criteria

10 CHANCE trial 41,561 pts screened 5170 pts enrolled 2548 clopidogrel-ASA 2586 to ASA Median age % were women 65.7% with HTN 21.1% with DM 43% smoking history 27.9% with TIA, 72% CVA Dual antiplatelet therapy vs aspirin alone for 21 days, followed by aspirin OR Plavix

11 CHANCE: Outcome definitions Primary efficacy outcome was new stroke event (ischemic OR hemorrhagic) at 90 days Primary safety outcome was moderate to severe bleeding event Key secondary efficacy outcomes – New clinical vascular event (CVA, MI, vascular death due to stroke, MI, CHF, PE, sudden death, arrhythmia)

12 CHANCE: Results  Stroke occurred in 8.2% in the clopidogrel-aspirin group, and 11.7% in the aspirin group.  Hazard ratio 0.68, 95% confidence interval 0.57 to 0.81, P<0.001  Composite outcome of vascular events in clopidogrel-aspirin group 8.4% and 11.9% in aspirin group  TIA in 1.5% in clopidogrel-aspirin group and 0.2% in aspirin group  Moderate or severe hemorrhage occurred in 7 pts (0.3%) in clopidogrel-aspirin group and in 8 (0.3%) in the aspirin group (P=0.73)

13

14 CHANCE: Conclusions Conclusions: Combination clopidogrel and aspirin is superior to aspirin alone for reducing risk of stroke in the first 90 days and does not increase risk of hemorrhage – Within 24 hrs of symptom onset – Dual antiplatelet therapy 21 days followed by aspirin or Plavix alone

15 CHANCE: Caveats Population at very high risk for recurrent stroke and at low risk for hemorrhage Study conducted entirely in China – Lower rates of treatment for HTN, DM and HL – Higher incidence of large artery intracranial atherosclerosis – Higher prevalence of genetic polymorphisms that affect the metabolism of clopidogrel

16 CHANCE: Take-Home Message  Dual antiplatelet therapy for 21 days with onset within 12 hrs of TIA or minor ischemic stroke reduces chance of recurrent ischemia without change in bleeding risk

17 Depression and Stroke Mean prevalence of post-stroke depression is 20% among patients receiving care for stroke Post-stroke major depression has mean duration of 9 months Significant minority continue to have depression 3 years out Several trials have evaluated treatment of depression post-stroke Some trials have suggested improved functional outcome

18 FLAME Trial

19 FLAME Rationale: Hemiparesis is common after stroke and prior small trials suggest that fluoxetine enhances motor recovery. Study to investigate whether fluoxetine enhances motor recovery if given soon after ischemic stroke in pts with motor deficits. Methods: – Double-blind, placebo controlled trial – 9 stroke centers in France with ischaemic CVA and hemiparesis – Fluoxetine 20 mg daily or placebo for 3 months starting days after stroke – Primary outcome change of motor symptoms between day 0 and 90

20 FLAME: Background Recovery of neurological function is associated with intracerebral reorganization Various interventions after CVA have been shown to modulate brain plasticity after CVA – Amphetamines show enhanced recovery in animal models – BZDs and neuroleptics reduce recovery in these models – SSRIs have shown acute neuroprotective action on ischemic brain and promote hippocampal neurogenesis – Functional MRI studies show that single doses of fluoxetine and paroxetine overactivate motor cortices compared with placebo in pts with and without CVA

21 FLAME: Inclusion Criteria Acute ischaemic CVA within prior 5-10 days that caused hemiparesis/hemiplegia 9 stroke units in France Age Fugl-Meyer motor scale 55 or less at baseline

22 FLAME: Exclusion Criteria Severe post-stroke disability NIHSS >20 Substantial premorbid disability Pre-existing deficit such as residual motor defict from prior CVA, comprehension deficits, severe aphasia Diagnosis of depression On antidepressants, MAOIs, neuroleptics, or BZDs during month prior to inclusion Impending carotid endarterectomy Pregnancy Major disease processes that would prevent f/u

23 FLAME Primary outcome was mean change in FMMS between day 0 and 90 as assessed by physiotherapist Secondary endpoints were NIHSS, modified Rankin Scale and MADRS

24 FLAME 118 patients, 6 lost 57 fluoxetine, 56 placebo Well balanced baseline, demographic and stroke severity characteristics Mean age and h/o CVA higher in fluoxetine group FMMS score at day 0 higher in fluoxetine group NIHSS, mRS and MADRS scores similar at day 0 All patients underwent physiotherapy

25 FLAME: Results Mean progression in FMMS total score from baseline to day 90 was significantly higher in fluoxetine group than in placebo group – Significant for both upper and lower limb scores

26 FLAME: Results NIHSS at day 90 did not differ Motor component score significantly higher in fluoxetine group Adjusted mean MADRS change was significantly lower in fluoxetine group than placebo Frequency of depression significantly higher in placebo FMMS change signifiantly greater in fluoxetine group even after adjusting for depression

27 FLAME: Caveats Small number of patients enrolled (but larger than prior trials) Pts selected for motor deficit and thus do not represent all CVA deficits Treatment stopped at 90 days; long-term outcome unknown

28 FLAME: Take-Home Fluoxetine 20 mg started in 5-10 days post stroke with motor deficits resulted in improvement in motor recovery compared to placebo when tested with FMMS, but not when tested with NIHSS mRS showed more independent patients in fluoxetine group than placebo Major role of motor function recovery in global recovery and return to independent activities

29 Intracranial Stenosis and Stroke Intracranial stenosis causes 8-10% of ischemic strokes in the USA Risk of stroke especially high in patients with recent stroke or TIA and severe arterial stenosis (70-99%) Treatment options have changed over time due to release of results of major multicenter randomized stroke prevention trials

30 WASID Trial: Comparison of Warfarin and Aspirin for Symptomatic Intracranial Arterial Stenosis

31 WASID: Summary Warfarin used to be used preferentially instead of aspirin for treatment of atherosclerotic intracranial arterial stenosis. 569 patients were randomized after found by angiogram to have 50-99% stenosis of a major intracranial artery to receive warfarin or aspirin Trial stopped early due to safety for patients assigned to warfarin, with higher rates of death, major hemorrhage, and MI Primary endpoint of ischemic stroke occurred in 22.1% of aspirin patients and 21.8% of warfarin pts Aspirin should be used instead of warfarin in patients with intracranial arterial stenosis

32 SAMMPRIS Trial: Stenting and Aggressive Medical Management for Preventing Recurrent stroke in Intracranial Stenosis

33 SAMMPRIS: Rationale Designed to assess whether percutaneous transluminal angioplasty and stenting (PTAS) with Wingspan stent system plus aggressive medical treatment is more effective than aggressive medical treatment alone in high risk patients Patients with recent (<30 days) TIA or non- disabling CVA attributed to angiographically verified severe intracranial stenosis

34 SAMMPRIS: Early results Early results of the SAMPRIS trial showed by 30 days 14.7% of patients in the stenting group and 5.8% of pts in medical group had died or had a stroke  Trial stopped early Patients continued to be followed to establish whether early benefit in medical group would persist over time, or whether they would have higher incidence of later strokes

35 SAMMPRIS The rate of stroke in the medical management group was much lower than expected

36 SAMMPRIS: Aggressive Medical Management Aspirin 325 mg daily Clopidogrel 75 mg daily for 90 days Management of primary risk factors (HTN and elevated LDL) – SBP <140 or <130 if comorbid DM Drug from each major class of antihypertensives – LDL <70 Rosuvastatin Management of secondary risk factors (DM, elevated non-HDL, smoking, obesity, lack of exercise) with help of lifestyle modification program

37

38 SAMMPRIS: Hyperlipidemia Mangament

39 SAMMPRIS: Lifestyle modification program Delivered by coach who contacted patients by telephone every 2 weeks for the first 3 months, then monthly Baseline assessment of risk factors Formulation of action plan with focus on exercise, nutrition, weight, tobacco cessation Follow up counseling sessions to help acquire skills, motivation and support to adhere to plan

40

41 SAMMPRIS vs WASID WASIDSAMMPRIS HTN84%89% DM38%46% Hyperlipidemia71%89.5% History of/current smoking65%64% Antithrombotic tx53%63% Vascular risk factors at trial entry Similar rates for hypertension, history of/current smoking SAMMPRIS had higher rates of DM and hyperlipidemia, higher proportion starting out already on antithrombotic tx

42 SAMMPRIS Trial: Stenting and Aggressive Medical Management for Preventing Recurrent stroke in Intracranial Stenosis

43 SAMMPRIS Follow Up Median follow up of 32 months 15% of medical group and 23% of stenting group had stroke or death Occurrence of adverse effects (any stroke, major hemorrhage) was higher in stenting group  Medical management superior to stenting

44 SAMMPRIS Aggressive medical management may help address vascular risk factors more readily and may help prevent future strokes – In patients with intracranial stenosis – Dual antiplatelet treatment for 30 days post TIA/CVA – Following trial’s protocol to lowering BP and LDL cholesterol

45 SAMMPRIS: Take Home For patients with recently symptomatic severe (70-99%) intracranial stenosis, aggressive medical management is superior to stenting Aggressive medical management practices can lower risk of future stroke

46 Cholesterol Management in Stroke Statins have been shown to reduce the risk of stroke among patients with CAD and those at increased risk for cardiovascular disease Prior studies (meta-analysis 90,000 pts) have shown that reduction in stroke risk was primarily related to extent to which LDL was lowered SPARCL trial designed to determine whether daily dose of 80 mg atorvastatin would reduce risk of stroke in pts with NO known CAD who had had a CVA or TIA in the previous 6 months

47

48 2365 pts in atorvastatin group 2366 pts in placebo group Entry event was CVA in about 70% in both, TIA in about 30% in both. Hemorrhagic CVA accounted for 2% in both Time since entry about 85 days Similar risk factors Similar concomitant therapy for risk factors LDL about 133, HDL 50, TG SPARCL: Baseline Characteristics

49 SPARCL: Decreased risk of fatal or nonfatal stroke or TIA with Atorvastatin

50 SPARCL: Decreased risk of any coronary and cardiovascular events with Atorvastatin

51 AHA/ASA Scientific Advisory 2013: Statins

52 AHA: High/Moderate/Low Intensity

53 AHA: Secondary Prevention and Statins Women and men with clinical atherosclerotic cardiovascular disease (ACS, MI, angina, coronary revascularization, CVA or TIA presumed to be of atherosclerotic origin, and PVD or revascularization) are at increased risk of recurrent ASCVD and ASCVD death. An extensive body of evidence demonstrates that high- intensity statin therapy reduces ASCVD events more than moderate-intensity statin therapy in individuals with clinical ASCVD. No evidence to support titrating meds to goal LDL, as clinical trials were mainly fixed dose

54

55 AHA Recommendations in Stroke/TIA Patients 75 and younger, initiate high intensity statin therapy Patients older than 75, initiate moderate intensity statin therapy Counsel both on healthy lifestyle habits

56 Recent Advances: Intra-arterial Treatment for Acute Ischemic Stroke MR CLEAN (the Netherlands) ESCAPE (Canada) EXTEND-IA (Australia)

57 Argument for Intra-Arterial Treatment IV alteplase within 4.5 hours after onset of symptoms is the only reperfusion therapy with proven efficacy in patients with acute ischemic stroke Narrow therapeutic window Contraindications including surgery, coagulation abnormalities, h/o ICH Less effective at opening proximal occlusions >1/3 of acute anterior circulation CVA Early recanalization after IV alteplase is seen in only 1/3 of patients with occlusion of the ICA terminus  poor prognosis

58 Among pts with a proximal vessel occlusion in the anterior circulation, 60-80% die within 90 days after stroke onset, or do not regain functional independence despite alteplase treatment. 9 Recent studies have shown superiority of retrievable stents over the prior generation of thrombectomy devices. 10

59 Intra-arterial therapy options Locally delivered thrombolytic agents with clot retrieval Thrombectomy with mechanical devices

60

61 MR CLEAN: Objective Is intraarterial treatment plus usual care more effective than usual care alone – Patients with proximal arterial occlusion in the anterior cerebral circulation – Treatment intraarterially within 6 hours after symptom onset

62 MR CLEAN: Design Phase 3, multicenter clinical trial, randomized, open-label treatment, blinded end-point evaluation Patients with acute ischemic stroke and proximal intracranial anterior ciculation occlusion on vessel imaging Intraarterial thrombolysis, mechanical treatment, or both plus usual care (which could include IV altepase)

63 MR CLEAN: Inclusion Criteria 16 centers in the Netherlands Age 18-no upper limit Acute ischemic stroke of proximal anterior circulation confirmed on vessel imaging (CTA or MRA or DSA) NIHSS 2 or higher Patients with additional extracranial ICA occlusion or dissection left to judgment of treating physician

64 502 patients Similar age (mid-60s), gender distribution, NIHSS, associated risk factors, blood pressure, prestroke modified Rankin score Similar time from stroke onset to start of IV alteplase (85-87 minutes) Similar distribution of clot Higher extracranial ICA occlusion in intervention group Time from stroke onset to randomization about 200 minutes Time from stroke onset to groin puncture 260 minutes MR CLEAN

65 MR CLEAN: Intervention Arterial catheterization with a microcatheter to the level of occlusion and delivery of a thromboylitic agent, mechanical thrombectomy, or both Use of alteplase or urokinase – Restricted total dosage if IV-alteplase was given Mechanical treatment: thrombus retraction, aspiration, wire disruption, use of retrievable stent Devices approved by FDA or CE and approved by steering committee 1+ members of each intervention team had to have completed at least 5 full procedures with a particular device

66 MR CLEAN: Outcome Primary outcome – score on modified Rankin scale at 90 days Secondary outcomes included: – NIHSS at 24 hrs and at 5-7 days or discharge if sooner – ADLs as measured with Barthel index – Health related quality of life with EuroQol Group 5-Dimension Self-Report Questionaire at 90 days Imaging outcomes: arterial recanalization mesausred with CTA or MRA at 24 hrs, final infarct volume on noncontrast head CT at 5-7 days Safety outcomes: hemorrhagic complications, progression of ischemia, new ischemic CVA in new vascular territory, death

67 MR CLEAN: Results Intraarterial therapy with or without mechanical thrombectomy performed in 196 of the 233 patients in the intervention group (84.1% – General anesthesia used in 37.8% – Second revascularization procedure (acute cervical carotid stenting) in 12.9% Mechanical treatment in 195 of the 233 pts (83.7%) – Retrievable stents in 190 patients (81.5%) – Other devices used in 5 pts (2.1%) Intraarterial thrombolytic agents used as monotherapy in 1 of the 233 pts (0.4%) No intervention in 37 pts (15.9%

68

69

70 No significant between-group difference in serious adverse effects in 90-day follow up period 13/233 pts (5.6% in intervention group had clinical signs of a NEW ischemic CVA in a different vascular territory within 90 days, and only 1/267 (0.4%) in the control group did No significant difference in mortality at 7, 30, 90 days of follow up Procedure related complications included embolization into new vascular territories outside the target downstream in 20/233 (8.6%), procedure related dissection in 4 (1.7%) and vessel perforations in 2 (0.9%)

71 MR CLEAN – Conclusions and Discussion Patients with acute ischemic stroke caused by a proximal intracranial arterial occlusion of the anterior circulation have BENEFIT with respect to FUNCTIONAL RECOVERY when intraarterial treatment is given within 6 hrs of onset  Clinically significant increase in functional independence in daily life by 3 months, without increase in mortality

72 MR CLEAN and prior IA trials Prior trials have not shown a benefit to IA treatment – Timing: Interventional Management of Stroke (IMS) III trial gave IV alteplase plus IA treatment with randomization 40 minutes after start of IV alteplase (about 120 minutes after IV alteplase This may have led to inclusion of more pts with favorable response to IV alteplase than MR CLEAN – Proximal occlusion vs more distal: IMS III and SYNTHESIS Expansion trial did not require radiographically proven intracranial occlusion Presence of proximal arterial occlusion uncertain in 47% of the study population

73 MR CLEAN Caveats 9% of pts in the intervention group had embolizationn into new vascular territories 13% underwent simultaneous second revascularization procedure (acute cervical carotid stent) – Unclear how this impacts results Low proportion of pts in control group had mRS of 0-2 in 90 day f/u – Broad inclusion with older pts and contraindications to IV alteplase? Poorer prognosis at baseline? Pts not blinded to intervention – May have influenced their opinions about health/functionality

74

75 ESCAPE Included: Patients with proximal intracranial occlusion in the anterior circulation up to 12 hours after symptom onset. Selected based on imaging with ASPECTS score Excluded: Large infarct core or poor collateral circulation on imaging. Randomly assigned participants to receive standard care or standard care plus endovascular treatment with use of available thrombectomy devices. Primary outcome: score on modified Rankin scale at 90 days. Trial stopped early because of efficacy.

76

77

78 Assessment at 90 days

79

80 ESCAPE: Conclusions Pts with acute ischemic CVA with small infarct core on imaging, proximal intracranial occlusion in the anterior circulation, and moderate to good intracranial collateral circulation, rapid endovascular treatment improved clinical outcome and reduced mortality. – Rapid reperfusion associated with better clinical outcome than slower reperfusion – Shorter interval times in this trial with median time from CT to first reperfusion of 84 minutes

81

82 EXTEND-IA Purpose: Prior trials of endovascular therapy for ischemic CVA have produced variable results. Study performed to test whether more advanced imaging selection, recently developed devices, and earlier intervention improve outcomes.

83 EXTEND-IA: Methods Pts with ischemic CVA, occlusion of ICA or MCA, evidence of small ischemic core, who were receiving IV-alteplase <4.5 hrs after onset of ischemic CVA Assigned to undergo endovascular thrombectomy with the Solitare Flow Restoration stent retriever, or to continue IV-alteplase alone Primary outcomes: Reperfusion at 24 hrs and early neurological improvement (>8 point reduction on NIHSS or score of 0 or 1 at day 3. Secondary outcomes included functional score on mRS at 90 days.

84

85

86 EXTEND-IA: Results Stopped early due to efficacy after 70 pts had undergone randomization. Percentage of ischemic territory that underwent reperfusion was higher in endovascular therapy at 24 hrs (median 100% vs 37%) and improved functional outcome at 90 days (71% achieving functional independence vs 40%) No significant differences in rates of death or symptomatic intracerebral hemorrhage.

87

88

89

90 EXTEND-IA: Conclusions In pts with ischemic CVA and proximal cerebral artery occlusion and salvagable tissue on CT perfusion imaging, early thrombectomy with Solitare FR stent retriever as compared with alteplase alone, improved reperfusion, early neuro recovery and functional outcome. Limitations: small number of patients limiting subgroup analyses

91 Conclusions Stroke is a common disorder with significant disability and mortality Dual antiplatelet therapy for 21 days, with treatment onset within 12 hrs of TIA or minor ischemic stroke, reduces chance of recurrent ischemia without change in bleeding risk Fluoxetine and potentially other SSRIs may improve motor function in CVA pts at 3 months, potentially through neuromodulation (animal models) For treatment of symptomatic intracranial stenosis: – Early studies showed aspirin is superior to warfarin – SAMMPRIS trial showed dual antiplatelet therapy for 3 months and aggressive medical treatment was superior to angioplasty and stenting SAMMPRIS trial also suggested that aggressive medical management alone led to a greater reduction in stroke risk factors than expected and can lower risk of future stroke – Dual antiplatelet therapy 90 days – High intensity statin therapy in pts 75 and younger – Aggressive BP management and LDL management (q30 day assessments) – Aggressive lifestyle modification with significant improvement seen at 4 months For patients with large artery intracranial occlusion in the anterior circulation, endovascular treatment within 6-12 hrs after stroke symptom onset is superior to alteplase alone without increased risk of hemorrhage or death

92 THANK YOU

93 References 1.Wang et al. Clopidogrel with Aspirin in Acute Minor Stroke or Transient Ischemic Attack. NEJM 2013;369: Narushima et al. Current Atheroscl Rep 2002;4: Chollet et al. Fluoxetine for motor recovery after acute ischaemic stroke. Lancet Neurol 2011;10: Chomowitz et al. Comparison of Warfarin and Aspirin for Symptomatic Intracranial Arterial Stenosis. NEJM 2005;352: Chimowitz et al. Stenting vs Aggressive Medical Therapy for Intracranial Arterial Stenosis. NEJM 2011;365: Derdeyn et al. Aggressive medical treatment with or without stenting in high-risk patients with intracranial artery stenosis (SAMMPRIS): the final results of a randomized trial. Lancet 2014;383: Amarenco et al. High-Dose Atorvastatin after Stroke or Transient Ischemic Attack. NEJM 2006;355: Stone et al. Circulation Published online at 9.Broderick et al. Endovascular therapy after IV t-PA vs t-PA alone for stroke. NEJM 3024;368: ) 10.Saver et al. Solitare Flow Restoration Device vs Merci Retriever in patients with acute ischaemic stroke (SWIFT): a randomized, parallel-group, non-inferiority trial. Lancet 2012;380: Berkhemer et al. A Randomized Trial of Intraarterial Treatment for Acute Ischemic Stroke. NEJM 2015;372: Goyal et al. Randomized Assessment of Rapid Endovascular Treatment of Ischemic Stroke. NEJM published online Feb 11, Campbell et al. Endovascular Therapy for Ischemic Stroke with Perfusion-Imaging Selection. NEJM published online on February 11, 2015.


Download ppt "Updates in Stroke Natalie Hendon MD Virginia Mason Medical Center Seattle, WA 2/21/2015."

Similar presentations


Ads by Google