Saturday, December 05, 2015

      Meeting the Criteria for Intravenous Thrombolysis

Dean Reinke
Deans' Stroke Musing
Sunday, August 16, 2015

There is so much that can be improved here with just a little bit of research. Our doctors still do not understand the whole etiology of stroke damage, without understanding the problem there is no way to even attempt to fix it - Opening the Window of Time in Stroke Therapy: New Evidence for the Emergency Department.

Thrombolytic therapy for treating qualified patients presenting with acute ischemic stroke (AIS) has been available for 20 years. Outcomes are critically related to the time from symptom onset until intervention is provided. Although intravenous thrombolysis has assumed an important role in managing patients with AIS, several factors must be considered to use this treatment safely.

Time is Brain

The importance of time in managing patients with AIS is emphasized in guidelines published by major organizations, including the American Heart Association/American Stroke Association (AHA/ASA), that define time goals for identifying, evaluating, and treating stroke patients in the emergency department (ED) (Table 1). Of special importance is the achievement of a “door-to-needle” time of less than 1 hour (Not good enough - In the ambulance!). Underlying this goal, there is a graduated diminution in the effect of recombinant tissue plasminogen activator (rt-PA) on stroke outcome with every minute that goes by.

These goals are totally wrong, the BHAG is delivery of tPA in the ambulance - Big Hairy Audacious Goal

Looking at the major trials that have informed our knowledge of thrombolysis of stroke in the past 20 years, there appears to be a correlation between the time-window in which most of the patients in each trial were treated and the overall outcome of that trial (Figure 1, page 4). The National Institute of Neurological Disorders and Stroke (NINDS) trials were first, and reported a positive effect of rt-PA treatment given within 0 and 3 hours of symptom onset.2 Two parts of the study investigated 2 hypotheses, but otherwise followed identical protocols. Part 1 explored the 24-hour change in National Institutes of Health Stroke Scale (NIHSS) score, while Part 2 examined 3-month recovery using 4 outcome measures. The Barthel Index (BI) measures ability to perform activities of daily living, the modified Rankin scale (mRS) assesses overall function with a score of 0 indicating absence of symptoms and 5 representing severe disability, the Glascow outcome scale (GOS) is a global function assessment ranging from 1 (good recovery) to 5 (death), and the 42-point NIHSS quantifies neurologic deficit in 11 categories.

Although the improvement trend in 291 patients in Part 1 failed to meet the primary outcome of improvement by ≥4 points in the NIHSS score or a complete resolution of the neurologic defect, there were statistically significant differences in secondary outcomes of BI (odds ratio [OR] 1.8; 95% confidence interval [CI] 1.1 to 2.8; P=.012), mRS (OR 2.3; 95% CI 1.4 to 3.6; P<.001), GOS (OR 2.0; 95% CI 1.2 to 3.1; P=.005) and total NIHSS score (OR 2.2; 95% CI 1.3 to 3.7; P=.002). Data from 333 Part 2 patients showed similar benefit in a primary outcome using BI, mRS, GOS, and NIHSS, with increased odds of a favorable outcome at 3 months in patients given rt-PA compared with placebo (OR 1.7; 95% CI 1.2 to 2.6; P=.008). Mortality at 90 days was similar between groups (17% vs. 21%; P=.30).

The whole previous paragraph can be explained by not recognizing the neuronal cascade of death continues the first week because tPA administration does not address that.

Results in 3 subsequent trials that included patients treated with rt-PA or placebo between 0 and 6 hours of symptom onset (ECASS 1, ECASS 2, and ATLANTIS Part A) were neutral.3-5 However, in each of these trials, only a minority (not more than one-fifth) of patients were actually treated within 3 hours of stroke onset.

The European Cooperative Acute Stroke Study 1 (ECASS 1) enrolled 620 patients, of whom only 14% were <3 hours from symptom onset.5 The study failed to show a significant difference between groups for the primary outcome of 90-day BI and mRS, although a secondary endpoint of combined BI and mRS scores was significantly different between groups, favoring rt-PA (P<.001). ECASS 2 enrolled 800 patients, stratified between 0 to 3 hours and 3 to 6 hours after symptom onset, with 20% of patients enrolled in the 0 to 3-hour window, and the remainder in the 3 to 6-hour window.4 Although nonsignificant trends in favor of rt-PA were observed for all outcomes, the primary endpoint, dichotomized mRS outcome (with a score of 0 to 1 considered favorable, and 2 to 6 considered unfavorable) at 90 days, was similar between groups (40.3% vs. 36.6%; P=.277).

The ATLANTIS Part A trial enrolled 142 patients, with a primary efficacy endpoint of number of patients with a decrease of :;:4 points on the NIHSS at 24 hours and day 30, and total infarct volume at day 30.3 Forty-six (about a third) of the patients in this study were enrolled between 5 and 6 hours after onset, and the safety concerns due to hemorrhaging that emerged in these patients resulted in early termination of the study by the Data and Safety Monitoring Board (DSMB). In the study population as a whole, the significantly greater number of patients with ≥4 point NIHSS improvement in the rt-PA compared with the placebo group at 24 hours (40% vs. 21%; P=.02) was offset by fewer rt-PA compared with placebo group patients having this outcome at 30 days (60% vs. 75%; P=.05). None of the efficacy endpoints, including functional recovery, favored rt-PA at 30 and 90 days, while symptomatic intracerebral hemorrhage (sICH) within 10 days (11% vs. 0%; P<.01) and 90-day mortality (23% vs. 7%; P<.01) were significantly increased in rt-PA compared with placebo patients. In the subgroup of rt-PA (n=22) and placebo (n=24) patients treated in the 5- to 6-hour window, both 30-day (27.3% vs. 4.2%; P=.03) and 90-day (36.1% vs. 4.2%; P=.01) mortality and rate of sICH at day 10 (18.2% vs. 0%; P=.03) were significantly greater in the rt-PA group. Significantly more rt-PA (23%) compared with placebo (8%; P<.05) patients in the 5- to 6-hour window had baseline NIHSS scores >20; which may have confounded the results. However, overall, only 15% of the patients were enrolled within 3 hours, and the negative results of this study were therefore attributed to patients treated >3 hours after stroke onset.

ATLANTIS Part B, which was a reconfiguration of the ATLANTIS trial based on the safety experience outlined above, and the interval FDA approval of rt-PA for the treatment of acute ischemic stroke within 3 hours of onset, had a treatment window restricted to 3 to 5 hours after symptom onset.6 The primary outcome was proportion of patients with an excellent neurologic recovery at day 90 (NIHSS ≤1). Following an interim analysis, ATLANTIS Part B was terminated for futility by the DSMB after 547 patients had received study medication, prior to completing the targeted enrollment of 968 patients. Significant benefits of rt-PA treatment on the 90-day primary and secondary efficacy endpoints were not observed. Patients in the rt-PA group, compared with placebo, had significantly increased sICH (7.0% vs. 1.1%; P<.001), asymptomatic ICH (11.4% vs. 4.7%; P=.004), and fatal ICH (3.0% vs. 0.3%; P<.001) within the first 10 days, although 90-day mortality was similar between groups (11.0% vs. 6.9%; P=.09).

ECASS 3 enrolled 821 patients randomized to alteplase or placebo treatment between 3 and 4.5 hours after stroke onset.7 The primary endpoint was dichotomized mRS (0 to 1 favorable, 2 to 6 unfavorable outcome) at 90 days. Significantly more patients in the rt-PA group had a favorable outcome compared with placebo (52.4% vs. 45.2%; P=.04). Mortality was similar in rt-PA and placebo groups (7.7% vs. 8.4%; P=.68), while intracranial hemorrhage (27.0% vs. 17.6%; P=.001), including sICH (2.4% vs. 0.2%; P=.008), was significantly greater in the rt-PA group.

A pooled analysis was performed that included individual patient data from the NINDS, ECASS (1-3), ATLANTIS, and EPIPHET trials, all of which compared intravenous rt-PA with placebo.8 A pooled analysis is different from a meta-analysis in that a meta-analysis assesses individual trial outcomes to provide an overall estimate of the effect size for the variables of interest, with a weighted contribution from each study; a pooled analysis is not analyzing individual trial outcomes, but is instead analyzing the individual patient data from subjects in those trials. A pooled analysis is predicated on the contributing studies having patient populations, interventions, and outcome measures that are similar enough that individual patient data can be pooled and analyzed together, as if obtained in a single “mega-trial,” which can provide more reliable results than can a standard meta-analysis.9

When the data for approximately 3,500 patients were combined for the pooled analysis, the odds of a favorable 3-month outcome with rt-PA treatment decreased as the time from onset to treatment increased, with no benefit observed after approximately 4.5 hours (Table 2).8 Adjusted odds of mortality increased with time from onset to start of treatment (P=.0444). The incidence of parenchymal hemorrhage was also significantly increased, with an adjusted OR of 5.37 (95% CI: 3.22, 8.95; P<.0001) for treatment started 0 to 360 minutes from stroke onset. A similar hemorrhage incidence was observed across the four 90-minute time windows studied.

Using rt-PA Safely: AHA/ASA 2013 Guidelines

The AHA/ASA guidelines for the early management of patients with AIS, updated in 2013, emphasize the importance of early treatment.1 However, the AHA/ASA policy statement on emergency medical services within stroke systems of care notes that symptom onset time, essential to determining treatment time from onset, is a common documentation omission.10
The guidelines define patient inclusion and exclusion characteristics, and stress the importance of documenting eligibility assessment. Eligibility characteristics are divided between patients for whom treatment can be initiated within 3 hours of symptom onset, and eligibility restrictions when treatment initiation would occur between 3 and 4.5 hours after symptom onset.

Exclusion Criteria Highlights

Leaky Brain or Leaking Body

Exclusion criteria for rt-PA can be divided into broad categories. The first considerations have to do with “leaky brain” (recent CNS parenchymal injury from stroke, trauma, or surgery) and “leaking body” (ie, active bleeding).

Any bleeding, including any that occurs outside the brain, excludes a patient from rt-PA treatment. This exclusion is further qualified to include any significant head trauma or prior stroke in the previous 3 months, symptoms suggesting subarachnoid hemorrhage, recent intracranial or intraspinal surgery, active internal bleeding, or arterial puncture at a noncompressible site in the previous 7 days.

Bad Brain

Conditions associated with compromised blood vessel integrity in the brain, such as tumors, arteriovenous malformation, and aneurysm (“bad brain”) are exclusion criteria for rt-PA treatment. This “bad brain” category also includes a radiographic correlate, a CT scan that demonstrates multilobar infarction; frank hypodensity in over one-third of the cerebral hemisphere on a non-contrast head CT is an exclusion criterion for rt-PA treatment.

Bleeding Diathesis

Exclusions to rt-PA therapy also include a platelet count <100,000/mm3, heparin within 48 hours with an elevated aPTT, or current use of an anticoagulant with an INR >1.7 or PT >15 seconds. If the patient is taking a direct thrombin or factor Xa inhibitor, laboratory markers such as a thrombin time and/or ecarin clotting time can be considered if feasible to obtain within a treatment time-frame, and if elevated, would qualify as exclusions.

Other Exclusions

Patients with blood glucose <50 mg/dL (2.7 mmo/L) or elevated blood pressure (systolic >185 mm Hg or diastolic >110 mm Hg) that cannot be readily controlled with a dose or 2 of a medication, such as labetolol, or a nicardipine drip, should not be given rt-PA treatment.

Relative rt-PA Exclusion Criteria for Treatment Within 3 Hours of Symptom Onset

There are also relative exclusion criteria, entities that are not absolute exclusions per se but war risks and benefits of administering rt-PA for a stroke within 3 hours of symptom onset. These relative contraindications include minor or rapidly improving symptoms, pregnancy, seizure at onset with residual neurological impairments, major surgery or serious trauma within the previous 14 days, recent gastrointestinal or urinary tract hemorrhage (within the previous 21 days), or recent acute myocardial infarction (within the previous 3 months).

Exclusions for Using Intravenous rt-PA 3 to 4.5 Hours after Symptom Onset

If a patient can be treated in the time period from 3 to 4.5 hours after stroke onset, the general exclusions are supplemented by additional relative exclusion criteria. These include patients aged >80 years, with a severe stroke (NIHSS >25), with imaging evidence of ischemic injury involving more than one-third of the middle cerebral artery territory, taking an oral anticoagulant regardless of INR, or with a history of both diabetes and prior ischemic stroke.

Safety of rt-PA Treatment in Stroke Mimics

Earlier treatment may increase the risk of providing rt-PA to a patient who presents with clinical features of a stroke but ultimately has an alternative diagnosis. Studies have shown that rt-PA treatment is not associated with adverse outcomes when given to these patients. In a retrospective study of approximately 70 stroke mimics treated with rt-PA, there were no instances of sICH.11 In a large study of 5581 consecutive patients treated with rt-PA, 100 (1.8%) were determined to be stroke mimics.12 An sICH occurred in 1 stroke mimic patient for a rate of 1.0%, compared with 7.9% in patients who had ischemic strokes.

Additional Treatment Decision Considerations

A patient with an acute ischemic stroke and no exclusion criteria, presenting within a window of time such that treatment can be initiated within 3 or 4.5 hours of stroke onset, may be a straightforward candidate for intravenous thrombolysis. There may, at times, be considerations that require additional clinical judgment, such as whether or not a patient may have a stroke that is too mild, or a stroke that is too severe, or if the patient may be too old.

Mild Patient

Deciding to use rt-PA in a patient with a mild stroke requires careful balancing of the potential benefits with the risk of sICH with rt-PA treatment. While the risk of sICH is approximately 6% overall, this risk is affected by stroke severity; a patient with a mild stroke, that is, with a baseline NIHSS of 0 to 5, has an approximate 2% risk of sICH, with a slight increase to approximately 3% in a patient with a mild-to-moderate stroke (NIHSS 6-10) (Figure 2).13-15 Contrasting this with the approximate 20% risk of having a poor outcome without treatment can help put things in perspective when considering if the potential outcome benefit is worth the risk of giving rt-PA to a patient who has had a mild stroke.

“Mild” strokes come in 2 varieties. The first is a stroke that is truly mild. The overall assessment of benefit versus risk in this first type needs to incorporate the possibility that a patient who appears to be minimally affected in the ED may not consider his condition to be mild after 6 weeks or 6 months as he is confronted with restrictions in his daily activities. Although the stroke may be associated with an isolated deficit, its focal impact may have a profound negative effect on the patient’s livelihood and quality of life, and in such circumstances the benefits of treating with rt-PA may outweigh the risk. Other “downstream” factors that often are not considered in the ED include depression,16,17 impaired executive function,18 and unmeasured motor dysfunction.19 These aspects of potential long-term disability should be considered when contemplating rt-PA treatment for a mild stroke.

Alternatively, a patient may appear to be having a “mild” stroke that looks mild, but in actuality is not a mild stroke at all. The patient may have mild symptoms that may represent a proverbial “calm before the storm” in an ongoing major vessel occlusion. Collateral vessels supplying the compromised brain tissue may be sustaining a patient like this, obscuring the full extent of the neurological deficit due. The literature suggests that up to 17% of ED stroke patients who are considered “too mild to treat” will acutely and significantly deteriorate after several hours, as these collaterals begin to fail.20-22 Approximately two-thirds of these deteriorating patients show rapid improvement from a significant deficit initially. The profile of the “too mild to treat” ED patient, who will progressively deteriorate appears to be of rapid improvement from a significant deficit to a mild one, accompanied by either (1) imaging that reveals a large-vessel occlusion or stenosis (especially if present while the deficit is mild) or (2) symptoms consistent with a large vessel occlusion or cortical stroke (such weakness combined with numbness, or weakness combined with aphasia). The exclusion of patients with this profile from thrombolytic therapy based on a classification as “too mid to treat” or “rapidly improving” should be done judiciously.

Patient with a Severe Stroke

Determining if a patient is too severe to treat with rt-PA includes consideration of NIHSS score and neuroimaging results, in a context of time from symptom onset. The alteplase prescribing information warns that patients with severe neurological deficit (eg, NIHSS >22) at presentation are in particular need of a diligent assessment of risks versus benefits.23 The hemorrhage risk increases from 5% at NIHSS 11 to 20 to reach 17% at NIHSS >20.14

Risks should be carefully weighed against benefits if CT imaging shows major early infarct signs (eg, substantial edema, mass effect, or midline shift).23 Data from the ECASS trial showed that patients with frank hypodensity covering greater than one-third of the territory of the MCA had an 8-fold increased rate of hemorrhage with rt-PA compared with those who were not rt-PA-treated.24 Accordingly, numerous major guidelines worldwide include this as a treatment exclusion criterion.25-30

Time since stroke onset has a major role in the treatment decision for these patients and clinical judgement has to be applied to the interplay of time, stroke severity, and imaging findings. For example, a patient who presents within a half hour of onset with an NIHSS of 25 and a normal CT scan may be a candidate for rt-PA despite the severity of the stroke. Alternatively, for a patient presenting 2.5 to 3.5 hours after onset, with a midrange NIHSS and a CT scan already showing hypodensities, the risks of giving rt-PA treatment may outweigh the benefits.

Influence of Patient Age on Treatment Choice

Older patients have worse outcomes after stroke than their younger counterparts.31-34 Their strokes are more severe, they deteriorate more readily, and they do not respond to therapies as well. These characteristics may be confounded by the tendency towards less aggressive management of elderly patients, as fewer investigative and invasive procedures are done with these patients.35 Primary randomized clinical trial literature showed that age >80 years was associated with an increased risk of ICH with rt-PA.36,37 That being said, although some studies showed rt-PA treatment is more effective in younger patients,38-40 older patients have been shown to benefit from rt-PA treatment.35,40 This is particularly evident in the more recent literature.41-44 In addition, more recent clinical reporting suggests that an age of 80 is not associated with an increased risk of sICH in selected patients treated with rt-PA.40,45-51 The caveat to these latter reports is that the data they are based on was largely obtained outside of randomized controlled trials and may be affected by selection bias; even if so, they still demonstrate that, from a real-world perspective, the medical community at large is selecting older patients to treat with rt-PA in a manner that leads to favorable outcomes and has a comparable safety profile to that in younger patients.


Time is brain. There is a minute-by-minute reduction of good outcome achievement with rt-PA, until it reaches a nadir by 4.5 hours from onset. Considerations for using rt-PA in patients with a mild stroke include that small strokes can have big consequences for the patient, and rapid improvement of cortical/large artery strokes may reverse within hours. In patients with a severe stroke, caution should be taken when choosing rt-PA for patients with NIHSS >22 who can be treated within 3 hours. Patients with NIHSS >25 who present >3 hours after onset of symptoms should not be given rt-PA (without careful consideration of the risks and benefits). Regardless of interval from onset to presentation, rt-PA should be avoided in patients with a CT hypodensity >1/3 of the hemisphere.

Patients of any age who present within 3 hours may be given rt-PA; however, caution is important when treating patients >80 years of age. If 3 to 4.5 hours have elapsed since symptom onset, extreme care is warranted when giving rt-PA treatment in patients aged >>80 years.

See the original article:

No comments:

Post a Comment