REBEL Cast Ep72: Systemic Alteplase in Stroke Mimics is Safe?

Background: Currently, alteplase is the mainstay of treatment of acute ischemic stroke.  Advocates of alteplase suggest that the benefit of alteplase is greatest when given early and declines with increasing time from stroke symptom onset (i.e. time is brain).  Therefore, the AHA/ASA guidelines recommend intravenous alteplase within 4.5 hours after stroke onset, which is based on very weak evidence (i.e. NINDS & ECASS III).

Due to weak evidence in support of it’s use and significant patient risks associated with alteplase, it’s use in acute ischemic stroke remains controversial.  One of the big issues is that by decreasing the time for evaluation and treatment, there is an increased risk of administrating alteplase to patients presenting with noncerebrovascular conditions that can resemble an acute ischemic stroke (i.e. stroke mimics).  This puts patients with no chance of improvement with alteplase at risk for increased mortality and symptomatic ICH.  There is some limited data on the safety of alteplase in stroke mimics and this study adds to that knowledge.

REBEL Cast Episode 72: Systemic Alteplase in Stroke Mimics is Safe?

What They Did:

• Used data from the Get With The Guidelines – Stroke Registry
• 72,582 patients with suspected ischemic stroked treated with alteplase from 485 US hospitals
• Documented use of alteplase in stroke mimics (i.e. patients who present with stroke-like symptoms, but after workup are determined not to have suffered from a stroke or TIA)
• Compared outcomes in stroke mimics versus acute ischemic stroke

Outcomes:

• Primary Safety Outcomes:
  • Symptomatic ICH (sICH) – Defined as ICH within 36 hours documented on CT or MRI with clinical deterioration attributable to hemorrhage
  • Life-threatening or serious systemic hemorrhage
  • Other serious complications
  • Composite endpoint of any alteplase complications
• In-Hospital Outcomes:
  • In-hospital mortality
  • Discharge destination (home, hospice, skilled nursing facility, or inpatient rehabilitation facility)
  • Independent ambulation at discharge 

Inclusion:

• Patients with suspected ischemic stroke
• Receiving IV alteplase
• Part of the Get With The Guidelines-Stroke Hospitals

Exclusion:

• Patients transferred from another hospital
• Had received alteplase at an outside hospital
• Undergoing investigational or experimental protocols or thrombolysis
• Received catheter-based reperfusion

Results:

• 72,582 patients with suspected stroke treated with IV alteplase
  • 2517 (3.5%) of alteplase treatments were given to stroke mimics
  • Baseline Characteristics of Stroke Mimic Patients vs Acute Ischemic Stroke Patients:
    • Median Age 54 years vs 71 years
    • Median NIHSS Score 5 vs 8
    • Median Time from Stroke Symptom Onset to Hospital Arrival 66 min vs 62 min
  • Most Common Final Non-Stroke Diagnoses:
    • Migraine 19.0%
    • Functional Disorder 9.4%
    • Seizure 7.9%
  • Symptomatic ICH:
    • Stroke Mimics: 0.4%
    • Acute Ischemic Stroke: 3.5%
    • aOR 0.29; 95% CI 0.17 – 0.50; p<0.001
  • Composite Endpoint of Any tPA Complication:
    • Stroke Mimic: 1.5%
    • Acute Ischemic Stroke: 6.9%
    • aOR 0.48; 95% CI 0.36 – 0.64; p<0.001
  • In-Hospital Mortality:
    • Stroke Mimics: 0.8%
    • Acute Ischemic Stroke: 6.2%
    • aOR 0.31; 95% CI 0.20 – 0.49; p<0.001
  • Ability to Ambulate at Discharge:
    • Stroke Mimic: 78.6%
    • Acute Ischemic Stroke: 50.6%
    • aOR 1.86; 95% CI 1.61 – 2.14; p<0.001

Strengths:

• Largest multicenter investigation of stroke mimics to date
• Asks a clinically important question of risk of harm from alteplase in stroke mimics

Limitations:

• Large reporting bias. Submitting stroke mimic cases to the registry were encouraged but optional, meaning many cases may not have been documented.
• The definition of sICH is open to bias due to the fact that this was an unblinded study. Reporters knew if patients were a stroke mimic, whether they received alteplase, etc
• The analysis was limited to hospitals that were submitting mimic cases by requiring at least 10 IV alteplase cases and 5 stroke mimics during the entire stroke period, but many stroke mimic cases may still have been missed in this analysis. This creates a selection bias as smaller community hospitals with lower volumes are excluded from the analysis
• There is no validated consensus on the definition of stroke mimics, therefore the true rate of alteplase-treated stroke mimics is most likely underestimated in this trial as many patients may have been classified as TIAs
• This study does not give us information on stroke mimic cases treated with tPA compared with stroke mimics not treated with alteplase.
• Get With The Guidelines-Stroke is a voluntary program so only hospitals that were interested in quality improvement in stroke care or had the capacity to fulfill requirements participated.

Discussion:

• At my hospital we get 3 – 8 suspected acute ischemic stroke cases a day. So, doing some math this is about 1,000 patients a year on the low end and 2800 patients a year on the high end.
  • This would equate to 4 to 11 sICH cases in patients with stroke mimics per year using the numbers from this study.
  • If we do the same calculation for inpatient mortality this would equate to 8 to 23 patients with in-hospital mortality who had a stroke mimic diagnosis.
  • In this publication the authors state there is a trend of increasing alteplase administration to stroke mimics in more recent years and centers not seeing as many stroke cases (<30cases/y) are more likely to give alteplase to mimics as well. So missing data plus increasing rate of alteplase administration would mean that the above calculations probably grossly underestimate the real risks of sICH and inpatient mortality in stroke mimic patients receiving IV alteplase
• Increased OR in low volume hospitals shines a light on the fact that figuring out who is eligible for alteplase is tricky. The selection of patients we see in studies will always be better than what we see in clinical practice
• It’s interesting to read in the discussion section that the authors state that if the estimate of stroke mimics is restricted to “specific” non-stroke diagnoses (i.e. migraine) the rate of stroke mimics would be 1.3%. This is absolutely ridiculous in my opinion.  It’s either a stroke or it’s not.  Who cares if there is a specific diagnosis.
• The authors perform a bunch of logistic regression models to find predictors of stroke mimic vs stroke. However none of the numbers are significant enough that I would base any decisions on alteplase off of them
• The authors of this publication still discuss acute ischemic stroke as a time is brain paradigm, making a strong emphasis on reducing door to needle times to increase treatment effectiveness in stroke, however more recent literature shows that perfusion based imaging may be a better and safer way to move forward in the administration of alteplase, similar to what we are seeing in endovascular studies.
• Faster treatment would translate into more eligible patients. Interestingly in this publication, the stress is on getting alteplase to more patients, but not improving outcomes. However as previous studies have shown, faster is not always better [2]. Decreasing door to balloon times in STEMI increased false positive STEMI activations with an increase in in-hospital mortality seen in patients with false positive STEMIs.
• When a patient stands no benefit to a treatment, there is only potential for harm. Any harm that a patient with a stroke mimic endures from alteplase should be thought of as unacceptable.  We shouldn’t be downplaying the risks of alteplase but rather working to improve our identification of actual stroke and differentiation from stroke mimics.  NO BENEFIT = NO Risk Acceptable Simply Because We Are Rushing!!!

Author Conclusion: “In this large cohort of patients treated with tPA, relatively few patients who received tPA for presumed stroke were ultimately not diagnosed with a stroke or transient ischemic attack.  The complication rates associated with tPA in stroke mimics were low. Despite the potential risk of administering tPA to stroke mimics, opportunity remains for continued improvement in the rapid and accurate diagnosis and treatment of ischemic stroke.” 

Clinical Take Home Point: We disagree with these authors conclusions that state the complication rates associated with alteplase in stroke mimics were low.  Reporting bias (not all stroke mimic cases reported), selection bias (exclusion of smaller hospitals with lesser patient volumes), and rates of increasing alteplase administration all lend themselves to gross underestimation of harms of alteplase in stroke mimics in patients who have zero chance of benefit from this medication.  Furthermore, stroke care should be moving to a perfusion-based paradigm and away from a time is brain paradigm which has consistently been replicated in endovascular studies in acute ischemic stroke.

References:

1. Ali-Ahmed F et al. Intravenous Tissue Plasminogen Activator in Stroke Mimics: Findings From the Get With the Guidelines Stroke Registry. Circ Cardiovas Qual Outcomes 2019. PMID: 31412730
2. Fanari Z et al. Aggressive Measures to Decrease “Door to Balloon” Time and Incidence of Unnecessary Cardiac Catheterization: Potential Risks and Role of Quality Improvement. Mayo Clin Proc. 2015 [epub ahead of print] PMID: 26549506
3. Meurer WJ et al. Stroke Thrombolysis: Mimicry, Accuracy, Safety, and 1.9 Million Neurons Per Minute. Circ Cardiovas Qual Outcomes 2019. PMID: 31412736

Post Peer Reviewed By: Anand Swaminathan, MD (Twitter: @EMSwami)