There have been three large randomized trials of streptokinase in acute ischemic stroke treatment, all of which were terminated early because of increased sICH and mortality in the treatment group (Table 3.1).42-44
There has been a single large randomized trial of the defibrinogenating agent ancrod in acute ischemic stroke. The Stroke Treatment with Ancrod Trial (STAT)45 randomized acute ischemic stroke patients, presenting within 3 hours of symptom onset, to ancrod (n = 248) or placebo (n = 252). Ancrod is a purified fraction of Malaysian pit viper venom, which induces rapid defibrinogenation by splitting fibrinopeptide A from fibrinogen. It was given as a 72-hour infusion with a rate of 0.082-0.167 IU/kg, depending on the pretreatment fibrinogen level, targeting a plasma fibrinogen level of 1.18-2.03 mmol/L. The primary endpoint was favorable outcome at 90 days, defined as BI >95. More patients in the ancrod group, compared to placebo, had a favorable outcome (42% vs. 34%, p = 0.04). There was a trend toward more sICH in the ancrod group (5.2% vs. 2.0%, p = 0.06). Mortality was not different between the groups (25% for the ancrod group and 23% for the placebo group, p = 0.62). Ancrod has, however, not been adopted for routine use in acute ischemic stroke therapy in the United States.
SUBGROUP ANALYSES: WHO BENEFITS THE MOST FROM rt-PA TREATMENT?
The NINDS rt-PA Stroke Study Group performed analyses, combining the data from parts I and II of the study, to identify specific patient subgroups that may have a higher likelihood of benefit or harm from IV rt-PA.46 The hypotheses tested, using interaction terms within a logistic regression model, were whether there was a significant difference in the magnitude of the rt-PA effect within the subgroups. In the final model, the independent predictors of outcome were rt-PA treatment, increased age, NIHSS score, diabetes, admission mean arterial blood pressure, and pretreatment CT findings of hypodensity or a hyperdense vessel sign (suggesting the presence of intravascular thrombus). Interactions were found between age, NIHSS, and mean arterial pressure, such that increased age and higher NIHSS, and increased age and higher mean arterial pressure were associated with reduced odds of a favorable outcome. rt-PA treatment remained strongly and independently associated with increased odds of a favorable outcome (OR 2.02, 95% CI 1.45-2.81, p < 0.001). Importantly, there were no subgroups in which there was statistical evidence of a differential treatment effect of rt-PA. The subgroups tested included age, sex, race, stroke severity (measured by NIHSS), stroke subtype (categorized as cardioembolic, large artery disease, or small vessel disease), admission mean arterial blood pressure, history of diabetes, history of hypertension, history of previous stroke, and pretreatment CT findings of hypodensity or hyperdense vessel sign. An apparent lack of benefit for rt-PA in the 49 subjects aged more than 75 years with NIHSS >20 proved, on review, to be caused by a ceiling effect; none of the subjects in this group had a favorable outcome as defined by the trial protocol. When the data were inspected it was, however, apparent that rt-PA treatment in the subjects aged more than 75 years with NIHSS >20 was associated with better outcomes on the stroke-rating scales. For example, 30% of rt-PA-treated patients had a 90-day mRS <3, compared to 14% of placebo-treated patients.
A combined analysis of the ATLANTIS, ECASS-II, and NINDS rt-PA study data found that females had a greater benefit from rt-PA than males (p = 0.04), despite similar initial stroke severity and rates of sICH.47 This finding may not be relevant to the clinical, FDA-approved use of rt-PA, because most of the analyzed subjects from ATLANTIS and ECASS-II were randomized greater than 3 hours after stroke onset. Therefore, sex should not be a criterion for patient selection for thrombolysis.
The relevant subgroup analyses therefore provide no additional criteria for patient selection for IV rt-PA. Subgroup analyses, however, must be interpreted with caution because of the possibility of either type I error, resulting from multiple hypothesis testing due to the large number of subgroups, or type II error, resulting from testing hypotheses on subsets of the study data, in a study sample that was designed to have adequate power only for testing the main trial endpoint.
The Stroke-Thrombolytic Predictive Instrument (Stroke-TPI) has recently been developed in order to provide patient-specific estimates of the probability of a more favorable outcome with rt-PA, and has been proposed as a decision-making aid to patient selection for rt-PA.48 The estimates from this tool should, however, be treated with caution. The prediction rule is dependent on post hoc mathematical modeling, uses clinical trial data from subjects randomized beyond 3 hours who are not rt-PA-eligible according to FDA labeling and current best practice, and has not been externally validated. It is, therefore, not appropriate to exclude patients from rt-PA treatment based solely on Stroke-TPI predictions.
IV rt-PA USE IN THE COMMUNITY
rt-PA in the United States: Prevalence and Outcomes
In 1996, the United States FDA approved the use of rt-PA for acute ischemic stroke of less than 3 hours duration. An early observational study raised concerns that rt-PA therapy, when given outside the context of a research trial, may be associated with worse outcomes than in the NINDS trial.49 In this study, involving several Cleveland hospitals, the sICH rate of 15.7% compared unfavorably with the rate of 6.4% from the NINDS trial. Although neurologists were directly involved in 96% of IV rt-PA treatment decisions in this patient cohort, in 50% of the cases, there were deviations from national treatment guidelines. The most frequent deviations were the use of antiplatelet drugs or anticoagulants within 24 hours of rt-PA administration, and treatment beyond 3 hours after stroke onset. In contrast, other cohort studies have, for the most part, shown rates of sICH that are similar to the trial data.50, 51 A follow -up study from the Cleveland group showed that, after the initiation of a stroke quality improvement program, the rate of sICH decreased to 6.4%.52 In order to obtain a valid nationally representative estimate of the prevalence of rt-PA use, and the risk of rt-PA-associated sICH, the United States Centers for Disease Control (CDC) has sponsored the Paul Coverdell National Acute Stroke Registry.53 Data collected as part of the pilot prototype, involving multiple centers within four states (Georgia, Massachusetts, Michigan, and Ohio), showed an sICH rate of 0-6.1%.53 Therefore, the preponderance of the data suggests that rt-PA may be used safely in clinical practice, with rates of sICH similar to that in the NINDS clinical trial.
Despite being listed in the official recommendation and guidelines for stroke management by the American Stroke Association, the American Heart Association, the American Academy of Neurology, and the American College of Chest Physi-cians,38,54,55 the rate of IV rt-PA use in the community has been disappointingly low. Several single-center and multicenter convenience samples have reported that 1.6-9% of acute ischemic stroke patients received treatment.49,52,56-62 Studies using a nationally representative administrative database suggest the true treatment rate is even lower, in the range of 1-2% of all ischemic strokes.63, 64 There is probably hospital and geographic variation in the use of rt-PA,65 although little is known about the causes of this variability. In the pilot prototype of the CDC-sponsored Paul Coverdell National Acute Stroke Registry, the treatment rate varied from 3% in Georgia to 8.5% in Massachusetts.53 This was despite the fact that 20-25% of ischemic stroke patients arrived within 3 hours of symptom onset and had no documented contraindications to rt-PA.53 Only 10-20% of treated patients received the drug within 60 minutes of presentation to the ER,53 as recommended by guidelines.66 There is a strong suspicion that stroke systems of care are one of the factors that influence the safety and efficacy of delivery of IV rt-PA therapy.67,68 In 2000, the Brain Attack Coalition recommended criteria, based for the most part on consensus opinion rather than scientific evidence, for the establishment of primary stroke cen-ters.66 These criteria, summarized in Table 3.3, have formed the basis for a voluntary stroke certification program offered by the Joint Commission on the Accreditation of Healthcare Organizations.69 The departments of public health of several states, including New York, have incorporated similar criteria in state-based stroke certification programs. The initial experience in New York suggests that hospital compliance with certification is likely to be associated with improvements in care delivery, such as shortened door-to-CT time.70
The most common reason for lack of rt-PA use in otherwise eligible patients remains, however, delay in presentation to the hospital. The California Acute Stroke Pilot Registry (CASPR) investigators examined the effect of various hypothetical interventions on the rate of rt-PA use.71 Their data suggested that if all patients with a known time of onset presented to medical attention immediately, the expected overall rate of thrombolytic treatment within 3 hours would have increased from 4.3% to 28.6%. By comparison, the expected rate of treatment that would result from instantaneous prehospital response was 5.5%, from perfect hospital care was 11.5%, and from extension of time window to 6 hours was 8.3%.
TABLE 3.3 Brain Attack Coalition—Recommended Major Elements of a Primary Stroke Center.
Patient care areas Acute stroke teams Written care protocols Emergency medical services Emergency department Stroke unit
Neurosurgical services Support services
Stroke center director with support of medical organization Neuroimaging services Laboratory services
Outcome and quality improvement activities Continuing medical education
The authors concluded that campaigns to educate patients to seek treatment sooner should be major components of system-wide interventions to increase the rate of thrombolysis for acute ischemic stroke. There is some evidence that public education may help to increase the rate of rt-PA utilization by encouraging earlier presentation when stroke symptoms occur.68
IV rt-PA may be associated with a net cost savings to the health care system. Data from the NINDS rt-PA trial24 showed that hospital length-of-stay was shorter in the rt-PA-treated group (10.9 days vs. 12.4 days, p = 0.02) and more rt-PA patients were discharged to home than to inpatient rehabilitation or a nursing home (48% vs. 36% p = 0.002). A 1998 analysis used the NINDS rt-PA trial data and Medicare data to estimate, using Markov regression models, the costs associated with rt-PA therapy.72 Per 1000 treated patients, rt-PA use was associated with a model-predicted increase in hospitalization costs of $1.7 million United States dollars, a decrease in rehabilitation costs of $1.4 million, and a decrease in nursing home costs of $4.8 million. Multiway sensitivity analysis revealed a greater than 90% probability of cost savings. The estimated impact on long-term health outcomes was 564 (95% CI 3-850) quality-adjusted life-years saved, over 30 years, per 1000 patients.
Was this article helpful?