Stroke-MRI: Extending the Time-Window: Recent Trials and Clinical Practice

Abstract

Even 10 years after the approval of rtPA in the USA, rtPA is still widely underused. The main reasons for this are the fear of complications – especially thrombolysis-related intracranial hemorrhages (ICH), and the strict inclusion criteria, most prominently the narrow 3 h time-window. Thus, intensive research is going on to find ways to extend this window for treatment. All CT-based trials have so far failed to demonstrate an effect of thrombolysis beyond 3 h. The pooled analysis of the NINDS, ATLANTIS and the ECASS trials demonstrates a persistent effect up to 4·5 h but also shows that the treatment effect, not safety diminishes with time (1). This means that in the extended time-window, there are fewer patients who benefit from treatment. The question since then has been how to identify this patient group. During the last few years, multiparametric MRI techniques using diffusion weighted imaging (DWI) and perfusion sequences (PI) have been introduced into clinical practice to select patients for thrombolysis in the extended time-window (2). The rationale behind the MRI approach is to substitute an arbitrary time-window with an individual ‘tissue-window’ defined for the single patient based on imaging findings. In a simplified approach, the area of decreased diffusion on DWI represents the ischemic core of the infarct, and the perfusion/diffusion (PI-DWI) mismatch is believed to be a marker for critically hypoperfused yet potentially salvageable brain tissue (presumptive ischemic penumbra) representing the target for thrombolytic therapy (3). Although ongoing research works on refining this concept and improving the means to visualize the ‘tissue at risk’, one of the main advantages of the PI/DWI mismatch concept is simplicity; it is sufficiently accurate and therefore feasible in emergency conditions (4). In addition, it has been shown that MRI as a first-line imaging tool is sufficient to exclude ICH in the emergency setting (5, 6), a fact that needs to be emphasized as even experienced stroke centers still perform CT scans before bringing the patient into the MRI for further selection.

Recently, a study showing proof of principle for using the PI/DWI mismatch has been published (7). DEFUSE was a seven-center trial, which recruited 74 patients treated with rtPA within 3–6 h after symptom onset based on CT findings. An MRI was obtained immediately prior to and after rtPA. Only patients with a mismatch benefited from thrombolysis; matchers did not. Large infarctions already seen on DWI experienced fatal sICH upon reperfusion; small (lacunar) infarcts did reasonably well with or without a mismatch. Another multicenter trial is the ongoing EPITHET study (8). This study randomizes PI/DWI mismatch patients within 3–6 h to placebo or rtPA; results are expected in early 2007.

After the earlier smaller studies, recently, two large observational studies using stroke MRI in an extended time-window in clinical practice as well as two randomized phase II trials, – the DIAS and the DEDAS trial, have been published. Thomalla et al. compared outcome and symptomatic bleeding complications of intravenous rtPA within 6 h of symptom onset in MRI-selected patients with acute MCA infarction with the pooled data of the large stroke rtPA trials (9). Favorable outcome was more frequent in MRI-selected rtPA patients compared with pooled placebo and pooled rtPA patients. Interestingly, the rate of sICH in MRI-selected rtPA patients was lower than in the pooled rtPA group and comparable to the pooled placebo group. Our group compared, in a single-center study patients, who were treated with rtPA based on CT findings within 3 h with patients who received MRI-based treatment within, or later than, 3h (10). Clinical outcome and occurrence of sICH were prospectively assessed in 382 consecutive patients. Patients were divided into three groups: (1) CT-based <3h (N=209 patients); (2) MRI-based <3h (N=103 patients); and (3) MRI-based >3h (N=70 patients). The rate of independent outcomes in groups 1–3 was 47·8%, 50·5% and 55·7%. Mortality was trend-wise reduced, and sICH were significantly reduced in the MRI-based groups. MRI selected patients overall had a significantly lower risk for sICH and mortality. In multivariate analysis, only age and treating MRI based were significant predictors of sICH, whereas for independent outcome and mortality age, NIHSS score and sICH were predictive. Time to treatment proved to be irrelevant for all outcomes in univariate and multivariate analyses.

Two parallel phase II trials (DIAS, DEDAS) with a 3–9-h time-window using another thrombolytic drug (desmoteplase) have recently been completed. In DIAS, reperfusion rates up to 71·4% were observed with desmoteplase (125 μg/kg) compared with 19·2% with placebo (11). Favorable 90-day clinical outcome was found in 22·2% of placebo-treated patients and between 13·3% (62·5 μg/kg; P=0·757) and 60·0% (125 μg/kg; P=0·009) of desmoteplase-treated patients. Early reperfusion correlated favorably with clinical outcome (P=0·0028). Favorable outcome occurred in 52·5% of patients experiencing reperfusion vs. 24·6% of patients without reperfusion. The DEDAS study yielded confirmatory results. Interestingly, in both studies, imaging parameters parallel the clinical outcome, proving that the MRI parameters function as adequate surrogate parameters (12). Also, as shown in the other studies using an MRI-based selection process within 3–9 h it does not matter when the therapy is given. If a mismatch is present, time from symptom onset was not a treatment effect-modifying factor. The DIAS-2 as the follow-up phase III study is currently recruiting in two dosage arms (90 μg/kg, and 125 μg/kg) vs. placebo. As of October 2006, 156 patients have been included and another trial the DIAS-C(onfirmatory) trial is already in preparation.

Some results from the presented studies are worth a closer look. All studies show that despite the extended time-window, the treatment using an MRI-based selection algorithm is at least as safe if not safer than the standard CT-based treatment. So far, there are no data to explain this result. Possible explanations are the more reliable exclusion of large infarcts as well as the exclusion of stroke mimics with a higher risk for sICH (e.g., neoplasm). More data are needed to understand this effect. Another important result is that in neither study time to treatment was a relevant predictor regarding safety as well as treatment effect. This suggests that in the extended time-window and maybe even within 3 h, patient selection is more important than the plain time-window. In conclusion, MRI has the potential to extend the time-window for thrombolysis, thus helping to overcome one of the main obstacles for making this life-saving therapy available for a substantially higher number of stroke patients in the future.

References

Hacke

Donnan

Fieschi

. ATLANTIS Trials Investigators; ECASS Trials Investigators; NINDS rt-PA Study Group Investigators. Association of outcome with early stroke treatment: Pooled analysis of ATLANTIS, ECASS, and NINDS rt-PA stroke trials. Lancet 2004; 363: 768–74.

Hjort

Butcher

Davis

. UCLA Thrombolysis Investigators. Magnetic resonance imaging criteria for thrombolysis in acute cerebral infarct. Stroke 2005; 36: 388–97.

Schellinger

Fiebach

Hacke

: Imaging-based decision making in thrombolytic therapy for ischemic stroke: Present status. Stroke 2003; 34: 575–83.

Kang

Chalela

Dunn

Warach

: MRI screening before standard tissue plasminogen activator therapy is feasible and safe. Stroke 2005; 36: 1939–43.

Kidwell

Chalela

Saver

. Comparison of MRI and CT for detection of acute intracerebral hemorrhage. JAMA 2004; 292: 1823–30.

Fiebach

Schellinger

Gass

. Stroke Magnetic Resonance Imaging is accurate in hyperacute intracerebral hemorrhage: A multicentre study on the validity of stroke imaging. Stroke 2004; 35: 502–7.

Albers

Thijs

Wechsler

. Magnetic resonance imaging profiles predict clinical response to early reperfusion: The diffusion and perfusion imaging evaluation for understanding stroke evolution (DEFUSE) study. Ann Neurol 2006; 60: 508–17.

Davis

Donnan

: Echoplanar Imaging Thrombolysis Evaluation Trial: EPITHET. In International Stroke Conference, Orlando, FL, USA, 2006.

Thomalla

Schwark

Sobesky

. Outcome and symptomatic bleeding complications of intravenous thrombolysis within 6-hours in MRI-selected stroke patients: Comparison of a German multicentre study with the pooled data of ATLANTIS, ECASS, and NINDS tPA Trials. Stroke 2006; 37: 852–8.

10.

Kohrmann

Juttler

Fiebach

. MRI versus CT-based thrombolysis treatment within and beyond the 3-hour time window after stroke onset: A cohort study. Lancet Neurol 2006; 5: 661–7.

11.

Hacke

Albers

Al-Rawi

. The Desmoteplase in Acute Ischemic Stroke Trial (DIAS): A phase II MRI-based 9-hour window acute stroke thrombolysis trial with intravenous desmoteplase. Stroke 2005; 36: 66–73.

12.

Furlan

Eyding

Albers

. Dose Escalation of Desmoteplase for Acute Ischemic Stroke (DEDAS): Evidence of safety and efficacy 3–9-hours after stroke onset. Stroke 2006; 37: 1227–31.