Intravenous Thrombolysis for Acute Ischemic Stroke Associated to Extracranial Internal Carotid Artery Occlusion: The ICARO-2 StudyPaciaroni M.a · Agnelli G.a · Caso V.a · Pieroni A.b · Bovi P.c · Cappellari M.c · Zini A.d · Nichelli P.d · Inzitari D.e · Nesi M.e · Nencini P.e · Pezzini A.f · Padovani A.f · Tassinari T.g · Orlandi G.h · Chiti A.h · Gialdini G.h · Alberti A.a · Venti M.a · Acciarresi M.a · D’Amore C.a · Luda E.i · Tassi R.j · Martini G.j · Ferrarese C.k · Beretta S.k · Trentini C.k · Silvestrelli G.l · Lanari A.l · Previdi P.l · Ciccone A.l · DeLodovici M.L.m · Bono G.m · Galletti G.n · Marcheselli S.o · Del Sette M.p · Traverso E.p · Riva M.q · Silvestrini M.r · Cerqua R.r · Consoli D.s · Monaco S.t · Toni D.b
aStroke Unit and Division of Cardiovascular Medicine, University of Perugia, Perugia, bDepartment of Neurological Sciences, Sapienza University of Rome, Rome, cStroke Unit, Dipartimento di Scienze Neurologiche e della Visione, Azienda Ospedaliera-Universitaria, Verona, dStroke Unit, Neurology Clinic, Nuovo Ospedale Civile ‘S.Agostino – Estense’, University of Modena and Reggio Emilia, AUSL Modena, Modena, eDepartment of Neurological and Psychiatric Sciences, University of Florence, Florence, fDipartimento di Scienze Cliniche e Sperimentali, Clinica Neurologica, University of Brescia, Brescia, gStroke Unit – Department of Neurology, Santa Corona Hospital, Pietra Ligure, hClinica Neurologica – Azienda Ospedaliero-Universitaria, Pisa, iNeurology, Rivoli Hospital, Torino, jStroke Unit, AOU Senese, Siena, kSan Gerardo Hospital, University of Milano-Bicocca, Monza, lStroke Unit, Department of Neurology, Poma Hospital, Mantova, mStroke Unit, Neurology, Insubria University, Varese, nMorgagni-Pierantoni Hospital, Forlì, oStroke Unit, Humanitas Hospital, Milano, pStroke Unit, Department of Neurology, Sant’Andrea Hospital, La Spezia, qNeurology, Azienda Ospedaliera della Provincia di Lodi, Lodi, rDepartment of Neuroscience, Marche Polytechnic University, Ancona, sStroke Unit, Iazzolino Hospital, Vibo Valentia, and tStroke Unit, Ospedale Civico, Palermo, Italy Corresponding Author
Background and Purposes: In a case-control study in patients with acute ischemic stroke and extracranial internal carotid artery (eICA) occlusion, thrombolytic treatment was associated with increased mortality. The aim of this cohort study was to assess the efficacy and safety of thrombolysis in patients with eICA occlusion compared to those without eICA occlusion. Methods: Consecutive patients treated with intravenous tissue-type plasminogen activator within 4.5 h from symptom onset included in the Safe Implementation of Thrombolysis in Stroke – International Stroke Thrombolysis Registry (SITS-ISTR) in 20 Italian centres were analyzed. Acute carotid occlusion was diagnosed using ultrasound examination, angio-CT scan or angio-MRI. Since the SITS-ISTR database did not plan to report the site of vessel occlusion, each participating center provided the code of the patient with eICA occlusion. Patients were divided into 2 groups, those with and those without eICA occlusion. Main outcome measures were: death, disability (modified Rankin Scale, mRS, 3–6) and any intracranial bleeding at 3 months. Multiple logistic regression analysis was performed to reveal predictors for main outcomes. The following variables of interest were included in the analysis: presence of eICA occlusion, age, gender, diabetes mellitus, hyperlipidemia, atrial fibrillation, congestive heart failure, previous stroke, current smoking, antiplatelet treatment at stroke onset, baseline NIHSS score, baseline blood glucose, cholesterol and blood pressure, history of hypertension and stroke onset to treatment time. Results: A total of 1,761 patients without eICA occlusion and 137 with eICA occlusion were included in the study. At 3 months, 42 patients were lost to follow-up (3 with eICA occlusion). Death occurred in 30 (22.4%) patients with eICA occlusion and in 175 (10.2%) patients without (p < 0.0001). Death or disability at 3 months occurred in 91 of 134 patients with eICA occlusion (67.9%) compared with 654 of 1,722 patients without eICA occlusion (37.9%, p < 0.0001). No or minimal disability at 3 months (mRS 0–1) was reported in 25 (18.7%) patients with eICA occlusion and in 829 (48.2%) patients without (p < 0.0001). Any intracranial bleeding detected by CT or MRI at posttreatment imaging was seen in 16 (11.7%) patients with eICA occlusion and in 314 (17.8%) of those without (p = 0.09). The proportion of symptomatic intracerebral hemorrhage was 5.8% for patients with eICA occlusion and 8.0% for patients without (p = 0.16). At logistic regression analysis, eICA occlusion was associated with mortality (odds ratio, OR 5.7; 95% confidence interval, CI 2.9–11.1) and mortality or disability (OR 5.0; 95% CI 2.9–8.7) at 90 days. Conclusions: This cohort study in patients with acute ischemic stroke treated with thrombolysis showed an association between eICA occlusion and adverse outcome.
© 2012 S. Karger AG, Basel
Intravenous tissue-type plasminogen activator (rt-PA), if given within 4.5 h of symptom onset, improves the clinical outcome after acute ischemic stroke [1,2]. The number needed to treat to reverse one stroke completely at 3 months is 8 [3,4]; however, half of treated patients remain severely disabled or die. Early reperfusion saves tissue at risk of infarction and is a predictor of positive outcome. rt-PA-induced reperfusion may be influenced by several factors including time-to-treatment, stroke subtype, size and location of arterial occlusion. Angiography-controlled studies of intra-arterial or intravenous thrombolysis in acute ischemic stroke reported low reperfusion rates in patients with extracranial internal carotid artery (eICA) occlusion ; this suggests that in these patients, rt-PA treatment may be less beneficial. In the only randomized trial of intravenous thrombolysis which used baseline vascular imaging (EPITHET study), patients with ICA occlusion treated with rt-PA in the 3- to 6-hour window had worse outcomes compared to placebo-treated patients .
In a case-control study, patients with acute ischemic stroke associated with ICA occlusion had an improved functional outcome after intravenous thrombolysis with the trade-off of an increase in mortality and cerebral bleeding . The aim of this cohort study was to assess the efficacy and safety of thrombolysis in patients with eICA occlusion compared to those without eICA occlusion included in the Safe Implementation of Thrombolysis in Stroke – International Stroke Thrombolysis Registry (SITS-ISTR).
Patients and Methods
Patients with acute ischemic stroke were treated with rt-PA according to the provisional European license and local regulatory requirements. Patients with intra-arterial rescue thrombolysis were excluded from this analysis. Patient baseline characteristics, including stroke severity using the National Institutes of Health Stroke Scale (NIHSS) score, were recorded .
According to the modified Trial of Org 10172 in Acute Stroke Treatment (TOAST) classification , stroke etiologies were defined as either large vessel disease or nonlarge vessel disease (cardioembolic, small vessel disease, other determined etiologies, undetermined etiologies including multiple causes, and stroke mimic).
Outcomes were assessed at 3 months by the modified Rankin Scale (mRS) score. All patients had brain imaging before treatment and follow-up brain imaging was performed after 24–36 h to exclude any intracranial bleeding. This was categorized by using the ECASS trial definitions for hemorrhagic infarction (HI1, HI2), parenchymal hemorrhage (PH1, PH2), and remote parenchymal hemorrhage (PHr1, PHr2) . However, for the purpose of this analysis, all types of hemorrhagic transformations were summed up. Symptomatic intracerebral hemorrhage (sICH) was defined using the NINDS definition, that is, any increase in NIHSS score from baseline and any parenchymal intracerebral hemorrhage .
Acute carotid occlusion was diagnosed using ultrasound examination, angio-CT scan or angio-MRI. Since the SITS-ISTR database did not plan to report the site of vessel occlusion, each participating center provided the code of the patient with eICA occlusion. Patients were divided into 2 groups, those with and those without eICA occlusion. Baseline characteristics, the 90-day mortality/disability (mRS) and complication rates were compared between the 2 groups.
The primary efficacy outcome was disability at 90 days, as assessed by the mRS, dichotomized as favorable outcome (a score of 0–2) or unfavorable outcome (a score of 3–6). The score was assessed by clinical examination of the patient either by phone interview or letter reply form from the patient or a family member. Ninety-day mortality of patients with eICA occlusion and mortality of patients without eICA occlusion but with NIHSS score ≥10 were compared. This was planned to indirectly compare the outcome of patients with eICA occlusion with that of patients with presumed intracranial ICA occlusion . Safety outcomes included overall mortality at 90 days, any intracranial bleeding and symptomatic hemorrhages.
Descriptive statistics for baseline and demographic data were based on the presence or absence of eICA occlusion. Categorical variables were calculated by dividing the number of events by the total number of patients, excluding missing or unknown cases. Continuous variables were expressed as means and standard deviations.
Multiple logistic regression analysis was performed to reveal predictors for main outcomes. The following variables of interest were included in the analysis: presence of eICA occlusion, age, gender, diabetes mellitus, hyperlipidemia, atrial fibrillation, congestive heart failure, previous stroke, current smoking, antiplatelet treatment at stroke onset, baseline NIHSS score, baseline blood glucose, cholesterol and blood pressure, history of hypertension and stroke onset to treatment time.
Ninety-five percent confidence intervals (CI) were calculated for odds ratio (OR). In addition, multiple logistic regression analysis was performed to determine predictors for main outcomes in patients with eICA occlusion.
A total of 1,761 patients without eICA occlusion and 137 with eICA occlusion treated with thrombolysis within 4.5 h of stroke onset were included in this study. Table 1 shows the demographic and baseline characteristics, as well as the causes of stroke. Patients with eICA occlusion had more severe stroke on admission (3-point NIHSS score difference) and were younger with a higher proportion of males compared to patients without eICA occlusion. Additionally, patients with eICA occlusion had atrial fibrillation and hyperlipidemia less frequently and were more likely to be smokers.
|Table 1. Demographic and baseline patient characteristics|
At 3 months, 42 patients were lost to follow-up (3 with ICA occlusion), 540 patients (29.1%) were disabled and 205 had died (11.0%).
Death occurred in 30 (22.4%) patients with eICA occlusion and in 175 (10.2%) patients without eICA occlusion (p < 0.0001). Death or disability at 3 months occurred in 91 of 134 patients with eICA occlusion (67.9%) compared with 654 of 1,722 patients without eICA occlusion (37.9%; p < 0.0001). No or minimal disability at 3 months (mRS 0–1) was reported in 25 (18.7%) patients with eICA occlusion and in 829 (48.2%) patients without (p < 0.0001). Any intracranial bleeding detected by CT/MRI at posttreatment imaging was seen in 16 (11.7%) patients with eICA occlusion and in 314 (17.8%) of those without eICA occlusion (p = 0.09). The proportions of sICH were 5.8% for patients with eICA occlusion and 8.0% for patients without (p = 0.16; table 2). Table 3 shows the characteristics of the patients distinguished as alive and dead. The overall distribution of the mRS scores is shown in figure 1.
|Table 2. Outcome events|
|Table 3. Demographic and baseline characteristics of dead/alive patients|
|Fig. 1. Distribution of scores on the mRS at 3 months.|
Logistic regression analysis showed that the presence of eICA occlusion was associated with mortality at 90 days (OR 5.7; 95% CI 2.9–11.1). Other predictors of mortality were: older age (OR 1.03; 95% CI 1.00–1.06 for 1 added year), history of congestive heart failure (OR 10.5; 95% CI 1.5–71.0), high NIHSS score on admission (OR 1.12; 95% CI 1.07–1.17 for 1 added point of the score), high systolic blood pressure on admission (OR 1.01; 95% CI 1.00–1.03 for 1 added mm Hg), glucose on admission (OR 1.00; 95% CI 1.00–1.00 for 1 added mg/dl) and hemorrhagic transformation (OR 2.7; 95% CI 1.6–4.6). Furthermore, logistic regression analysis showed that the presence of eICA occlusion was also associated with adverse functional outcome at 90 days (mortality or disability) (OR 5.0; 95% CI 2.9–8.7). Other predictors of mortality or disability were: age (OR 1.02; 95% CI 1.01–1.04 for 1 added year), high NIHSS score on admission (OR 1.17; 95% CI 1.14–1.21 for 1 added point of the score), high systolic blood pressure on admission (OR 1.01; 95% CI 1.00–1.02 for 1 added mm Hg) and hemorrhagic transformation (OR 1.5; 95% CI 1.0–2.2).
In patients with eICA occlusion, logistic regression analysis revealed that glucose on admission (OR 1.03; 95% CI 1.00–1.05 for 1 added mg/dl of glucose) and baseline NIHSS score (OR 1.03; 95% CI 1.00–1.05 for 1 added point of the score), were associated with increased mortality, while age (OR 1.09; 95% CI 1.00–1.19 for 1 added year) and baseline NIHSS score (OR 1.33; 95% CI 1.12–1.58 for 1 added point of the score) were associated with adverse functional outcome at 90 days (mortality or disability).
Death occurred in 152 (15.1%) patients without eICA occlusion and NIHSS score ≥10 compared to a mortality rate of 22.4% in patients with eICA occlusion (χ2 p = 0.04).
In patients with acute ischemic stroke treated with rt-PA, this study found an association between eICA occlusion before treatment and adverse clinical outcome. In these patients, mortality was 2-fold higher in patients with eICA occlusion compared to patients without eICA occlusion. When considering functional outcome, more than 70% of patients with carotid occlusion treated with systemic thrombolysis had an mRS greater than 3 at 3 months after the acute stroke. The increase in mortality does not appear to be due to the increase in asymptomatic or symptomatic intracranial bleeding as these were lower in patients with eICA occlusion than in patients without eICA occlusion. The worst outcome seems to be due to the intrinsic pathophysiology of stroke triggered by carotid occlusion . Patients with eICA occlusion had a more severe stroke compared to patients without eICA occlusion (mean NIHSS score of 15 on admission compared to 12, respectively); the baseline NIHSS score is a very robust predictor of outcomes.
These results do not exclude the fact that thrombolysis may be beneficial in patients with eICA occlusion. This issue may be solved by comparing patients with eICA occlusion treated and not treated with systemic thrombolysis, which was the objective of the ICARO-1 study . In the ICARO-1 study, there was a high mortality rate in patients with eICA occlusion but, at 3-month follow-up, there was a significant increase in favorable outcome in thrombolysis-treated patients compared to those not treated.
Patients treated with thrombolysis in the ICARO-1 and ICARO-2 studies were comparable regarding both severity and risk factors. The only difference was seen in the rate of intracranial bleeding complications which was lower in the ICARO-2 study; however, this was probably due to the different definition of intracranial bleeding used in the two studies . Regarding the high rates of mortality in patients with eICA occlusion seen in the two ICARO studies, we hypothesize that CT angiography or other imaging examinations at the time of initial evaluation after stroke could predict who will respond poorly to systemic thrombolysis. A recent meta-analysis has shown that favorable outcomes (mRS 0–2) were significantly more frequent in patients who underwent endovascular treatment compared to those receiving intravenous thrombolysis alone (33.6 vs. 24.9%; p = 0.004). The rate of sICH was significantly higher in the endovascular group compared to the intravenous thrombolysis group (11.1 vs. 4.9%; p < 0.001). However, the mortality rate was not significantly different between the 2 groups (27.3% in the intravenous thrombolysis group vs. 32.0% in the endovascular group; p = 0.12) . Currently, randomized controlled trials are comparing intravenous thrombolysis with endovascular treatment in patients with acute occlusion of large vessels and their results are eagerly awaited in order to gain further insight.
In ICARO-2, patients with eICA occlusion seem to have had less classic cardiovascular risk factors compared with patients without eICA occlusion. In part, this may be explained by the fact that patients with eICA occlusion were younger. However, an expected higher incidence of unusual causes of stroke, including arterial dissection, was not seen.
ICARO-2 has some limitations. First, information on carotid occlusion was collected post hoc, though data were prospectively entered into the SITS-ISTR registry. Given this, there is the possibility of data underreporting. Second, the exact location of the occlusion of ICA and of any other associated intracranial vessel was not included in the database, a potential study limitation given the prognostic implications of the occlusion site of the cerebral artery [18,19]. Regarding the location of ICA occlusion, it is possible that some of the patients without eICA occlusion had an intracranial ICA occlusion. For this reason we compared the mortality rate of patients with eICA occlusion with patients without eICA occlusion but with an NIHSS score more than 10. The positive predictive value of an NIHSS score ≥10 is 97% for an occlusion of intracranial carotid artery. Also in this comparison, mortality was higher in patients with eICA occlusion. Another limitation of this study was the lack of data on the reperfusion of the involved vessels since recanalization is strongly associated with improved functional outcome and reduced mortality .
In conclusion, the present study shows that, despite treatment with intravenous rt-PA, outcomes among patients with stroke secondary to occlusion of the eICA remain poor. However, the presence of eICA occlusion should not deter physicians from giving thrombolysis, as further research is required to identify more effective treatments.
The study was conducted under the nonfinancial auspices of the Italian Stroke Association
G.A. received honoraria as a member of the speaker bureau of Astra Zeneca and Bayer. D.T. was paid expert testimony by Boehringer Ingelheim, Pfizer and Sanofi-Aventis. The other authors have no disclosures to report. The authors received no funding for this study.
Stroke Unit and Division of Cardiovascular Medicine
University of Perugia, Santa Maria della Misericordia Hospital
Sant’Andrea delle Fratte, IT–06126 Perugia (Italy)
Received: August 22, 2012
Accepted: October 11, 2012
Published online: December 1, 2012
Number of Print Pages : 6
Number of Figures : 1, Number of Tables : 3, Number of References : 20
Vol. 34, No. 5-6, Year 2012 (Cover Date: December 2012)
Journal Editor: Hennerici M.G. (Mannheim)
ISSN: 1015-9770 (Print), eISSN: 1421-9786 (Online)
For additional information: http://www.karger.com/CED