Acute Ischemic Stroke and Long-Term Outcome After Thrombolysis
Nationwide Propensity Score–Matched Follow-Up Study
Background and Purpose—Data on long-term outcome after intravenous tissue-type plasminogen activator (tPA) in ischemic stroke are limited. We examined the risk of long-term mortality, recurrent ischemic stroke, and major bleeding, including intracranial and gastrointestinal bleeding, in intravenous tPA-treated patients when compared with intravenous tPA eligible but nontreated patients with ischemic stroke.
Methods—We conducted a register-based nationwide propensity score–matched follow-up study among patients with ischemic stroke in Denmark (2004–2011). Cox regression analysis was used to compute adjusted hazard ratios for all outcomes.
Results—Among 4292 ischemic strokes (2146 intravenous tPA-treated and 2146 propensity score–matched nonintravenous tPA-treated patients), with a follow-up for a median of 1.4 years, treatment with intravenous tPA was associated with a lower risk of long-term mortality (adjusted hazard ratio, 0.66; 95% confidence interval, 0.49–0.88). The long-term risk of recurrent ischemic stroke (adjusted hazard ratio, 1.05; 95% confidence interval, 0.68–1.64) and major bleeding (adjusted hazard ratio, 0.59; 95% confidence interval, 0.24–1.47) did not differ significantly between the intravenous tPA-treated and nontreated patients.
Conclusions—Treatment with intravenous tPA in patients with ischemic stroke was associated with improved long-term survival.
Thrombolytic therapy with intravenous tissue-type plasminogen activator (tPA) is an effective treatment in acute ischemic stroke.1,2 Several studies have examined functional outcome and mortality at 3 months after intravenous tPA treatment and found that treatment improves neurological outcome but has no effect on mortality.1–4 In contrast, data on long-term outcome are limited.5–9 Treatment with intravenous tPA has been introduced gradually in the Danish stroke units offering a unique opportunity to compare the long-term outcome, including mortality, readmission with ischemic stroke, and major bleeding among intravenous tPA-treated and propensity score–matched intravenous tPA eligible, but nontreated patients with ischemic stroke in routine clinical practice.
We conducted a nationwide register-based follow-up study. The study was approved by the Danish Data Protection Agency (2011-41-6095).
The Danish Stroke Registry was used to identify all patients (≥18 years) admitted with ischemic stroke between 2004 and 2011, n=81477 (Figure I in the online-only Data Supplement). The registration of patients in the Danish Stroke Registry is mandatory for all hospitals treating patients with acute stroke. The sensitivity and predictive value of the registration of patients with acute stroke are >90%.10 Treatment with intravenous tPA was approved in Denmark on April 1, 2004, and eligible patients were treated with intravenous tPA according to the existing national guidelines in the study period.11 Treatment was initially restricted to a few stroke units because of available resources and was fully implemented nationwide to all eligible patients in 2008. We restricted the cohort of intravenous tPA-eligible but nontreated patients to those admitted to stroke units not offering intravenous tPA between 2004 and 2006. Patients with missing information about symptom onset or admission time were excluded. A total of 2146 intravenous tPA-treated patients could be propensity score–matched with 2146 intravenous tPA-eligible patients with acute ischemic stroke not offered intravenous tPA treatment.
The outcomes included all-cause mortality, admission with recurrent ischemic stroke, and major bleeding, defined as intracranial and gastrointestinal bleeding. Information on mortality was obtained by linkage with the Danish Civil Registration System. The Danish Stroke Registry provided data on admissions with recurrent ischemic stroke, whereas data on major bleedings were obtained from the Danish National Registry of Patients.
The Danish Stroke Registry provided information about sex, age, stroke severity, time from symptom onset to admission, vascular risk factors (previous or current atrial fibrillation, hypertension, diabetes mellitus, previous stroke, and smoking), living arrangements (living with someone or alone), and housing (own home or nursing home). Stroke severity was assessed using the Scandinavian Stroke Scale score, which ranges from 0 (coma with quadriplegia) to a total of 58 (intact neurological status). Data on the quality of in-hospital stroke care were obtained as a percentage of provided recommended early processes of care, including admission to a stroke unit (day of admission), administration of antiplatelet (second day of admission) or anticoagulant therapy (within 14 days), examination with computed tomography or MRI (day of admission), and assessment by a physiotherapist and an occupational therapist and assessment of nutritional risk (second day of admission). We only assessed the processes for patients without contraindications (eg, early antiplatelet therapy was contraindicated in intravenous tPA-treated patients). We computed the Charlson comorbidity index score for each patient based on all discharge diagnoses registered within the past 10 years before stroke admission in the Danish National Registry of Patients.12 The Charlson comorbidity index score was categorized according to the severity of comorbidity: low (0 comorbidities), moderate (1–2 comorbidities)’, and high (>2 comorbidities). From the Danish Medicines Agency’s Medical Register, we traced all prescriptions for antiplatelet drugs, vitamin K antagonists, statins, and blood pressure–lowering drugs that were filled during follow-up.
A propensity score was calculated for each patient based on key baseline covariates with a known association with intravenous tPA treatment or the outcomes studied, including age, Charlson comorbidity index score, and stroke severity.13 Each intravenous tPA-treated patient was matched with 1 nonintravenous tPA-treated patient with a similar propensity score (5-1-digit matching, Greedy method). An absolute standardized difference <10% and a variance ratio between 0.8 and 1.25 were considered to support the assumption of balance between the groups (Figures II and III in the online-only Data Supplement). All patients were followed up from the date of admission until the date of outcome or December 31, 2011, whichever came first. The outcomes were compared using multivariable Cox regression analysis stratified for matched pairs. Age and time from onset to admission were included as continuous variables, stroke preventive drugs as time-dependent covariates, and all other covariates as categorical variables (Table I in the online-only Data Supplement). Stroke severity according to the Scandinavian Stroke Scale score was stratified into 0 to 14 (very severe), 15 to 29 (severe), 30 to 44 (moderate), and 45 to 58 (mild) for the adjusted analysis.
Because of partly different inclusion periods for intravenous tPA-treated and nontreated patients, the maximum follow-up time for each propensity score–matched pair could potentially diverge. Therefore, we ensured identical maximum follow-up time for each pair defined by the shortest maximum follow-up time in the match. All analyses were performed using STATA version 12.1.
Table 1 presents the patient characteristics (Table I in the online-only Data Supplement). During a median follow-up time of 1.4 years (range, 0–7.6 years), a total of 633 patients (14.8%) died (Table 2). We identified 244 admissions with recurrent ischemic stroke (5.7%) and 98 with major bleedings (2.3%). Table 2 shows the crude and adjusted propensity score–matched hazard ratios for the outcomes. Patients treated with intravenous tPA had a lower risk of long-term mortality with an adjusted propensity score–matched hazard ratio of 0.66 (95% confidence interval, 0.49–0.88; Figure). The long-term risk of recurrent ischemic stroke or major bleeding did not differ significantly between the intravenous tPA-treated and nonintravenous tPA-treated patients (adjusted hazard ratio, 1.05; 95% confidence interval, 0.68–1.64 and 0.59; 95% confidence interval, 0.24–1.47, respectively).
Only 2 randomized controlled stroke trials have examined long-term mortality after intravenous tPA.8,9 The National Institute of Neurological Disorders and Stroke trial revealed no differences in mortality rates among treated and nontreated patients at 12-month follow-up.8 Likewise, the third International Stroke Trial (IST-3) recently reported no significant difference in mortality rates in the intravenous tPA-treated and the placebo group at 18 months of follow-up.9 However, the IST-3 randomized patients who did not meet the existing European approval criteria for intravenous tPA and the cohort is, therefore, most likely not representative of routine clinical practice. To date, there is only 1 small observational study examining long-term mortality after intravenous tPA treatment, but the lack of a comparison group has precluded direct assessment of the potential benefit of intravenous tPA on survival.5 Our study on long-term outcome after intravenous tPA is the first to include a propensity score–matched comparison group on a large number of patients with complete follow-up.
The exact mechanisms behind the lower risk of long-term mortality after intravenous tPA treatment remain to be clarified. However, patients with smaller final infarct sizes, and therefore often a better functional outcome, most likely have fewer poststroke complications, including pneumonia, hip fractures, and depressions, all of which can be associated with increased mortality.
We cannot rule out that our results remain influenced by confounding factors because of the observational nature of our study design. However, the selection of the eligible, but nonintravenous tPA-treated patients with ischemic stroke, was designed to minimize the risk of confounding using a combination of restriction, propensity score matching, and multivariable regression.
Intravenous tPA treatment in routine clinical practice was associated with improved long-term survival that emphasizes the need of continued efforts to raise stroke awareness and to ensure effective organization and availability of acute stroke treatment.
Sources of Funding
The study was funded by The Tryg Foundation, Denmark.
Dr Ellemann has received a grant from the Tryg Foundation. Dr Madsen is a member of the Boehringer-Ingelheim, advisory board, Denmark. Dr Johnsen has received a speaker’s fee from Boehringer-Ingelheim. The other authors report no conflicts.
The online-only Data Supplement is available with this article at http://stroke.ahajournals.org/lookup/suppl/doi:10.1161/STROKEAHA.114.006570/-/DC1.
- Received July 28, 2014.
- Revision received August 11, 2014.
- Accepted August 12, 2014.
- © 2014 American Heart Association, Inc.
- Dirks M,
- Niessen LW,
- van Wijngaarden J,
- Koudstaal PJ,
- Franke CL,
- van Oostenbrugge RJ,
- et al
- Schmülling S,
- Grond M,
- Rudolf J,
- Heiss WD
- Fischer U,
- Mono ML,
- Zwahlen M,
- Nedeltchev K,
- Arnold M,
- Galimanis A,
- et al
- Kwiatkowski TG,
- Libman RB,
- Frankel M,
- Tilley BC,
- Morgenstern LB,
- Lu M,
- et al
- 11.↵Dansk Selskab for Apopleksi, DSfA. Referenceprogram for Behandling af Patienter med Apopleksi. Version 3.0 ed. København: Dansk Selskab for apopleksi, DSfA; 2009.