Utilization of Intravenous Thrombolysis in 3–4.5 Hours: Analysis of the Minnesota Stroke RegistryTekle W.G.a · Chaudhry S.A.a · Hassan A.E.a · Peacock J.M.b · Lakshminarayan K.a, b · Tsai A.b · Luepker R.a, b · Anderson D.C.a · Qureshi A.I.a
aZeenat Qureshi Stroke Research Center, University of Minnesota Medical Center, Minneapolis, Minn., and bMinnesota Department of Health, St. Paul, Minn., USA Corresponding Author
Background: The American Heart Association and the American Stroke Association recommend intravenous (IV) thrombolysis up to 4.5 h from acute ischemic stroke symptom onset based on its proven benefit in improving patient outcomes. We analyzed data from the Minnesota Stroke Registry to assess the rates of IV thrombolytic utilization and the process of care in this expanded window. Methods: We identified patients who had received IV recombinant tissue plasminogen activator (rt-PA) at any of the 19 participating hospitals from January 1, 2008 till December 31, 2010. Treatment groups were those actually treated by IV rt-PA in 0–3 h and those treated by IV rt-PA in 3–4.5 h. Duration from symptom onset to arrival in the emergency department (ED) was dichotomized to cohorts of 0–2 and 2–3.5 h. We determined the overall utilization of IV rt-PA in the expanded window and calculated door-to-needle times for the two treatment windows. We also ascertained the rates of symptomatic intracerebral hemorrhage between the two treatment groups. Results: Out of the total 519 patients who received IV rt-PA for acute ischemic stroke, 433 (83%) were treated within 0–3 h and 86 (17%) within 3–4.5 h. Of all the patients who received IV rt-PA within 3–4.5 h, 45% arrived at the ED within 2 h of symptom onset. Median door-to-needle time for the 0- to 3-hour window was 74.5 min [interquartile range (IQR) 57–90] and 54 min (IQR 43.5–70.5) for the 3- to 4.5-hour window. Based on arrival time to the ED, door-to-needle time of ≤60 min was achieved by only 31% (142/458) of patients who arrived within 0–2 h of their symptom onset compared to 61% (37/61) of those who arrived at the ED within 2–3.5 h of their symptom onset. Fifty-nine (14%) patients in the 0- to 3-hour group and 17 (20%) patients in the 3- to 4.5-hour group received a combination of IV rt-PA and endovascular treatments. Among patients with documented admission National Institutes of Health Stroke Scale scores, the values (median with IQR) were different between the 0- to 3- and the 3- to 4.5-hour group, i.e. 10 (IQR 5–18) and 7 (IQR 4–14), respectively. Conclusion: Patients who received IV rt-PA within the 3- to 4.5-hour window comprised 17% of all IV rt-PA cases treated in the Minnesota Stroke Registry hospitals after the new guidelines recommended a time window expansion. Almost half of these patients would have qualified for treatment within the 0- to 3-hour window as they presented within 0–2 h of symptom onset. Patients arriving 2–3.5 h after symptom onset received thrombolysis on average 20 min faster than patients arriving within 2 h of symptom onset.
© 2012 S. Karger AG, Basel
Intravenous (IV) recombinant tissue plasminogen activator (rt-PA) has been approved by the Food and Drug Administration as a thrombolytic agent for the treatment of acute ischemic stroke (AIS) since 1996. Over the past two decades, investigators have focused on expanding the time window for IV rt-PA administration in order to increase its utilization. A pooled analysis  of ECASS 1 and 2 [2,3], the NINDS rt-PA trial  and the ATLANTIS trials [5,6] suggested that IV rt-PA could be beneficial if administered up to 4.5 h from symptom onset. Previous studies [3,5,6] were unable to demonstrate a benefit of IV rt-PA when administered between 3 and 6 h of symptom onset, suggesting that treatment of patients after 4.5 h could obscure the treatment benefit. A meta-analysis that focused on IV rt-PA treatment between 3 and 6 h after symptom onset did not find an increase in the rate of a favorable outcome in the rt-PA-allocated group, compared with the placebo group .
Subsequently, ECASS 3  randomized 821 patients eligible for treatment 3–4.5 h after symptom onset to IV rt-PA or to placebo treatment. At 3 months, patients who were treated with IV rt-PA had a higher rate of functional independence as measured by a modified Rankin Scale (mRS) score of 0–1, with an absolute difference of 7.2% favoring rt-PA compared with placebo. In addition, the Safe Implementation of Thrombolysis in Stroke International Stroke Thrombolysis Registry (SITS-ISTR)  compared the outcomes of 11,865 patients who received IV rt-PA within 3 h of symptom onset with 664 patients treated within 3–4.5 h. The rates of symptomatic intracerebral hemorrhage (sICH; 1.6 vs. 2.2%) and functional independence (mRS 0–2) at 3 months (56.3 vs. 58.0%) were similar between the 0- to 3- and the 3- to 4.5-hour group, respectively.
In May 2009, the American Heart Association and the American Stroke Association issued a science advisory that recommended the use of IV rt-PA up to 4.5 h . The eligibility criteria for selecting patients within 3– 4.5 h was more stringent, requiring additional clinical criteria which excluded patients who were on any anticoagulation treatment, or had a combination of previous stroke and diabetes mellitus, or had severe strokes assessed clinically [e.g., by the National Institutes of Health Stroke Scale (NIHSS) score >25] or by imaging (early ischemic changes involving more than one third of the middle cerebral artery territory).
The implementation of IV rt-PA administration within the expanded time window has been studied in the Canadian Alteplase for Stroke Effectiveness Study  and in a follow-up analysis of the SITS-ISTR study  in Europe. Both studies demonstrated that the rates of sICH and favorable outcome are similar between patients treated within 3–4.5 h and those treated within 3 h.
However, large-scale studies on utilization of IV rt-PA have not been reported from the United States. Implementation of the expanded time window for IV rt-PA administration in the United States faces two unique challenges. First, an increasing rate of ‘drip-and-ship’ , whereby patients are given IV rt-PA bolus in emergency rooms en route to destination hospitals, provides an alternative to using a delayed time window for many cases. Second, endovascular treatment is provided in many major centers in the United States for patients presenting after 3 h of symptom onset , and all patients in the expanded IV time window do not represent previously untreated patients.
Based on these considerations, we analyzed data collected from the population-based Minnesota Stroke Registry (MSR) program to assess the utilization and the process of care with window expansion and, in real-world settings, the concurrent existence of ‘drip-and-ship’ and endovascular treatment.
The MSR is part of the Centers for Disease Control and Prevention Paul Coverdell National Acute Stroke Registry program . A total of 19 hospitals in the state of Minnesota provided standardized data for all admitted stroke patients during this study period. We identified patients who had received IV rt-PA at any of the 19 participating hospitals from January 1, 2008 till December 31, 2010. Patients who were first treated with IV rt-PA and later received a follow-up endovascular treatment (bridging therapy) were also included as IV treatment cases. A detailed description of methodology and data elements of the Paul Coverdell National Acute Stroke Registry has been provided in previous publications  and may also be found on the MSR website .
Briefly, data are collected for all patients aged ≥18 years on the date of admission, with documented final clinical diagnosis of stroke on hospital discharge, including subarachnoid hemorrhagic stroke, intracerebral hemorrhagic stroke, ischemic stroke, transient ischemic attack (TIA), stroke (not otherwise specified), and/or patients whose hospitalizations are assigned a principal International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) discharge diagnosis code 430-438, 671.5X, 674.0X, 997.02 .
Participating hospitals in the MSR are expected to utilize two methods to identify cases for inclusion and abstraction: first, by reviewing case logs from the emergency department (ED), hospital admissions and neurology consults, and second, by monitoring reports from medical or billing department records of cases with ICD-9-CM principal diagnosis discharge codes for stroke or TIA.
Two treatment groups of patients were identified based on the timing of IV rt-PA treatment from symptom onset. These treatment groups included patients treated with IV rt-PA within 3 h after symptom onset and those who received IV rt-PA within 3–4.5 h. Duration from symptom onset to arrival in the ED was dichotomized to cohorts of 0–2 and 2–3.5 h for both treatment groups based on the 60-min door-to-needle quality measure by the Joint Commission .
Clinical and demographic characteristics were determined. These included age, sex, time interval between symptom onset and treatment and door-to-needle times for the two treatment groups, admission ambulation status, pre-existing hypertension, diabetes mellitus, cigarette smoking, coronary artery disease, stroke or TIA, hyperlipidemia, congestive heart failure, atrial fibrillation, as well as the use of antithrombotic, cholesterol-reducing and antihypertensive agents before admission. To assess the severity of disease at presentation, we used ‘adverse initial examination findings’, which are the presence or absence of depressed level of consciousness, aphasia and weakness. The ‘adverse initial examination findings’ are consistently documented for almost all patients in the MSR database starting from March 1, 2009. The admission NIHSS score was also used to assess severity at presentation, but this variable was not documented in approximately 30% of the patients in the database. Ambulation status at discharge, rates of sICH (defined by a CT within 36 h that shows ICH and physician’s notes indicating clinical deterioration due to hemorrhage), in-hospital mortality, discharge destination, length of hospital stay, and other in-hospital complications, including pneumonia and urinary tract infection, were ascertained and analyzed.
We excluded patients who were treated before arriving at the MSR hospital using the drip-and-ship paradigm, because there was inadequate information available on last-known well time, arrival time at the original ED and on IV rt-PA administration time. In addition, patients who received IV rt-PA in the ED of a hospital participating in the MSR and were then transferred to another hospital for further care were excluded from the study.
All data are descriptively presented using the mean ± standard deviation or the median (interquartile range, IQR) for continuous data and frequencies for categorical data. The analysis was conducted using SAS software (version 9.2; SAS Institute, Inc., Cary, N.C., USA). Differences between potential confounders across both treatment groups were assessed by the χ2 test for categorical variables and by the F test (means) and Kruskal-Wallis test (medians) for continuous variables.
We identified a total of 519 patients who met the inclusion criteria for our analysis and had complete data. Of these 519 IV rt-PA-treated patients, 433 (83%) were treated within 0–3 h and 86 (17%) within the 3- to 4.5-hour interval after symptom onset. Of all the patients who received IV rt-PA within 3–4.5 h, 45% arrived to the ED within 2 h of symptom onset. Median door-to-needle time for the 0- to 3-hour window was 74.5 min (IQR 57–90) and 54 min (IQR 43.5–70.5) for the 3- to 4.5-hour window. Among all patients who received IV rt-PA, 31% of patients who arrived within 0–2 h and 61% of patients who arrived within 2–3.5 h received IV rt-PA within 60 min of arrival to the ED. Follow-up endovascular treatment was performed in 20 and 14% of IV rt-PA-treated patients in the 3- to 4.5- and 0- to 3-hour categories, respectively. Baseline demographic and clinical characteristics for each group are presented in table 1. Figure 1 demonstrates the overall trend in adopting the American Heart Association guideline among hospitals in Minnesota.
|Table 1. Baseline demographic and clinical characteristics of patients in the MSR treated with IV rt-PA|
|Fig. 1. Patients (n) treated with IV rt-PA in the 0- to 3- and 3- to 4.5-hour windows by quarter (Q).|
The rates of the presence of a reduced level of consciousness and aphasia were similar for both treatment groups; however, motor weakness was less common in the 3- to 4.5-hour group compared with the 0- to 3-hour group (70 vs. 83%; p = 0.01). Among patients with available admission NIHSS scores, the values were different between the 0- to 3- and 3- to 4.5-hour groups, i.e. median 10 (IQR 5–18) and 7 (IQR 4–14), respectively (p = 0.03). The proportion of patients with hypertension was higher in the 3- to 4.5-hour treatment group compared to the 0- to 3-hour group (78 vs. 67%; p = 0.04).
Thirty-seven percent of patients in the 0- to 3-hour and 31% in the 3- to 4.5-hour window were able to ambulate independently with or without device at discharge. In-hospital mortality was observed in 8% of patients treated in the 0- to 3-hour group and in 7% of the 3- to 4.5-hour group. Thirty-three and 30% of patients treated in the 0- to 3- and 3- to 4.5-hour window, respectively, were discharged home. The length of hospitalization was 5–6 days for both treatment groups.
The rates of sICH were 4.6 and 2.3% for the 0- to 3- and the 3- to 4.5-hour group, respectively.
During the period of this observational study, the 19 hospitals received 41 additional patients who were treated using the drip-and-ship paradigm. They have been excluded from the analysis for reasons already stated. They represent 8% (41/519) of all IV rt-PA cases hospitalized in the MSR hospitals.
Of all AIS patients treated with IV rt-PA in the MSR, 17% were treated within 3–4.5 h after symptom onset. Our results are comparable to a single-center study that we recently conducted  and in which we found a 14% rate of rt-PA utilization within 3–4.5 h. The Canadian Alteplase for Stroke Effectiveness Study  and the SITS-ISTR  study found that 11.6 and 10% of all IV rt-PA-treated patients were treated within the 3- to 4.5-hour window, respectively. We avoided direct comparison regarding outcome between patients treated within 3 h and those treated within 3–4.5 h after symptom onset because of our inability to adjust for prominent confounders such as the NIHSS score and etiological stroke subtype. The two groups are different at baseline.
The proportion of patients who received IV rt-PA within 3–4.5 h being much lower than the proportion of those who received thrombolysis within 3 h is probably secondary to a lower overall number of stroke patients who present within 3–4.5 h compared to those who present within 3 h. The Buffalo Metropolitan Area and Erie County Stroke Study  found that 11% of all AIS patients presented to the ED within 3–6 h of symptom onset. Another study from Shanghai, China , reported that 21% of all AIS patients present in this time frame. Schwamm and Fonarow  using the Get With The Guidelines-Stroke database reported that 13% of AIS patients presented to participating hospitals within 2–3.5 h of symptom onset.
The finding that 45% of the patients who received IV rt-PA within 3–4.5 h actually arrived at the ED within 2 h of symptom onset suggests that these patients could potentially have been treated with IV rt-PA within 3 h if the <60 min door-to-needle time recommendation of the Joint Commission was met. In fact, only about 31% of patients who arrived within 0–2 h received IV rt-PA within 60 min of arrival to the ED, compared with 61% of those who arrived within 2–3.5 h. This seemingly systemic delay in initiation of IV rt-PA in ‘early arrivals’ compared with ‘late arrivals’ is also reflected by the finding that patients arriving 2–3.5 h after symptom onset received IV rt-PA 20 min faster on average than patients arriving within 2 h of symptom onset.
An important question yet to be answered is the following: ‘could the availability of the 3- to 4.5-hour window potentially reduce the sense of urgency in AIS treatment among practitioners?’ We are unable to answer this question from our analysis because all the patients who arrived within 0–2 h, despite experiencing a delay in treatment, were captured in the 3- to 4.5-hour treatment window data. However, patients experiencing a delay among the 2- to 3.5-hour arrival cohort were unlikely to receive IV rt-PA and would remain unidentified in our study sample, thereby biasing the calculation of door- to-needle time. The anticipated consequence is that the mean door-to-needle time would appear artifactually prolonged in the cohort treated within 0–3 h. Nevertheless, it is evident that many more patients could be treated earlier with IV thrombolytics if the Joint Commission guidelines for the door-to-needle time were followed.
It is possible that sICH may have been underreported in a retrospective database such as ours, as CT scans and serial examinations may not have been performed consistently and analyzed for sICH. It is routine practice in most hospitals to perform a CT scan for any neurological deterioration detected by the treating team. Therefore, only patients with a smaller magnitude of deterioration are likely to be missed during ascertainment.
As a retrospective review, our study has limitations due to incomplete data, e.g., unknown admission and discharge ambulation status in approximately 10% of patients, undetermined admission NIHSS scores in 30% of patients, and the lack of standardized outcome scale measures such as the mRS at 90 days. However, the estimates of IV rt-PA treatment in 0–3 and 3–4.5 h are reflective of the general practice derived from hospitals that treat the majority of patients with ischemic stroke but are not well represented in prospective controlled studies.
In conclusion, patients who received IV rt-PA within the 3- to 4.5-hour time window comprised 17% of all IV rt-PA cases treated in the MSR hospitals after new guidelines recommended a time window expansion. The observation that nearly half of the patients who presented within 0–2 h of symptom onset received IV rt-PA in the expanded window raises the question if availability of an expanded window might have reduced the urgency of the response to acute stroke treatment. Therefore, continuous efforts are required to ensure that patients who present within 0–2 h receive IV rt-PA within 3 h of symptom onset to avoid delay-related loss of treatment benefit.
Dr. Qureshi has received funding from the National Institutes of Health RO1-NS44976-01A2 (medication provided by ESP Pharma), American Heart Association Established Investigator Award 0840053N, National Institute of Health U01-NS062091-01A2, and the Minnesota Medical Foundation, Minneapolis, Minn., USA.
Wondwossen G. Tekle, MD
Zeenat Qureshi Stroke Research Center
Department of Neurology, University of Minnesota
516 Delaware St. SE, Minneapolis, MN 55455 (USA)
Received: November 18, 2011
Accepted: September 14, 2012
Published online: December 4, 2012
Number of Print Pages : 6
Number of Figures : 1, Number of Tables : 1, Number of References : 21
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