Clinical Review & Education
Special Communication
Implementing a Mobile Stroke Unit Program in the United States Why, How, and How Much? Suja Rajan, PhD; Sarah Baraniuk, PhD; Stephanie Parker, RN, BSN; Tzu-Ching Wu, MD; Ritvij Bowry, MD; James C. Grotta, MD
IMPORTANCE There are many ways a mobile stroke unit (MSU) might prove valuable for
patients with ischemic and hemorrhagic stroke, such as earlier recognition, more accurate triage, improved management of blood pressure and other critical physiological variables, and eventually earlier implementation of effective therapies. The MSU may be particularly valuable for treatment of patients with acute ischemic stroke with tissue plasminogen activator (tPA) within 4.5 hours of symptom onset, the most evidence-based effective emergency treatment for the most prevalent stroke diagnosis. OBJECTIVES To review existing data on prehospital stroke treatment, especially relevant to MSU technology, to identify gaps in our understanding of MSU feasibility, especially relevant to applying the MSU strategy in the United States, and to describe the Houston MSU program and clinical trial. EVIDENCE REVIEW Published data from English-language journals in PubMed from 1995 to present reviewing early treatment with tPA and prehospital stroke evaluation and treatment. FINDINGS The MSU may result in an overall shift toward earlier evaluation and treatment with tPA, particularly into the first hour after symptom onset, leading to substantially better outcomes. As a result of improved clinical outcomes owing to earlier treatment, the costs of an MSU program may be offset by a reduction in the costs of long-term stroke care and an increase in quality-adjusted life-years, thereby supporting more widespread use of this technology. To make MSU deployment more practical, the vascular neurologist aboard the MSU must be replaced by a remote vascular neurologist connected to the MSU by telemedicine, reducing manpower requirements and costs. CONCLUSIONS AND RELEVANCE The MSU strategy could dramatically transform the way acute stroke is managed in the United States. A prospective study evaluating the logistics, outcomes, and cost-effectiveness of this approach is needed and under way. JAMA Neurol. 2015;72(2):229-234. doi:10.1001/jamaneurol.2014.3618 Published online December 8, 2014.
I
ntravenous tissue plasminogen activator (tPA) remains the only level 1A treatment for acute ischemic stroke based on several prospective randomized trials comparing tPA with standard management (SM).1 The results of these trials and of pooled analyses confirm the relationship of treatment success with time from symptom onset to initiation of treatment.2-7 However, despite 2 decades of efforts to streamline systems of care, including formation of designated stroke centers, placement of computed tomographic (CT) scanners in the emergency department (ED), dedicated 24/7 in-house stroke teams, and ED pathways to speed treatment, most patients are treated beyond 2 hours when tPA is most effective.8 While there are many reasons for this delay, inherent delay is caused by ED triage, registration, evaluation, testing, and treatment. The median door-to-needle times in stroke center EDs in the jamaneurology.com
Author Affiliations: The University of Texas School of Public Health, Houston (Rajan, Baraniuk); Department of Neurology, The University of Texas Health Science Center at Houston, Houston (Parker, Wu); Memorial Hermann–Texas Medical Center, Houston (Bowry, Grotta). Corresponding Author: James C. Grotta, MD, Memorial Hermann–Texas Medical Center, Clinical Innovation and Research Institute, 6410 Fannin St, Ste 1423, Houston, TX 77030 (james [email protected]).
United States approximate 60 minutes.9 Such delay not only likely results in fewer patients completely recovering but also reduces the total number of patients who can be treated within the 4.5-hour maximum window of tPA effectiveness, contributing to the overall low national treatment rate estimated to be about 5% of all acute ischemic strokes. Recently, the long-awaited concept of faster prehospital treatment with tPA has become a reality following the demonstration by Walter et al10 and Ebinger et al11 in Germany that CT scanners can be mounted on an ambulance to take treatment to the patient. Walter and colleagues showed that by using this MSU concept, treatment with tPA can be carried out safely and accurately with a median time between onset and treatment of 70 to 80 minutes. Ebinger and colleagues showed that an MSU resulted in increased treat(Reprinted) JAMA Neurology February 2015 Volume 72, Number 2
Copyright 2015 American Medical Association. All rights reserved.
Downloaded From: http://archneur.jamanetwork.com/ by a CAMBRIDGE UNIVERSITY LIBRARY User on 05/14/2015
229
Clinical Review & Education Special Communication
ment rates (from 21% to 33%), 25-minute-shorter time to treatment, and increased proportion of patients treated within 90 minutes of symptom onset (58% with an MSU vs 37% without). This success has the potential to result in substantially improved outcomes for patients with acute ischemic stroke and dramatically alter the way patients with acute stroke are managed. Achieving faster treatment is the primary mechanism by which MSU management might result in better outcomes from tPA treatment. Better outcomes should result from an overall shift toward earlier evaluation and treatment with tPA, particularly into the first hour after symptom onset. Another result of faster treatment via the MSU is that there might be an increase in the absolute number of tPAtreated patients. It is logical that if all patients are evaluated sooner using the MSU, more will be treated within the 4.5-hour window that is currently recommended by published guidelines.1 If use of the MSU can increase the total number of patients with stroke being evaluated and treated within this window, it should increase the total number of patients recovering from their stroke. The MSU strategy could dramatically transform the way acute stroke is managed in the United States, but there are critically important gaps not addressed by pilot studies already completed in Germany. These include the ability to adapt the MSU to the logistics of the more complex US health care system, reliability of telemedicine (TM) on the MSU, clinical outcomes, and costs. These must be addressed by a trial executed in the United States.
Logistics To our knowledge, there is no experience using the MSU in the United States; therefore, we do not know how much time can be saved by their use in the United States, where traffic patterns, distances, market forces, and local regulations differ from Europe. Certainly, the impact of any change in a stroke system of care such as MSU deployment will be location specific and will differ between urban and rural areas. For instance, in Houston, Texas, where we have a longestablished coordinated network of stroke centers and emergency medical services (EMS) providers as well as a central database of all stroke centers, current stroke management is excellent. In 2013, 300 patients (22% of patients with ischemic stroke transported) arrived at one of the 3 comprehensive stroke centers (CSCs) in the city within 2 hours of symptom onset and 199 (14% of the total number of patients with ischemic stroke and 90% of tPA-eligible patients) were treated with tPA within 3 hours of symptom onset. At Memorial Hermann–Texas Medical Center between January and August 2013 (8 months), 74 patients were delivered by EMS and treated with intravenous tPA within 4.5 hours of symptom onset, with a median time between onset and treatment of 123.5 minutes. It remains to be seen how much these statistics could be improved by an MSU strategy grafted onto the existing EMS. There are several potential complexities with earlier treatment on the MSU that also need to be addressed. One is the increased chance of treating patients with stroke mimics, eg, other pathologies such as migraine or seizures, or patients with transient ischemic attacks, eg, those who would recover within 24 hours even without treatment (this does not include strokes with negative findings on magnetic resonance imaging or “aborted” strokes that are reversed by tPA). Regarding stroke mimics, Ebinger et al11 reported 230
Mobile Stroke Unit Program
the rate of stroke mimic treatment with the MSU to be 2% and no different from that with SM. These results were dependent on having a physician experienced in stroke treatment aboard the MSU, and such expertise in person or via TM will be necessary in the United States as well. An important potential advantage of the MSU concept that extends beyond treating more patients more quickly may be earlier, accurate identification of stroke mimics and appropriate prioritization of patients with true stroke. The transient ischemic attack issue may be more difficult to assess. The incidences of patients completely recovering within 24 hours in the placebo arm of the National Institute of Neurological Disorders and Stroke (NINDS) study were 2.4% in patients treated 91 to 180 minutes after symptom onset and 2.1% in patients treated within 0 to 90 minutes.2 Because of the temporal distribution of randomized patients in the NINDS study (as will be described later), this does not suggest a dramatic increase with earlier treatment between 80 and 180 minutes. However, this incidence could be higher in patients evaluated by the MSU within the first 60 minutes after onset. To solve this dilemma, we need to observe patients in the prehospital stage who we determine would meet criteria for treatment but for whom we do not activate the MSU and then to determine what proportion of them completely recover by the time of tPA decision in the ED. This would allow us to approximate the percentage of patients with transient ischemic attack in the cohort of tPAtreated patients with MSU dispatch. A third potential problem might be an increase in the number of intracerebral hemorrhages, angioedema, or other complications of tPA treatment with MSU management. German studies have not reported an increase in these complications with MSU management,10,11 and assuming stroke expertise aboard the MSU or via TM, we would expect the same results in the United States. Another question is the impact of MSU deployment on intraarterial treatment (IAT). Recent trials12-14 failed to show a benefit for IAT compared with tPA or as an adjunctive approach to tPA. However, post hoc analysis from the Interventional Management of Stroke III study13 showed that patients who achieve recanalization within 4 to 5 hours from symptom onset have potential to benefit most from IAT, so if and when IAT becomes widely used, reducing time to treatment will be important. Deployment of the MSU might potently reduce time to IAT treatment by allowing prehospital identification of patients with probable large artery occlusion, facilitating their inhospital treatment by prehospital notification, earlier assembly of the endovascular team and angiography suite preparation, and shorter in-house delays incurred by acquiring imaging and laboratory data and treating with tPA, perhaps allowing bypass of ED evaluation altogether.
Telemedicine Eventually, the widespread use of MSUs will depend on adequate manpower to guide treatment. Our preliminary experience and the data from Germany suggest that the ratio of MSU alerts from EMS dispatch to tPA treatments is at least 10 to 1, making it impractical to have a vascular neurologist (VN) aboard the MSU for all calls. However, the decision whether to give tPA based on clinical criteria requires training, experience, and careful judgment. Experience with TM, including at our own center, has shown it to
JAMA Neurology February 2015 Volume 72, Number 2 (Reprinted)
Copyright 2015 American Medical Association. All rights reserved.
Downloaded From: http://archneur.jamanetwork.com/ by a CAMBRIDGE UNIVERSITY LIBRARY User on 05/14/2015
jamaneurology.com
Mobile Stroke Unit Program
Special Communication Clinical Review & Education
be a feasible, accurate, and safe way to extend this expertise to hospitals where it is lacking and to guide and increase tPA treatment in such locations.15,16 We recently demonstrated the feasibility and accuracy of TM assessment of actors simulating patients with stroke in ambulances using existing technology. In the Pre-hospital Utility of Rapid Stroke Evaluation Using In-Ambulance Telemedicine (PURSUIT) study,15 trained actors portrayed 10 unique stroke scenarios, each conducted 4 times, and were retrieved and transported by Houston Fire Department EMS to our stroke center. A remote VN performed remote assessments in real time and obtained clinical data points and National Institutes of Health Stroke Scale scores using the RP-Xpress device (InTouch Health). In 34 of the 40 scenarios (85%), the teleconsultation was conducted without major technical complication. The absolute agreement for intraclass correlation was 0.997 (95% CI, 0.992-0.999) for the National Institutes of Health Stroke Scale scores obtained during the real-time sessions. Matching of real-time assessments occurred for 88% of the National Institutes of Health Stroke Scale scores (30 of 34 scenarios) within 2 points and 96% of the clinical information. However, TM has not been tested for treating actual patients with stroke with tPA in the prehospital environment. This will likely be a more complex exercise given the need for multiple parallel tasks (neurological assessment; vital signs; patient positioning, attention to comfort, and possible restraint; CT scanning; point-ofcare laboratory measurement; drug mixing and administration) being carried out urgently by several people within the confined space of an MSU.
Clinical Outcomes Faster treatment using the MSU will allow us, for the first time, to answer an important scientific question by making treatment possible within the first 60 to 80 minutes after symptom onset when tPA is likely to have its greatest effect. Preliminary data from Ebinger et al17 showed that 31% of patients treated with tPA using their MSU were treated within 60 minutes of onset compared with 4.9% with SM. These rates, although based on small numbers, compare very favorably with treatment rates within the first 60 minutes from previous trials and current databases. Among the 302 patients treated with tPA vs placebo within 90 minutes of onset in the NINDS Stroke Study, only 2 were randomized within 60 minutes of onset (both were randomized to the placebo group) and only 41 were randomized between 61 and 80 minutes after onset.2 The latest pooled data from all randomized tPA trials added only 2 additional patients randomized within the first 60 minutes. Only 1.6% of patients were treated within 60 minutes of onset in the Safe Implementation of Treatments in Stroke–International Stroke Thrombolysis Registry (SITS-ISTR).18 In Helsinki, Finland, using a streamlined system of stroke triage, 10% of patients were treated within 70 minutes of symptom onset,19 and of 58 353 patients treated with tPA in the Get With the Guidelines–Stroke program, fewer than 1000 were treated within 60 minutes of symptom onset.8 There is very little information on how much improvement in clinical outcomes will occur with treatment within the first 60 to 80 minutes. While data from multiple studies and pooled analyses conjamaneurology.com
firm a strong inverse relationship between time to treatment and recovery, the slope and shape of that relationship within the first 90 minutes after stroke symptom onset are very uncertain as reflected in the wide confidence intervals surrounding outcomes in various pooled analyses and the total absence of data before 60 minutes. It is possible that, owing to collateral flow, human penumbral tissue can survive long enough so that there would be little advantage to earlier treatment within the 90-minute epoch. However, data presented by Ebinger et al17 demonstrated that patients treated within the first 60 minutes of onset by their MSU had an odds ratio of 1.93 (95% CI, 1.09-3.41) of discharge to home compared with later treatment. These data have not yet been subjected to peer review to be certain that confounding variables were considered in the comparison, and longer-term outcomes were not reported. Deployment of the MSU in the United States will allow us to determine the benefit of tPA on long-term outcome in prospectively studied patients treated within 0 to 60 or 60 to 80 minutes vs later treatment. Following the precedent of recent tPA pooling projects,3,4 US data can be pooled with European data to increase our power to arrive at an accurate answer. Eventually, to completely assess the value of the MSU strategy, we need to answer the larger question of whether all patients eligible for tPA in the prehospital setting have better outcomes if managed via an MSU or SM. This requires an intention-to-treat study that will include all eligible patients (whether or not they eventually receive tPA) evaluated at the same time in the prehospital setting and by the same personnel and will compare patients assigned to MSU vs SM. Assessing and enrolling all patients while they are still in the prehospital phase will ensure that patients receiving SM who are enrolled are comparable to the patients managed with the MSU and will also allow us to account for those patients who would have met criteria for prehospital treatment but could not be treated within the 4.5-hour window because they were managed on the SM pathway.
Cost-Effectiveness Stroke is among the top 15 most expensive conditions treated in US hospitals and among the top 10 most expensive conditions billed to Medicare.20 Medicare bears the highest cost burden of the disease; almost 60% of stroke-related hospital costs and more than 60% of overall stroke-related costs are borne by Medicare.20,21 Non– nursing home stroke care constitutes more than 10% of Medicare’s budget.22 As the US population ages, the incidence and prevalence of this disease will increase, and hence costs associated with stroke and the cost burden of Medicare will substantially increase. It is projected that by 2030, 4% of the American population will have had a stroke and the total medical cost of stroke will be nearly $200 billion (in 2010 US dollars), which is a 250% increase as compared with the medical costs as of 2012.23 Ischemic stroke accounts for 87% of all stroke events. The use of tPA in ischemic stroke, and specifically the early use of tPA, has been shown to be both clinically efficacious and cost-effective. Fagan et al24 demonstrated that the use of tPA reduced hospital length of stay, with higher discharge to homes instead of inpatient rehabilitation or nursing homes. Their Markov analysis predicted an increase in quality-adjusted life-years (QALYs) with 94% probability (Reprinted) JAMA Neurology February 2015 Volume 72, Number 2
Copyright 2015 American Medical Association. All rights reserved.
Downloaded From: http://archneur.jamanetwork.com/ by a CAMBRIDGE UNIVERSITY LIBRARY User on 05/14/2015
231
Clinical Review & Education Special Communication
Mobile Stroke Unit Program
Figure 1. Mobile Stroke Unit A
B
Exterior (A) and interior (B) of the mobile stroke unit.
and a decrease in poststroke first-year costs with 93% probability among patients receiving tPA, clearly demonstrating the value of using tPA. Tan Tanny et al25 documented that tPA use within 4.5 hours of stroke occurrence had an incremental cost-effectiveness ratio of $1478/QALY (Australian) during the first year following the tPA treatment; tPA use improved QALYs but also marginally increased costs. Tung et al26 performed a lifetime cost-effectiveness analysis for the use of tPA and found an increase in both lifetime costs and QALYs with tPA administered within 4.5 hours of ischemic stroke, with an incremental cost-effectiveness ratio of $21 978/QALY. Treatments with an incremental cost-effectiveness ratio of less than $50 000/QALY are well accepted as cost-effective and generally reimbursed by health care decision makers. The literature provides ample evidence on the economic impact of stroke and cost-effectiveness of tPA. However, to our knowledge, there are no studies within the United States that specifically evaluate the economic impact of an MSU using TM to improve early tPA administration. The MSU strategy entails significant capital investment in outfitting an ambulance with a CT scanner and TM equipment as well as physician and nurse staffing changes and staff training to support MSU operations. These significant fixed costs imply that the agency operating the MSU and the staff supporting it should receive additional reimbursements to ensure the financial viability and sustainability of the MSU operation. The health care payers are more likely to provide additional reimbursement if we can demonstrate improvement in patient outcomes and QALYs and possible poststroke cost savings for the payers and also provide estimates of incremental fixed cost increases per patient due to the dispatch of an MSU. Regarding initial setup costs, purchasing, equipping, and deploying an ambulance with a CT scanner, point-of-care laboratory testing capabilities, TM capability, and other needed equipment in Houston has a cost of approximately $600 000. Formal cost analysis would have to balance these costs plus costs for staffing and maintenance against any reduction in the total hospital and longterm care costs to the health care system for each patient with an ischemic stroke treated on the MSU, estimated to average approximately $140 000 per patient in 1999 US dollars, undoubtedly much higher today. 232
Houston MSU To address these gaps in our knowledge, we have introduced at the Texas Medical Center in Houston the nation’s first MSU funded by donations from grateful patients, local philanthropists, and the Frazer ambulance company. We are not only the first center in the United States to put an MSU into operation but also the first (and only) group to use it for clinical research purposes. The MSU was constructed at the Frazer factory (Figure 1 and Figure 2; http://www.frazerbilt.com/Videos/watch.php?id=784 and https: //www.youtube.com/watch?v=y1m64EL-k5I&index=6&list =UU7MwkvzzoUJ1SOHHl-PvLBQ). Many steps were required for licensing, inspecting, equipping, insuring, and credentialing the MSU and integrating MSU operations into the Houston Fire Department EMS dispatch, firstresponder, and paramedic system, which services the various Houston stroke centers. A full description of these efforts, which required 6 to 12 months of dedicated effort, is beyond the scope of this review, but 2 important overriding concepts that led to our ability to be up and running within 1 year of starting the project were an understanding that our overall goal is to conduct clinical research rather than to assume that the MSU should be implemented prima facie and that this is to be a collaborative effort of community stroke stakeholders, including EMS and all 3 of the CSCs and stroke teams at the Texas Medical Center.
The Benefits of Stroke Treatment Delivered Using a Mobile Stroke Unit Compared to Standard Management by Emergency Medical Services Study We have begun the Benefits of Stroke Treatment Delivered Using a Mobile Stroke Unit Compared to Standard Management by Emergency Medical Services (BEST-MSU) Study with the specific aims to determine the following: (1) the logistics and clinical outcome of the MSU vs SM; (2) the agreement between a VN remotely assessing a patient with suspected stroke via TM and in-person assessment by a VN in the MSU; and (3) the incremental cost and effectiveness as-
JAMA Neurology February 2015 Volume 72, Number 2 (Reprinted)
Copyright 2015 American Medical Association. All rights reserved.
Downloaded From: http://archneur.jamanetwork.com/ by a CAMBRIDGE UNIVERSITY LIBRARY User on 05/14/2015
jamaneurology.com
Mobile Stroke Unit Program
Special Communication Clinical Review & Education
Figure 2. Treating the First Patient A
B
Treating our first patient 78 minutes after symptom onset (A) with simultaneous telemedicine backup (B).
sociated with MSU vs SM using standard cost-effectiveness analyses. This is a prospective cohort study with block-randomized MSU or SM deployment weeks and blinded assessment of both trial entry and clinical outcomes. After we are notified of a patient with possible stroke by EMS, the patient is included in the study if he or she was last seen healthy possibly within 4.5 hours of symptom onset, has a medical history and physical and/or neurological examination findings consistent with acute stroke, and meets no tPA exclusions per guidelines1 prior to CT scanning or baseline laboratory tests. Written informed consent is obtained from the patient (if competent) or legal representative to include the patient in the database, but it is not needed for treatment, which follows published guidelines. We intend to enroll 248 patients, of whom we expect 186 will receive tPA. The MSU operates in parallel with EMS routine. After receiving a 9-1-1 call, the EMS dispatcher immediately dispatches the nearest available emergency medical technician or paramedic team. Immediately after emergency medical technician or paramedic dispatch, the MSU team is activated via dedicated pager. The MSU is dispatched on only 50% of weeks according to a blocked randomization schedule. On non-MSU dispatch weeks (SM weeks), the MSU team is still dispatched to the scene but travels in a private vehicle. The MSU is staffed by a Houston Fire Department paramedic, certified CT technician, VN (VN faculty or fellow), and registered nurse experienced in both clinical research and acute stroke management. The MSU team alerts the on-call TM VN, who immediately connects from the central TM office at the University of Texas Medical School at Houston to the mobile TM device on the MSU. The MSU is stationed in the heart of the Texas Medical Center and surrounded by the 3 CSCs, which are the destination of all patients with stroke who are picked up by EMS within a predetermined 5-mile radius catchment area surrounding the Texas Medical Center. We have found that the 5-mile radius catchment area will allow dispatch and arrival of the MSU at the emergency site while EMS is still on scene evaluating the patient. This catchment area is expanding as MSU team efficiency improves. Once on scene, the patient’s medical history, vital signs, fingerstick glucose level, and physical examination findings are jointly evaluated by the EMS paramedics and MSU VN and nurse, and if the jamaneurology.com
patient has signs and symptoms of stroke possibly within 4.5 hours of onset meeting published guidelines for tPA, he or she is moved into the MSU, enrolled into the study for purposes of answering the specific aims, and assigned to the MSU arm. If the patient does not have signs and symptoms of a stroke, is clearly outside the 4.5hour window, has other tPA exclusions, or is clinically unstable (such as requiring pressor or ventilator support), he or she is managed and transported per EMS routine. These latter patients will be considered screen failures. Once a patient is moved into the MSU, a venous blood sample is analyzed by a point-of-care laboratory (blood glucose level, hematocrit, platelet level, international normalized ratio if needed) by the MSU paramedic and nurse. Noncontrast CT scanning of the head is performed by the CT technician, and the results are uploaded onto the picture archiving and communication system for immediate visualization on the MSU laptop computer and read by the on-site MSU VN. Blood pressure is managed per guidelines. If the patient meets all published inclusion and exclusion criteria for thrombolysis, intravenous tPA is given without delay. Simultaneous to these events, the patient is evaluated by the TM VN using a portable RP-Xpress unit (InTouch Health) attached to the MSU gurney or handheld by the paramedic to optimize viewing. The MSU paramedic helps the TM VN record the essentials of the medical history, vital signs, physical examination findings, and laboratory results. The CT images on the MSU are securely and wirelessly sent via on-board 4G connection in real time to a secure picture archiving and communication system for review by the TM VN. The TM VN makes his or her own independent decision about whether the patient qualifies for tPA treatment. After the tPA bolus is given and infusion is started or the decision is made to withhold tPA, the patient is transported in the MSU with the research nurse and VN to the appropriate CSC. On SM weeks, the MSU is not dispatched but the MSU nurse or VN travels to the scene by car. Once on scene, the patient’s medical history, vital signs, finger-stick glucose level, and physical examination findings are jointly evaluated by the EMS paramedics and MSU VN or nurse. If the patient meets all inclusion criteria except laboratory and CT results, the patient is enrolled into the study and as(Reprinted) JAMA Neurology February 2015 Volume 72, Number 2
Copyright 2015 American Medical Association. All rights reserved.
Downloaded From: http://archneur.jamanetwork.com/ by a CAMBRIDGE UNIVERSITY LIBRARY User on 05/14/2015
233
Clinical Review & Education Special Communication
Mobile Stroke Unit Program
signed to the SM arm. The MSU research nurse or VN travels from the sceneandmeetsEMSandthepatientatthedesignatedCSCED.There, the hospital-based stroke team manages the patient as per CSC routine, the same standard-of-care analyses are carried out as with the MSU treatment, and intravenous tPA is given if the patient meets inclusion and exclusion criteria according to published guidelines. All enrolled patients are followed up for both clinical and economic outcome measures with clinical outcomes assessed by an investigator blinded to treatment group. To date, the study has been up and running for 1 month. We have been enrolling 1 patient approximately every 2 days and treating 2 ARTICLE INFORMATION Accepted for Publication: October 6, 2014. Published Online: December 8, 2014. doi:10.1001/jamaneurol.2014.3618. Author Contributions: Dr Grotta had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Rajan, Baraniuk, Parker, Bowry, Grotta. Acquisition, analysis, or interpretation of data: Wu, Grotta. Drafting of the manuscript: Rajan, Parker, Bowry, Grotta. Critical revision of the manuscript for important intellectual content: Baraniuk, Parker, Wu, Grotta. Statistical analysis: Baraniuk. Administrative, technical, or material support: Rajan, Parker, Wu, Bowry, Grotta. Study supervision: Parker, Grotta. Conflict of Interest Disclosures: Dr Grotta reported having consulting contracts with Frazer Ltd, Specialists on Call, Haemnonetics, and Lundbeck. No other disclosures were reported. Funding/Support: The Houston Mobile Stroke Unit project has received limited financial support from Genentech and Covidien. Role of the Funder/Sponsor: Genentech and Covidien had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication. REFERENCES
Conclusions The MSU strategy could dramatically transform the way acute stroke is managed in the United States. A prospective study evaluating the logistics, outcomes, and cost-effectiveness of this approach is needed and under way.
4. Lees KR, Bluhmki E, von Kummer R, et al; ECASS, ATLANTIS, NINDS and EPITHET rt-PA Study Group. Time to treatment with intravenous alteplase and outcome in stroke: an updated pooled analysis of ECASS, ATLANTIS, NINDS, and EPITHET trials. Lancet. 2010;375(9727):1695-1703. 5. Marler JR, Tilley BC, Lu M, et al. Early stroke treatment associated with better outcome: the NINDS rt-PA Stroke Study. Neurology. 2000;55(11): 1649-1655. 6. Lansberg MG, Schrooten M, Bluhmki E, Thijs VN, Saver JL. Treatment time-specific number needed to treat estimates for tissue plasminogen activator therapy in acute stroke based on shifts over the entire range of the modified Rankin Scale. Stroke. 2009;40(6):2079-2084. 7. Saver JL, Fonarow GC, Smith EE, et al. Time to treatment with intravenous tissue plasminogen activator and outcome from acute ischemic stroke. JAMA. 2013;309(23):2480-2488. 8. Fonarow GC, Smith EE, Saver JL, et al. Timeliness of tissue-type plasminogen activator therapy in acute ischemic stroke: patient characteristics, hospital factors, and outcomes associated with door-to-needle times within 60 minutes. Circulation. 2011;123(7):750-758. 9. Fonarow GC, Zhao X, Smith EE, et al. Door-to-needle times for tissue plasminogen activator administration and clinical outcomes in acute ischemic stroke before and after a quality improvement initiative. JAMA. 2014;311(16):1632-1640. 10. Walter S, Kostopoulos P, Haass A, et al. Diagnosis and treatment of patients with stroke in a mobile stroke unit versus in hospital: a randomised controlled trial. Lancet Neurol. 2012;11(5):397-404.
1. Jauch EC, Saver JL, Adams HP Jr, et al; American Heart Association Stroke Council; Council on Cardiovascular Nursing; Council on Peripheral Vascular Disease; Council on Clinical Cardiology. Guidelines for the early management of patients with acute ischemic stroke: a guideline for healthcare professionals from the American Heart Association/American Stroke Association. Stroke. 2013;44(3):870-947.
11. Ebinger M, Winter B, Wendt M, et al; STEMO Consortium. Effect of the use of ambulance-based thrombolysis on time to thrombolysis in acute ischemic stroke: a randomized clinical trial. JAMA. 2014;311(16):1622-1631.
2. National Institute of Neurological Disorders and Stroke rt-PA Stroke Study Group. Tissue plasminogen activator for acute ischemic stroke. N Engl J Med. 1995;333(24):1581-1587.
13. Broderick JP, Palesch YY, Demchuk AM, et al; Interventional Management of Stroke (IMS) III Investigators. Endovascular therapy after intravenous t-PA versus t-PA alone for stroke. N Engl J Med. 2013;368(10):893-903.
3. Hacke W, Donnan G, Fieschi C, et al; 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(9411):768-774.
234
patients per week with tPA. Of the 9 tPA-treated patients to date, 2 were treated within 60 minutes and 3 between 61 and 80 minutes from symptom onset.
12. Ciccone A, Valvassori L; SYNTHESIS Expansion Investigators. Endovascular treatment for acute ischemic stroke. N Engl J Med. 2013;368(25): 2433-2434.
14. Kidwell CS, Jahan R, Gornbein J, et al; MR RESCUE Investigators. A trial of imaging selection and endovascular treatment for ischemic stroke. N Engl J Med. 2013;368(10):914-923.
15. Wu TC, Nguyen C, Ankrom C, et al. Prehospital utility of rapid stroke evaluation using in-ambulance telemedicine: a pilot feasibility study. Stroke. 2014; 45(8):2342-2347. 16. Johansson T, Wild C. Telemedicine in acute stroke management: systematic review. Int J Technol Assess Health Care. 2010;26(2):149-155. 17. Ebinger M, Wendt M, Rozanski M, et al. Golden hour-thrombolysis by starting treatment before hospital arrival: the Pre-hospital Acute Neurological Treatment and Optimization of Medical Care in Stroke Study (PHANTOM-S) [abstract 104]. Stroke. 2014;45: A104. 18. Wahlgren N, Ahmed N, Dávalos A, et al; SITS Investigators. Thrombolysis with alteplase 3-4.5 h after acute ischaemic stroke (SITS-ISTR): an observational study. Lancet. 2008;372(9646): 1303-1309. 19. Meretoja A, Strbian D, Mustanoja S, Tatlisumak T, Lindsberg PJ, Kaste M. Reducing in-hospital delay to 20 minutes in stroke thrombolysis. Neurology. 2012;79(4):306-313. 20. Torio CM, Andrews RM. National Inpatient Hospital Costs: The Most Expensive Conditions by Payer, 2011. Rockville, MD: Agency for Health Care Policy & Research; 2013. HCUP statistical brief 160. 21. Trogdon JG, Finkelstein EA, Nwaise IA, Tangka FK, Orenstein D. The economic burden of chronic cardiovascular disease for major insurers. Health Promot Pract. 2007;8(3):234-242. 22. Cohen JW, Krauss NA. Spending and service use among people with the fifteen most costly medical conditions, 1997. Health Aff (Millwood). 2003;22(2):129-138. 23. Ovbiagele B, Goldstein LB, Higashida RT, et al; American Heart Association Advocacy Coordinating Committee and Stroke Council. Forecasting the future of stroke in the United States: a policy statement from the American Heart Association and American Stroke Association. Stroke. 2013;44(8): 2361-2375. 24. Fagan SC, Morgenstern LB, Petitta A, et al; NINDS rt-PA Stroke Study Group. Cost-effectiveness of tissue plasminogen activator for acute ischemic stroke. Neurology. 1998;50(4):883-890. 25. Tan Tanny SP, Busija L, Liew D, Teo S, Davis SM, Yan B. Cost-effectiveness of thrombolysis within 4.5 hours of acute ischemic stroke: experience from Australian stroke center. Stroke. 2013;44(8):2269-2274. 26. Tung CE, Win SS, Lansberg MG. Cost-effectiveness of tissue-type plasminogen activator in the 3- to 4.5-hour time window for acute ischemic stroke. Stroke. 2011;42(8):2257-2262.
JAMA Neurology February 2015 Volume 72, Number 2 (Reprinted)
Copyright 2015 American Medical Association. All rights reserved.
Downloaded From: http://archneur.jamanetwork.com/ by a CAMBRIDGE UNIVERSITY LIBRARY User on 05/14/2015
jamaneurology.com
Special Communication
Implementing a Mobile Stroke Unit Program in the United States Why, How, and How Much? Suja Rajan, PhD; Sarah Baraniuk, PhD; Stephanie Parker, RN, BSN; Tzu-Ching Wu, MD; Ritvij Bowry, MD; James C. Grotta, MD
IMPORTANCE There are many ways a mobile stroke unit (MSU) might prove valuable for
patients with ischemic and hemorrhagic stroke, such as earlier recognition, more accurate triage, improved management of blood pressure and other critical physiological variables, and eventually earlier implementation of effective therapies. The MSU may be particularly valuable for treatment of patients with acute ischemic stroke with tissue plasminogen activator (tPA) within 4.5 hours of symptom onset, the most evidence-based effective emergency treatment for the most prevalent stroke diagnosis. OBJECTIVES To review existing data on prehospital stroke treatment, especially relevant to MSU technology, to identify gaps in our understanding of MSU feasibility, especially relevant to applying the MSU strategy in the United States, and to describe the Houston MSU program and clinical trial. EVIDENCE REVIEW Published data from English-language journals in PubMed from 1995 to present reviewing early treatment with tPA and prehospital stroke evaluation and treatment. FINDINGS The MSU may result in an overall shift toward earlier evaluation and treatment with tPA, particularly into the first hour after symptom onset, leading to substantially better outcomes. As a result of improved clinical outcomes owing to earlier treatment, the costs of an MSU program may be offset by a reduction in the costs of long-term stroke care and an increase in quality-adjusted life-years, thereby supporting more widespread use of this technology. To make MSU deployment more practical, the vascular neurologist aboard the MSU must be replaced by a remote vascular neurologist connected to the MSU by telemedicine, reducing manpower requirements and costs. CONCLUSIONS AND RELEVANCE The MSU strategy could dramatically transform the way acute stroke is managed in the United States. A prospective study evaluating the logistics, outcomes, and cost-effectiveness of this approach is needed and under way. JAMA Neurol. 2015;72(2):229-234. doi:10.1001/jamaneurol.2014.3618 Published online December 8, 2014.
I
ntravenous tissue plasminogen activator (tPA) remains the only level 1A treatment for acute ischemic stroke based on several prospective randomized trials comparing tPA with standard management (SM).1 The results of these trials and of pooled analyses confirm the relationship of treatment success with time from symptom onset to initiation of treatment.2-7 However, despite 2 decades of efforts to streamline systems of care, including formation of designated stroke centers, placement of computed tomographic (CT) scanners in the emergency department (ED), dedicated 24/7 in-house stroke teams, and ED pathways to speed treatment, most patients are treated beyond 2 hours when tPA is most effective.8 While there are many reasons for this delay, inherent delay is caused by ED triage, registration, evaluation, testing, and treatment. The median door-to-needle times in stroke center EDs in the jamaneurology.com
Author Affiliations: The University of Texas School of Public Health, Houston (Rajan, Baraniuk); Department of Neurology, The University of Texas Health Science Center at Houston, Houston (Parker, Wu); Memorial Hermann–Texas Medical Center, Houston (Bowry, Grotta). Corresponding Author: James C. Grotta, MD, Memorial Hermann–Texas Medical Center, Clinical Innovation and Research Institute, 6410 Fannin St, Ste 1423, Houston, TX 77030 (james [email protected]).
United States approximate 60 minutes.9 Such delay not only likely results in fewer patients completely recovering but also reduces the total number of patients who can be treated within the 4.5-hour maximum window of tPA effectiveness, contributing to the overall low national treatment rate estimated to be about 5% of all acute ischemic strokes. Recently, the long-awaited concept of faster prehospital treatment with tPA has become a reality following the demonstration by Walter et al10 and Ebinger et al11 in Germany that CT scanners can be mounted on an ambulance to take treatment to the patient. Walter and colleagues showed that by using this MSU concept, treatment with tPA can be carried out safely and accurately with a median time between onset and treatment of 70 to 80 minutes. Ebinger and colleagues showed that an MSU resulted in increased treat(Reprinted) JAMA Neurology February 2015 Volume 72, Number 2
Copyright 2015 American Medical Association. All rights reserved.
Downloaded From: http://archneur.jamanetwork.com/ by a CAMBRIDGE UNIVERSITY LIBRARY User on 05/14/2015
229
Clinical Review & Education Special Communication
ment rates (from 21% to 33%), 25-minute-shorter time to treatment, and increased proportion of patients treated within 90 minutes of symptom onset (58% with an MSU vs 37% without). This success has the potential to result in substantially improved outcomes for patients with acute ischemic stroke and dramatically alter the way patients with acute stroke are managed. Achieving faster treatment is the primary mechanism by which MSU management might result in better outcomes from tPA treatment. Better outcomes should result from an overall shift toward earlier evaluation and treatment with tPA, particularly into the first hour after symptom onset. Another result of faster treatment via the MSU is that there might be an increase in the absolute number of tPAtreated patients. It is logical that if all patients are evaluated sooner using the MSU, more will be treated within the 4.5-hour window that is currently recommended by published guidelines.1 If use of the MSU can increase the total number of patients with stroke being evaluated and treated within this window, it should increase the total number of patients recovering from their stroke. The MSU strategy could dramatically transform the way acute stroke is managed in the United States, but there are critically important gaps not addressed by pilot studies already completed in Germany. These include the ability to adapt the MSU to the logistics of the more complex US health care system, reliability of telemedicine (TM) on the MSU, clinical outcomes, and costs. These must be addressed by a trial executed in the United States.
Logistics To our knowledge, there is no experience using the MSU in the United States; therefore, we do not know how much time can be saved by their use in the United States, where traffic patterns, distances, market forces, and local regulations differ from Europe. Certainly, the impact of any change in a stroke system of care such as MSU deployment will be location specific and will differ between urban and rural areas. For instance, in Houston, Texas, where we have a longestablished coordinated network of stroke centers and emergency medical services (EMS) providers as well as a central database of all stroke centers, current stroke management is excellent. In 2013, 300 patients (22% of patients with ischemic stroke transported) arrived at one of the 3 comprehensive stroke centers (CSCs) in the city within 2 hours of symptom onset and 199 (14% of the total number of patients with ischemic stroke and 90% of tPA-eligible patients) were treated with tPA within 3 hours of symptom onset. At Memorial Hermann–Texas Medical Center between January and August 2013 (8 months), 74 patients were delivered by EMS and treated with intravenous tPA within 4.5 hours of symptom onset, with a median time between onset and treatment of 123.5 minutes. It remains to be seen how much these statistics could be improved by an MSU strategy grafted onto the existing EMS. There are several potential complexities with earlier treatment on the MSU that also need to be addressed. One is the increased chance of treating patients with stroke mimics, eg, other pathologies such as migraine or seizures, or patients with transient ischemic attacks, eg, those who would recover within 24 hours even without treatment (this does not include strokes with negative findings on magnetic resonance imaging or “aborted” strokes that are reversed by tPA). Regarding stroke mimics, Ebinger et al11 reported 230
Mobile Stroke Unit Program
the rate of stroke mimic treatment with the MSU to be 2% and no different from that with SM. These results were dependent on having a physician experienced in stroke treatment aboard the MSU, and such expertise in person or via TM will be necessary in the United States as well. An important potential advantage of the MSU concept that extends beyond treating more patients more quickly may be earlier, accurate identification of stroke mimics and appropriate prioritization of patients with true stroke. The transient ischemic attack issue may be more difficult to assess. The incidences of patients completely recovering within 24 hours in the placebo arm of the National Institute of Neurological Disorders and Stroke (NINDS) study were 2.4% in patients treated 91 to 180 minutes after symptom onset and 2.1% in patients treated within 0 to 90 minutes.2 Because of the temporal distribution of randomized patients in the NINDS study (as will be described later), this does not suggest a dramatic increase with earlier treatment between 80 and 180 minutes. However, this incidence could be higher in patients evaluated by the MSU within the first 60 minutes after onset. To solve this dilemma, we need to observe patients in the prehospital stage who we determine would meet criteria for treatment but for whom we do not activate the MSU and then to determine what proportion of them completely recover by the time of tPA decision in the ED. This would allow us to approximate the percentage of patients with transient ischemic attack in the cohort of tPAtreated patients with MSU dispatch. A third potential problem might be an increase in the number of intracerebral hemorrhages, angioedema, or other complications of tPA treatment with MSU management. German studies have not reported an increase in these complications with MSU management,10,11 and assuming stroke expertise aboard the MSU or via TM, we would expect the same results in the United States. Another question is the impact of MSU deployment on intraarterial treatment (IAT). Recent trials12-14 failed to show a benefit for IAT compared with tPA or as an adjunctive approach to tPA. However, post hoc analysis from the Interventional Management of Stroke III study13 showed that patients who achieve recanalization within 4 to 5 hours from symptom onset have potential to benefit most from IAT, so if and when IAT becomes widely used, reducing time to treatment will be important. Deployment of the MSU might potently reduce time to IAT treatment by allowing prehospital identification of patients with probable large artery occlusion, facilitating their inhospital treatment by prehospital notification, earlier assembly of the endovascular team and angiography suite preparation, and shorter in-house delays incurred by acquiring imaging and laboratory data and treating with tPA, perhaps allowing bypass of ED evaluation altogether.
Telemedicine Eventually, the widespread use of MSUs will depend on adequate manpower to guide treatment. Our preliminary experience and the data from Germany suggest that the ratio of MSU alerts from EMS dispatch to tPA treatments is at least 10 to 1, making it impractical to have a vascular neurologist (VN) aboard the MSU for all calls. However, the decision whether to give tPA based on clinical criteria requires training, experience, and careful judgment. Experience with TM, including at our own center, has shown it to
JAMA Neurology February 2015 Volume 72, Number 2 (Reprinted)
Copyright 2015 American Medical Association. All rights reserved.
Downloaded From: http://archneur.jamanetwork.com/ by a CAMBRIDGE UNIVERSITY LIBRARY User on 05/14/2015
jamaneurology.com
Mobile Stroke Unit Program
Special Communication Clinical Review & Education
be a feasible, accurate, and safe way to extend this expertise to hospitals where it is lacking and to guide and increase tPA treatment in such locations.15,16 We recently demonstrated the feasibility and accuracy of TM assessment of actors simulating patients with stroke in ambulances using existing technology. In the Pre-hospital Utility of Rapid Stroke Evaluation Using In-Ambulance Telemedicine (PURSUIT) study,15 trained actors portrayed 10 unique stroke scenarios, each conducted 4 times, and were retrieved and transported by Houston Fire Department EMS to our stroke center. A remote VN performed remote assessments in real time and obtained clinical data points and National Institutes of Health Stroke Scale scores using the RP-Xpress device (InTouch Health). In 34 of the 40 scenarios (85%), the teleconsultation was conducted without major technical complication. The absolute agreement for intraclass correlation was 0.997 (95% CI, 0.992-0.999) for the National Institutes of Health Stroke Scale scores obtained during the real-time sessions. Matching of real-time assessments occurred for 88% of the National Institutes of Health Stroke Scale scores (30 of 34 scenarios) within 2 points and 96% of the clinical information. However, TM has not been tested for treating actual patients with stroke with tPA in the prehospital environment. This will likely be a more complex exercise given the need for multiple parallel tasks (neurological assessment; vital signs; patient positioning, attention to comfort, and possible restraint; CT scanning; point-ofcare laboratory measurement; drug mixing and administration) being carried out urgently by several people within the confined space of an MSU.
Clinical Outcomes Faster treatment using the MSU will allow us, for the first time, to answer an important scientific question by making treatment possible within the first 60 to 80 minutes after symptom onset when tPA is likely to have its greatest effect. Preliminary data from Ebinger et al17 showed that 31% of patients treated with tPA using their MSU were treated within 60 minutes of onset compared with 4.9% with SM. These rates, although based on small numbers, compare very favorably with treatment rates within the first 60 minutes from previous trials and current databases. Among the 302 patients treated with tPA vs placebo within 90 minutes of onset in the NINDS Stroke Study, only 2 were randomized within 60 minutes of onset (both were randomized to the placebo group) and only 41 were randomized between 61 and 80 minutes after onset.2 The latest pooled data from all randomized tPA trials added only 2 additional patients randomized within the first 60 minutes. Only 1.6% of patients were treated within 60 minutes of onset in the Safe Implementation of Treatments in Stroke–International Stroke Thrombolysis Registry (SITS-ISTR).18 In Helsinki, Finland, using a streamlined system of stroke triage, 10% of patients were treated within 70 minutes of symptom onset,19 and of 58 353 patients treated with tPA in the Get With the Guidelines–Stroke program, fewer than 1000 were treated within 60 minutes of symptom onset.8 There is very little information on how much improvement in clinical outcomes will occur with treatment within the first 60 to 80 minutes. While data from multiple studies and pooled analyses conjamaneurology.com
firm a strong inverse relationship between time to treatment and recovery, the slope and shape of that relationship within the first 90 minutes after stroke symptom onset are very uncertain as reflected in the wide confidence intervals surrounding outcomes in various pooled analyses and the total absence of data before 60 minutes. It is possible that, owing to collateral flow, human penumbral tissue can survive long enough so that there would be little advantage to earlier treatment within the 90-minute epoch. However, data presented by Ebinger et al17 demonstrated that patients treated within the first 60 minutes of onset by their MSU had an odds ratio of 1.93 (95% CI, 1.09-3.41) of discharge to home compared with later treatment. These data have not yet been subjected to peer review to be certain that confounding variables were considered in the comparison, and longer-term outcomes were not reported. Deployment of the MSU in the United States will allow us to determine the benefit of tPA on long-term outcome in prospectively studied patients treated within 0 to 60 or 60 to 80 minutes vs later treatment. Following the precedent of recent tPA pooling projects,3,4 US data can be pooled with European data to increase our power to arrive at an accurate answer. Eventually, to completely assess the value of the MSU strategy, we need to answer the larger question of whether all patients eligible for tPA in the prehospital setting have better outcomes if managed via an MSU or SM. This requires an intention-to-treat study that will include all eligible patients (whether or not they eventually receive tPA) evaluated at the same time in the prehospital setting and by the same personnel and will compare patients assigned to MSU vs SM. Assessing and enrolling all patients while they are still in the prehospital phase will ensure that patients receiving SM who are enrolled are comparable to the patients managed with the MSU and will also allow us to account for those patients who would have met criteria for prehospital treatment but could not be treated within the 4.5-hour window because they were managed on the SM pathway.
Cost-Effectiveness Stroke is among the top 15 most expensive conditions treated in US hospitals and among the top 10 most expensive conditions billed to Medicare.20 Medicare bears the highest cost burden of the disease; almost 60% of stroke-related hospital costs and more than 60% of overall stroke-related costs are borne by Medicare.20,21 Non– nursing home stroke care constitutes more than 10% of Medicare’s budget.22 As the US population ages, the incidence and prevalence of this disease will increase, and hence costs associated with stroke and the cost burden of Medicare will substantially increase. It is projected that by 2030, 4% of the American population will have had a stroke and the total medical cost of stroke will be nearly $200 billion (in 2010 US dollars), which is a 250% increase as compared with the medical costs as of 2012.23 Ischemic stroke accounts for 87% of all stroke events. The use of tPA in ischemic stroke, and specifically the early use of tPA, has been shown to be both clinically efficacious and cost-effective. Fagan et al24 demonstrated that the use of tPA reduced hospital length of stay, with higher discharge to homes instead of inpatient rehabilitation or nursing homes. Their Markov analysis predicted an increase in quality-adjusted life-years (QALYs) with 94% probability (Reprinted) JAMA Neurology February 2015 Volume 72, Number 2
Copyright 2015 American Medical Association. All rights reserved.
Downloaded From: http://archneur.jamanetwork.com/ by a CAMBRIDGE UNIVERSITY LIBRARY User on 05/14/2015
231
Clinical Review & Education Special Communication
Mobile Stroke Unit Program
Figure 1. Mobile Stroke Unit A
B
Exterior (A) and interior (B) of the mobile stroke unit.
and a decrease in poststroke first-year costs with 93% probability among patients receiving tPA, clearly demonstrating the value of using tPA. Tan Tanny et al25 documented that tPA use within 4.5 hours of stroke occurrence had an incremental cost-effectiveness ratio of $1478/QALY (Australian) during the first year following the tPA treatment; tPA use improved QALYs but also marginally increased costs. Tung et al26 performed a lifetime cost-effectiveness analysis for the use of tPA and found an increase in both lifetime costs and QALYs with tPA administered within 4.5 hours of ischemic stroke, with an incremental cost-effectiveness ratio of $21 978/QALY. Treatments with an incremental cost-effectiveness ratio of less than $50 000/QALY are well accepted as cost-effective and generally reimbursed by health care decision makers. The literature provides ample evidence on the economic impact of stroke and cost-effectiveness of tPA. However, to our knowledge, there are no studies within the United States that specifically evaluate the economic impact of an MSU using TM to improve early tPA administration. The MSU strategy entails significant capital investment in outfitting an ambulance with a CT scanner and TM equipment as well as physician and nurse staffing changes and staff training to support MSU operations. These significant fixed costs imply that the agency operating the MSU and the staff supporting it should receive additional reimbursements to ensure the financial viability and sustainability of the MSU operation. The health care payers are more likely to provide additional reimbursement if we can demonstrate improvement in patient outcomes and QALYs and possible poststroke cost savings for the payers and also provide estimates of incremental fixed cost increases per patient due to the dispatch of an MSU. Regarding initial setup costs, purchasing, equipping, and deploying an ambulance with a CT scanner, point-of-care laboratory testing capabilities, TM capability, and other needed equipment in Houston has a cost of approximately $600 000. Formal cost analysis would have to balance these costs plus costs for staffing and maintenance against any reduction in the total hospital and longterm care costs to the health care system for each patient with an ischemic stroke treated on the MSU, estimated to average approximately $140 000 per patient in 1999 US dollars, undoubtedly much higher today. 232
Houston MSU To address these gaps in our knowledge, we have introduced at the Texas Medical Center in Houston the nation’s first MSU funded by donations from grateful patients, local philanthropists, and the Frazer ambulance company. We are not only the first center in the United States to put an MSU into operation but also the first (and only) group to use it for clinical research purposes. The MSU was constructed at the Frazer factory (Figure 1 and Figure 2; http://www.frazerbilt.com/Videos/watch.php?id=784 and https: //www.youtube.com/watch?v=y1m64EL-k5I&index=6&list =UU7MwkvzzoUJ1SOHHl-PvLBQ). Many steps were required for licensing, inspecting, equipping, insuring, and credentialing the MSU and integrating MSU operations into the Houston Fire Department EMS dispatch, firstresponder, and paramedic system, which services the various Houston stroke centers. A full description of these efforts, which required 6 to 12 months of dedicated effort, is beyond the scope of this review, but 2 important overriding concepts that led to our ability to be up and running within 1 year of starting the project were an understanding that our overall goal is to conduct clinical research rather than to assume that the MSU should be implemented prima facie and that this is to be a collaborative effort of community stroke stakeholders, including EMS and all 3 of the CSCs and stroke teams at the Texas Medical Center.
The Benefits of Stroke Treatment Delivered Using a Mobile Stroke Unit Compared to Standard Management by Emergency Medical Services Study We have begun the Benefits of Stroke Treatment Delivered Using a Mobile Stroke Unit Compared to Standard Management by Emergency Medical Services (BEST-MSU) Study with the specific aims to determine the following: (1) the logistics and clinical outcome of the MSU vs SM; (2) the agreement between a VN remotely assessing a patient with suspected stroke via TM and in-person assessment by a VN in the MSU; and (3) the incremental cost and effectiveness as-
JAMA Neurology February 2015 Volume 72, Number 2 (Reprinted)
Copyright 2015 American Medical Association. All rights reserved.
Downloaded From: http://archneur.jamanetwork.com/ by a CAMBRIDGE UNIVERSITY LIBRARY User on 05/14/2015
jamaneurology.com
Mobile Stroke Unit Program
Special Communication Clinical Review & Education
Figure 2. Treating the First Patient A
B
Treating our first patient 78 minutes after symptom onset (A) with simultaneous telemedicine backup (B).
sociated with MSU vs SM using standard cost-effectiveness analyses. This is a prospective cohort study with block-randomized MSU or SM deployment weeks and blinded assessment of both trial entry and clinical outcomes. After we are notified of a patient with possible stroke by EMS, the patient is included in the study if he or she was last seen healthy possibly within 4.5 hours of symptom onset, has a medical history and physical and/or neurological examination findings consistent with acute stroke, and meets no tPA exclusions per guidelines1 prior to CT scanning or baseline laboratory tests. Written informed consent is obtained from the patient (if competent) or legal representative to include the patient in the database, but it is not needed for treatment, which follows published guidelines. We intend to enroll 248 patients, of whom we expect 186 will receive tPA. The MSU operates in parallel with EMS routine. After receiving a 9-1-1 call, the EMS dispatcher immediately dispatches the nearest available emergency medical technician or paramedic team. Immediately after emergency medical technician or paramedic dispatch, the MSU team is activated via dedicated pager. The MSU is dispatched on only 50% of weeks according to a blocked randomization schedule. On non-MSU dispatch weeks (SM weeks), the MSU team is still dispatched to the scene but travels in a private vehicle. The MSU is staffed by a Houston Fire Department paramedic, certified CT technician, VN (VN faculty or fellow), and registered nurse experienced in both clinical research and acute stroke management. The MSU team alerts the on-call TM VN, who immediately connects from the central TM office at the University of Texas Medical School at Houston to the mobile TM device on the MSU. The MSU is stationed in the heart of the Texas Medical Center and surrounded by the 3 CSCs, which are the destination of all patients with stroke who are picked up by EMS within a predetermined 5-mile radius catchment area surrounding the Texas Medical Center. We have found that the 5-mile radius catchment area will allow dispatch and arrival of the MSU at the emergency site while EMS is still on scene evaluating the patient. This catchment area is expanding as MSU team efficiency improves. Once on scene, the patient’s medical history, vital signs, fingerstick glucose level, and physical examination findings are jointly evaluated by the EMS paramedics and MSU VN and nurse, and if the jamaneurology.com
patient has signs and symptoms of stroke possibly within 4.5 hours of onset meeting published guidelines for tPA, he or she is moved into the MSU, enrolled into the study for purposes of answering the specific aims, and assigned to the MSU arm. If the patient does not have signs and symptoms of a stroke, is clearly outside the 4.5hour window, has other tPA exclusions, or is clinically unstable (such as requiring pressor or ventilator support), he or she is managed and transported per EMS routine. These latter patients will be considered screen failures. Once a patient is moved into the MSU, a venous blood sample is analyzed by a point-of-care laboratory (blood glucose level, hematocrit, platelet level, international normalized ratio if needed) by the MSU paramedic and nurse. Noncontrast CT scanning of the head is performed by the CT technician, and the results are uploaded onto the picture archiving and communication system for immediate visualization on the MSU laptop computer and read by the on-site MSU VN. Blood pressure is managed per guidelines. If the patient meets all published inclusion and exclusion criteria for thrombolysis, intravenous tPA is given without delay. Simultaneous to these events, the patient is evaluated by the TM VN using a portable RP-Xpress unit (InTouch Health) attached to the MSU gurney or handheld by the paramedic to optimize viewing. The MSU paramedic helps the TM VN record the essentials of the medical history, vital signs, physical examination findings, and laboratory results. The CT images on the MSU are securely and wirelessly sent via on-board 4G connection in real time to a secure picture archiving and communication system for review by the TM VN. The TM VN makes his or her own independent decision about whether the patient qualifies for tPA treatment. After the tPA bolus is given and infusion is started or the decision is made to withhold tPA, the patient is transported in the MSU with the research nurse and VN to the appropriate CSC. On SM weeks, the MSU is not dispatched but the MSU nurse or VN travels to the scene by car. Once on scene, the patient’s medical history, vital signs, finger-stick glucose level, and physical examination findings are jointly evaluated by the EMS paramedics and MSU VN or nurse. If the patient meets all inclusion criteria except laboratory and CT results, the patient is enrolled into the study and as(Reprinted) JAMA Neurology February 2015 Volume 72, Number 2
Copyright 2015 American Medical Association. All rights reserved.
Downloaded From: http://archneur.jamanetwork.com/ by a CAMBRIDGE UNIVERSITY LIBRARY User on 05/14/2015
233
Clinical Review & Education Special Communication
Mobile Stroke Unit Program
signed to the SM arm. The MSU research nurse or VN travels from the sceneandmeetsEMSandthepatientatthedesignatedCSCED.There, the hospital-based stroke team manages the patient as per CSC routine, the same standard-of-care analyses are carried out as with the MSU treatment, and intravenous tPA is given if the patient meets inclusion and exclusion criteria according to published guidelines. All enrolled patients are followed up for both clinical and economic outcome measures with clinical outcomes assessed by an investigator blinded to treatment group. To date, the study has been up and running for 1 month. We have been enrolling 1 patient approximately every 2 days and treating 2 ARTICLE INFORMATION Accepted for Publication: October 6, 2014. Published Online: December 8, 2014. doi:10.1001/jamaneurol.2014.3618. Author Contributions: Dr Grotta had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Rajan, Baraniuk, Parker, Bowry, Grotta. Acquisition, analysis, or interpretation of data: Wu, Grotta. Drafting of the manuscript: Rajan, Parker, Bowry, Grotta. Critical revision of the manuscript for important intellectual content: Baraniuk, Parker, Wu, Grotta. Statistical analysis: Baraniuk. Administrative, technical, or material support: Rajan, Parker, Wu, Bowry, Grotta. Study supervision: Parker, Grotta. Conflict of Interest Disclosures: Dr Grotta reported having consulting contracts with Frazer Ltd, Specialists on Call, Haemnonetics, and Lundbeck. No other disclosures were reported. Funding/Support: The Houston Mobile Stroke Unit project has received limited financial support from Genentech and Covidien. Role of the Funder/Sponsor: Genentech and Covidien had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication. REFERENCES
Conclusions The MSU strategy could dramatically transform the way acute stroke is managed in the United States. A prospective study evaluating the logistics, outcomes, and cost-effectiveness of this approach is needed and under way.
4. Lees KR, Bluhmki E, von Kummer R, et al; ECASS, ATLANTIS, NINDS and EPITHET rt-PA Study Group. Time to treatment with intravenous alteplase and outcome in stroke: an updated pooled analysis of ECASS, ATLANTIS, NINDS, and EPITHET trials. Lancet. 2010;375(9727):1695-1703. 5. Marler JR, Tilley BC, Lu M, et al. Early stroke treatment associated with better outcome: the NINDS rt-PA Stroke Study. Neurology. 2000;55(11): 1649-1655. 6. Lansberg MG, Schrooten M, Bluhmki E, Thijs VN, Saver JL. Treatment time-specific number needed to treat estimates for tissue plasminogen activator therapy in acute stroke based on shifts over the entire range of the modified Rankin Scale. Stroke. 2009;40(6):2079-2084. 7. Saver JL, Fonarow GC, Smith EE, et al. Time to treatment with intravenous tissue plasminogen activator and outcome from acute ischemic stroke. JAMA. 2013;309(23):2480-2488. 8. Fonarow GC, Smith EE, Saver JL, et al. Timeliness of tissue-type plasminogen activator therapy in acute ischemic stroke: patient characteristics, hospital factors, and outcomes associated with door-to-needle times within 60 minutes. Circulation. 2011;123(7):750-758. 9. Fonarow GC, Zhao X, Smith EE, et al. Door-to-needle times for tissue plasminogen activator administration and clinical outcomes in acute ischemic stroke before and after a quality improvement initiative. JAMA. 2014;311(16):1632-1640. 10. Walter S, Kostopoulos P, Haass A, et al. Diagnosis and treatment of patients with stroke in a mobile stroke unit versus in hospital: a randomised controlled trial. Lancet Neurol. 2012;11(5):397-404.
1. Jauch EC, Saver JL, Adams HP Jr, et al; American Heart Association Stroke Council; Council on Cardiovascular Nursing; Council on Peripheral Vascular Disease; Council on Clinical Cardiology. Guidelines for the early management of patients with acute ischemic stroke: a guideline for healthcare professionals from the American Heart Association/American Stroke Association. Stroke. 2013;44(3):870-947.
11. Ebinger M, Winter B, Wendt M, et al; STEMO Consortium. Effect of the use of ambulance-based thrombolysis on time to thrombolysis in acute ischemic stroke: a randomized clinical trial. JAMA. 2014;311(16):1622-1631.
2. National Institute of Neurological Disorders and Stroke rt-PA Stroke Study Group. Tissue plasminogen activator for acute ischemic stroke. N Engl J Med. 1995;333(24):1581-1587.
13. Broderick JP, Palesch YY, Demchuk AM, et al; Interventional Management of Stroke (IMS) III Investigators. Endovascular therapy after intravenous t-PA versus t-PA alone for stroke. N Engl J Med. 2013;368(10):893-903.
3. Hacke W, Donnan G, Fieschi C, et al; 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(9411):768-774.
234
patients per week with tPA. Of the 9 tPA-treated patients to date, 2 were treated within 60 minutes and 3 between 61 and 80 minutes from symptom onset.
12. Ciccone A, Valvassori L; SYNTHESIS Expansion Investigators. Endovascular treatment for acute ischemic stroke. N Engl J Med. 2013;368(25): 2433-2434.
14. Kidwell CS, Jahan R, Gornbein J, et al; MR RESCUE Investigators. A trial of imaging selection and endovascular treatment for ischemic stroke. N Engl J Med. 2013;368(10):914-923.
15. Wu TC, Nguyen C, Ankrom C, et al. Prehospital utility of rapid stroke evaluation using in-ambulance telemedicine: a pilot feasibility study. Stroke. 2014; 45(8):2342-2347. 16. Johansson T, Wild C. Telemedicine in acute stroke management: systematic review. Int J Technol Assess Health Care. 2010;26(2):149-155. 17. Ebinger M, Wendt M, Rozanski M, et al. Golden hour-thrombolysis by starting treatment before hospital arrival: the Pre-hospital Acute Neurological Treatment and Optimization of Medical Care in Stroke Study (PHANTOM-S) [abstract 104]. Stroke. 2014;45: A104. 18. Wahlgren N, Ahmed N, Dávalos A, et al; SITS Investigators. Thrombolysis with alteplase 3-4.5 h after acute ischaemic stroke (SITS-ISTR): an observational study. Lancet. 2008;372(9646): 1303-1309. 19. Meretoja A, Strbian D, Mustanoja S, Tatlisumak T, Lindsberg PJ, Kaste M. Reducing in-hospital delay to 20 minutes in stroke thrombolysis. Neurology. 2012;79(4):306-313. 20. Torio CM, Andrews RM. National Inpatient Hospital Costs: The Most Expensive Conditions by Payer, 2011. Rockville, MD: Agency for Health Care Policy & Research; 2013. HCUP statistical brief 160. 21. Trogdon JG, Finkelstein EA, Nwaise IA, Tangka FK, Orenstein D. The economic burden of chronic cardiovascular disease for major insurers. Health Promot Pract. 2007;8(3):234-242. 22. Cohen JW, Krauss NA. Spending and service use among people with the fifteen most costly medical conditions, 1997. Health Aff (Millwood). 2003;22(2):129-138. 23. Ovbiagele B, Goldstein LB, Higashida RT, et al; American Heart Association Advocacy Coordinating Committee and Stroke Council. Forecasting the future of stroke in the United States: a policy statement from the American Heart Association and American Stroke Association. Stroke. 2013;44(8): 2361-2375. 24. Fagan SC, Morgenstern LB, Petitta A, et al; NINDS rt-PA Stroke Study Group. Cost-effectiveness of tissue plasminogen activator for acute ischemic stroke. Neurology. 1998;50(4):883-890. 25. Tan Tanny SP, Busija L, Liew D, Teo S, Davis SM, Yan B. Cost-effectiveness of thrombolysis within 4.5 hours of acute ischemic stroke: experience from Australian stroke center. Stroke. 2013;44(8):2269-2274. 26. Tung CE, Win SS, Lansberg MG. Cost-effectiveness of tissue-type plasminogen activator in the 3- to 4.5-hour time window for acute ischemic stroke. Stroke. 2011;42(8):2257-2262.
JAMA Neurology February 2015 Volume 72, Number 2 (Reprinted)
Copyright 2015 American Medical Association. All rights reserved.
Downloaded From: http://archneur.jamanetwork.com/ by a CAMBRIDGE UNIVERSITY LIBRARY User on 05/14/2015
jamaneurology.com
