Telestroke: Delivery and Design




© Springer International Publishing Switzerland 2015
Patrick D Lyden (ed.)Thrombolytic Therapy for Acute Stroke10.1007/978-3-319-07575-4_11


11. Telestroke: Delivery and Design



Konrad H. Schlick  and Brett C. Meyer 


(1)
Department of Neurosciences, UCSD School of Medicine, 200 W. Arbor Drive, MON 3rd Floor, Suite 3, San Diego, CA 92103, USA

(2)
UCSD Stroke Center, UC San Diego Health Sciences, San Diego, CA, USA

 



 

Konrad H. Schlick



 

Brett C. Meyer (Corresponding author)




Introduction


Telemedicine for stroke is a dynamic and evolving process. Since the inception of medicine, care providers have developed new and improved methods of providing medical evaluations and treatments to their patients. Telemedicine is no different. Over the past 15 years, telemedicine has spread exponentially into the stroke care landscape, providing a robust emergency stroke management option for providing care to even more patients. Telestroke is a new way of doing the same business of caring for patients, now just on a wider scale and less limited by geographic barriers.

Telemedicine is a process and requires numerous essential elements for successful implementation and sustainable use [1]. Telestroke applications have undergone important periods of development, testing, efficacy analyses, technology iterations, model adaptations, and surpassed multiple hurdles to reach the current stage of development. Detailing each element is beyond the scope of this chapter as variability exists depending on chosen strategy and definitions of success. The numerous systems, vendors, techniques, and fast-paced arena of technology development preclude the ability to detail each specific type of telestroke application sufficiently. The goal of this chapter is to bring specific concept familiarity to those desiring a deeper understanding of telestroke applications, in an attempt to enable its implementation in standard stroke systems of care in a dynamic marketplace.


History of rT-PA


The NINDS rt-PA trial demonstrated efficacy of rt-PA in acute ischemic stroke, and emphasized the benefit of early administration [2]. This treatment however remains widely underused, with less than 5 % of ischemic stroke patients receiving rt-PA [3]. Some of the treatment disparities are due to geographic limitations and lack of specialist availability for acute stroke victims. Countries with systems of streamlined stroke care tend to have developed larger stroke centers (primary or comprehensive level) that are generally located in urban areas. Disparities are further created by lack of stroke specialist expertise in many rural and community hospitals. These disparities elucidate a need for other strategies to augment the administration of rt-PA for acute stroke patients.


History of Telemedicine


Telemedicine has been broadly defined as “the use of telecommunications technologies to provide medical information and services” [4]. As such, the use of telephone, fax and even email could be considered ‘telemedicine’. However, in keeping with current accepted guidelines, telemedicine will be referred to as using high definition 2 way audio and 2 way video conferencing technology to provide medical services [5, 6]. Telemedicine was initially spearheaded by teleradiology services. Over the years, more specialties began to utilize technologies both in the form of synchronous telemedicine evaluations (live video conferencing) and asynchronous “store and forward” applications (such as dermatology, pathology or radiology imaging review).

The history of telemedicine dates as far back as the beginnings of telephone use. Aronson describes a physician listening to a child’s cough over the telephone in 1879, and deciding it did not warrant emergent evaluation because it did not sound like croup [7]. The early twentieth century saw the transmission of electrocardiographic an electroencephalographic data over analogue telephone networks [8]. In 1959 Wittson used two-way interactive television for telepsychiatry consultations between two Nebraska hospitals [9]. That same year, Jutra described his initial experiences with teleradiology, in which he transmitted fluoroscopic data over coaxial cable between hospitals [10].

Given the importance of imaging in the evaluation of acute stroke, it is prudent to describe the evolution of teleradiology as well. Jutras’s initial exchanges of fluoroscopy data between two Montreal hospitals via coaxial cable demonstrated the ability to transmit radiographic data between two different locations; as technology advanced this ability became more rapid and refined. Portability of imaging devices has afforded imaging outside of conventional radiology suites—portable CXR devices have simplified obtaining chest films (especially in the inpatient settings). Today, teleradiology is standard in many hospitals, with radiologic interpretation transmitted by telephone or electronically to other caregivers within the hospital, other cities, or even other countries.

The widespread use of cellular phones has stimulated investigation into its use as medical technology, and Yamada et al. used mobile phone cameras to evaluate ED brain imaging to assist in triaging patients [11]. The 1990s saw the use of the first mobile MRI scanners [12]. More recently, the PHANTOM-S investigators transported CT scanners in the field and found that neuroimaging in the field (as well as rt-PA administration in the field) were feasible [13].


History of Telestroke


Telestroke represents the application of telemedicine technology in the stroke care field to provide stroke expertise in settings where it was not previously available due to limitations (usually geographic). The importance of telestroke is reflected both by its inclusion in two scientific statements [5, 6] and by its inclusion in the 2013 American Stroke Association guidelines for early management of patients with acute ischemic stroke [14].

The use of the term “telestroke” to describe telemedicine for stroke management was initially coined by Levine and Gorman in 1999 [15]. They proposed using telemedicine modalities in the diagnosis and treatment of acute stroke, in part to address the dearth of stroke specialists especially in rural and community hospitals, where treatment of acute ischemic stroke is frequently hampered by limited experience. Even at that time, it was suggested as a method to disseminate stroke expertise, augment rt-PA administration, assist with secondary stroke prevention, help with stroke rehabilitation, improve education and even perform clinical research enrollments. Today the use of this technology for stroke treatment is becoming more prevalent, and with continued telestroke experiences stroke treatment rates are increasing.

At its inception, a typical telestroke model involved “point-to-point” audiovisual interactions, requiring the stroke specialist to travel to a dedicated workstation (typically in a hospital setting) from which remote consultation could be performed. The travel time to a workstation and on-site staffing requirements set some limitations to this model. A “Telestroke 2.0” model that is gaining popularity can now facilitate site-independent, mobile web-based stroke care solutions as well [16]. This allows greater mobility while retaining high-quality video teleconferencing.


Telestroke: The Need



Acute Stroke Management with rt-PA


Along with other potential uses, telestroke can help disseminate stroke expertise. Since rt-PA treatment rates continue to be low even many years after drug approval, telestroke’s primary driving force has been to augment rt-PA treatments. Without support, small hospitals, without immediate access to stroke specialists, may miss potential treatments or providers may choose not to treat fearing liability if the patient were to subsequently bleed. The need for telestroke in remote and rural communities is clear. It is estimated that only 55 % of Americans live within 60 min of a primary stroke center, leaving a significant amount of patients without direct access to acute stroke specialists in a timely manner [17]. This need also exists in more metropolitan areas, as rapid treatments are correlated with improved outcome [18]. The use of telestroke to begin the rt-PA evaluation itself, even if a stroke specialist soon arrives, can potentially even decrease the time required to provide treatment by the bedside specialist.


Other Therapies


Similarly, in the current spectrum of stroke care, other therapies are emerging that may also benefit from telestroke applications. Currently, rt-PA is the only proven effective therapy for acute ischemic stroke. In the future, other interventions or even neuroprotective strategies may be utilized as treatment modalities offered after telestroke evaluations [19]. Endovascular-based mechanical reperfusion is becoming more commonplace, though the optimal patient selection for this approach is still actively being evaluated. Timely telestroke evaluation may even help select patients for intra-arterial therapies, and may facilitate rapid transfer to a center (typically the hub site) where such catheter-based interventions are offered [2022]. This changing landscape of stroke care is well served by the telestroke technique.


Hospital Certification


Many hospitals do not have stroke specialists located within their geographic regions and therefore cannot provide 24 h a day/7 day a week/365 days a year coverage for acute stroke victims presenting to their facilities. In response, telestroke has become a popular technique for providing this coverage in full, or augment times when bedside coverage is not available. The Brain Attack Coalition (BAC) deems the use of telestroke modalities as a useful tool for Primary Stroke Centers (PSCs) for supporting spoke sites [23]. The PSC designation, under the auspices of the Joint Commission is often used to distinguish hospitals in a competitive marketplace.


Telestroke: Spectrum of Care and Evidence


Telemedicine for use in stroke has been evaluated first as a novelty, followed by true clinical measurements of feasibility, reliability and efficacy. With advances in technology, this technique progressed to a stage where providers could apply telestroke in clinical care. Evidence supporting this use actually lagged due to a high perception of feasibility coupled with a perception of low risk. Clinical and technical standards eventually needed to be developed. Numerous investigators, too numerous to mention here, have helped move telemedicine and telestroke into a stage where providers “should” realistically use this technique in clinical care. Further investigations, and research trials, will help further elucidate different applications, different populations, and more robust methods of telestroke delivery. In the sections below, this chapter focuses on telestroke uses throughout the spectrum of stroke care. Where appropriate, current Class and Level of evidence information has been included. It is extremely likely that this data will become more refined over time.


Pre-hospital Assessments


The effectiveness of rt-PA for acute stroke is highly time-dependent. Earlier administration is associated with improved outcomes and reduced rates of intracranial hemorrhage [24]. Decreasing the time from stroke onset to treatment will optimize care by minimizing the millions of neurons lost due to even brief delays [25].

Telemedicine applications for stroke have largely centered on evaluation of stroke in the ED. In order to minimize the time from onset to rt-PA treatment, telestroke applications have also been proposed in the prehospital realm. Applications have extended to prehospital clinical research as well.

Various prehospital stroke scales are in use by EMS personnel in order to assess patients, though the accuracy of these simple screening evaluations is limited. The majority of ambulances do not have practitioners specifically trained in evaluating stroke or its mimics. Telemedicine allows for remote evaluation prior to hospital arrival, which has the potential for improved diagnostic accuracy, appropriate triage, and stroke code activation.

Prehospital provider uncertainty of stroke diagnosis may lead to delays in care, and therefore delays in rt-PA use especially in rural areas that tend to be farther away from stroke centers where there is more experience with stroke diagnosis and treatment. Evaluation by stroke specialists before the patient arrives in an emergency department may permit early stroke diagnosis, improved coordination of care, and identification of mimics.

Rapid examination of potential stroke patients is typically performed using standardized assessments such as the NIHSS. Evaluation of video transmission from an ambulance telemedicine system has been reported by the Brain Attack Team (TeleBAT) project [26, 27]. TeleBAT ambulance assessment use was shown to be reliable, with 40–47 % of examination elements showing excellent kappa reliability, and feasible though with a low image transfer rate that was insufficient for practical use.

More recently, a pilot trial assessing the use of telestroke in the pre-hospital setting was performed [28]. The aim was to assess feasibility and reliability of performing NIHSS evaluations in simulated ambulance runs. In 40 % of the scenarios, NIHSS assessments were feasible with fair to good reliability (weighted κ-values of 0.69). The primary limitation of this work was in the failure to perform all NIHSS examinations due to wireless bandwidth inadequacies. Though the authors concluded that current technology did not result in clinical feasibility, it does show potential of telestroke in the prehospital setting. The evolution of cellular data networks may allow for improved quality of audiovisual data transmission. The nature of an ambulance moving through individual cells of a network may lead to interruptions in data transmission, thus hampering the prehospital evaluation. Increasing bandwidth of modern mobile technology may allow for higher-quality and more consistent data transmission.

The evidence surrounding the reliability and efficacy of pre-hospital use of telestroke is limited. Current technologies can provide some degree of interactive video and audio communication with ambulances, but current publications still note low frame rates and insufficient reliability. A 2009 Scientific Statement reviewing the evidence for telestroke in various settings reported that there are insufficient data to support any recommendation regarding its use [6].

Pre-hospital management has changed in other ways. Recently, clinician investigators enabled ambulances to provide acute CT imaging in an effort to decrease door to needle times [29]. This method may also allow for treatment initiation prior to hospital arrival. For now, this approach may be limited to cities with sufficient patient volume in a relatively small radius, in order to maximize cost-effectiveness. It is conceivable that advances in mobile CT technology may make this approach more widespread in the future.

Similarly, the highly time-dependent nature of acute stroke treatment hampers the investigation of new therapeutic agents. Enrolling patients into clinical trials requires time and resource allocation, and at the same time must not delay standard of care treatment. Telestroke techniques can be used to facilitate enrollment of potential study subjects prior to hospital arrival. Early feasibility work has enabled the administration of potential neuroprotective therapies in the pre-hospital setting and the initiation of clinical trial consent for potential neuroprotective therapies in the ambulance [19, 30]


Non-emergency and Emergency Settings: Reliability


The NIHSS is generally regarded as the reference standard for stroke clinical deficit scale assessments. The NIHSS is a graded neurological examination that assesses consciousness, visual field abnormalities, motor and sensory abilities, speech and language functions and inattention. The scale was developed for use in acute stroke therapy trials [31, 32], and requires only a limited time to perform [31]. Overall bedside inter-rater reliability has been documented and [31, 32] the percentage of items with excellent inter-rater reliability ranged from 31 to 38 % [31, 32]. Much of the work surrounding the use of telestroke in the non-urgent and urgent settings has focused on the reliability of performing clinical deficit scales, such as the NIHSS. In order to ensure the adequacy of remote stroke evaluation by telemedicine, the feasibility and reliability of performing clinical deficit scales using this technique was assessed in both non-acute and acute environments.

There have been studies of performing a NIHSS via telestroke in the nonacute setting. Shafqat et al. performed the first investigation of interrater agreement between bedside and telstroke NIHSS [33]. Twenty patients were examined using a point-to-point telemedicine link. Thirty-one percent of NIHSS items showed excellent agreement as measured with a weighted kappa statistic. Meyer et al. performed another study utilizing a site-independent system capable of either wired or wireless connectivity over public Internet [34]. Twenty-five patients with stroke symptoms were examined via the STRokE DOC (Stroke Team Remote Evaluation using a Digital Observation Camera) system. Sixty-seven percent of NIHSS items showed excellent weighted kappa agreement. Wiborg et al. demonstrated good to excellent agreement in testing 44 patients using 2 other stroke severity scales, the European Stroke Scale (weighted kappa 0.72–0.95) and Scandinavian Stroke Scale (weighted kappa 0.70–0.97) [35]. These, and other assessments in the non-acute setting have demonstrated reliability of performing clinical deficit scales using telestroke. A 2009 Scientific Statement reported a Class I, Level of Evidence A recommendation for the reliability of performing an NIHSS evaluation using telestroke in the non-acute setting. Similar recommendations apply for the European and Scandinavian Stroke scales (Class I, Level of Evidence A) [6].

In the emergency setting, when scales are applied via telemedicine, the remote stroke specialist typically relies on the presence of healthcare staff (who may or may not have particular stroke training) at the bedside to assist in executing exam maneuvers. Several studies suggest that telestroke application of the NIHSS is both feasible and reliable. Wang et al. investigated the reliability of performing the NIHSS in the Emergency Room or inpatient hospital [36]. Twenty patients with acute ischemic stroke were examined. There was no difference of greater than 3 points on total NIHSS score between bedside and remote examiner. Handschu et al. assessed the German version of the NIHSS within 6 and 36 h of stroke onset in 41 patients [37]. Excellent reliability was found in 100 % of 41 patients examined within 36 h of stroke onset and in 92 % of 12 patients examined within 6 h. A 2009 Scientific Statement reported a Class I, Level of Evidence A recommendation for the reliability of performing an NIHSS evaluation using telestroke in the acute setting when a bedside stroke specialist is not immediately available [6].

There have been further reports of NIHSS use in actual acute stroke evaluations. Reliability of the NIHSS has been tested in various settings, including comparing telemedicine and bedside evaluations [6]. They serve to further the feasibility, reliability and potential efficacy of using telestroke to perform clinical deficit examinations in numerous settings.


Emergency Department Setting: Efficacy


Efficacy measures have generally focused on the use of telestroke in the Emergency Department setting and have included outcomes of rt-PA administration rates, correctness of decision-making, and even patient long-term outcomes.


rt-PA Rates


Most commonly, stroke is clinically evaluated and treated in the ED. Timely evaluation and treatment of acute stroke is resource and labor-intensive, requiring 24-h availability of providers trained in its management. Use of intravenous rt-PA is particularly low in rural hospitals, where coverage by stroke teams is lacking and experience with thrombolytic medications is low. Stroke risk factors are more prevalent in rural areas, which makes access to time-sensitive stroke treatment even more critical in those areas. Transport to accredited stroke centers may be hindered by the narrow time window for stroke treatment. Regarding the initial driving force behind telestroke, that of improving rt-PA rates, it is now apparent that rt-PA rates can be significantly increased using this technology. In fact, although denominators are sometimes difficult to determine, estimates note a ten-fold increase in rt-PA rates [38].

The original REACH (Remote Evaluation of Acute isCHemic stroke) program included telemedicine consultation to hospitals in rural Georgia [39]. Over 15 months, 30 rt-PA treatments occurred at hospitals with little or no prior rt-PA administration rates. A telemedicine service in Ontario reported a 31 % treatment rate [40]. A Maryland experience reported an rt-PA rate of 24 % in telemedicine vs. 4 % in telephone care [41]. Schwamm et al. reported the results of telemedicine-guided rt-PA where treatment was initiated in 6 of 10 patients (60 %) presenting within 3 h of stroke onset and in 6 of 8 (75 %) in whom telestroke consultation was begun within 3 h after onset [42]. Meyer et al. reported a telestroke rt-PA rate of 28 % [43]. In some of these studies, the denominator may not have included all possible stroke admissions, which limits the certainty of rt-PA rate increase. The TEMPiS project (the Telemedic Pilot Project for Integrative Stroke Care) in Bavaria did track this total number of admissions and reported a 10-fold relative increase in rt-PA treatments at telestroke hospitals vs. hospital rates prior to telestroke implementation, increasing from 10 to 115 per year [44]. In many experiences, telestroke services are coupled with stroke education and provider training, features which may have further contributed to increased rt-PA rates. Given the above, a 2009 Scientific Statement reported a Class IIa, Level of Evidence B recommendation for using telestroke in conjunction with education and training for healthcare providers to increase rt-PA rates at community hospitals without access to adequate onsite stroke expertise [6].


rt-PA Decision-Making


Various series have demonstrated the feasibility of telestroke as a means to assist in decision-making for rt-PA administration. One prospective telestroke comparison, the STRoKE DOC trial, assessed the efficacy of telestroke vs. telephone-only technique for the purposes of medical decision-making in acute stroke [43]. The primary outcome measure, assessed via central adjudication, showed correct decision-making in 98 % of telemedicine cases vs. 82 % of telephone cases (p = 0.0009). This primary outcome result was also noted for the rt-PA subgroup comparisons (p = 0.047). Further analyses showed a telemedicine sensitivity of 100 %, specificity of 98 %, likelihood ratio of 41 and number needed to assess of 6 for telestroke [45]. STRokE DOC-AZ was a subsequent feasibility trial that used the same protocol, but in a different cohort of patients in Arizona and showed similar results [46]. A pooled analysis of these two studies showed correct decision-making in 96 % vs. 83 % of cases (p = 0.002) [47]. A 2009 Scientific Statement reported a Class I, Level of Evidence B recommendation for using telestroke to provide a medical opinion in favor of or against the use of rt-PA in patients with suspected acute ischemic stroke when on-site stroke expertise is not immediately available [6].


Patient Outcomes


Trials to date have demonstrated that acute telestroke consultations are feasible, can increase rates of IV rt-PA administration and show superior efficacy over telephone modalities for acute medical decision-making in stroke. However, the impact of telestroke on clinical outcomes remains to be conclusively demonstrated. Safety and efficacy outcomes assessed usually include ICH rates and functional outcome scale scores.

In the STRokE DOC trial, 90-day functional outcomes in the telephone versus telemedicine groups respectively were similar for the BI (95–100) (43 % vs. 54 %; p = 0.13) and mRS(0-1) (34 % vs. 47 %; p = 0.09) [43]. There was no difference in rate of ICH (7 % vs. 8 %; p = 1.0) or overall mortality (19 % versus 13 %; p = 0.27). Similar results were noted for the rt-PA subgroup. In this subgroup, when adjusting for baseline NIHSS differences, mortality was not found to be statistically significant (39 % vs. 12 %; p = 0.17). The TEMPiS study was a cluster-control study of 132 rt-PA-treated patients at stroke centers compared with 170 rt-PA-treated patients at telemedicine-enabled spoke hospitals [38]. This trial reported similar outcomes for both groups, with equivalent 6 month mortality (14.2 % vs. 13 %; p = 0.45), 6 month favorable mRS (39.5 % vs. 30.9 %; p = 0.10) and 6 month BI (47.1 % vs. 44.8 %; p = 0.44) scores for the stroke centers versus the telemedicine-enabled hospitals, respectively. A Finnish telestroke experience of 61 patients treated with rt-PA via telestroke was compared with control patients treated with rt-PA in-person from the Helsinki Stroke Thrombolysis Registry. This demonstrated similar symptomatic ICH rates (6.7 % vs. 9.4 %; p = 0.427), 3-month mortality (11.5 % vs. 10.2 %; p = 0.662), 3-month mRS of 0-2 (49.1 % vs. 58.1 %; p = 0.214), and 3-month mRS of 0-1 (29.4 % vs. 36.8 %; p = 0.289) for the telestroke vs. in-person treatments, respectively [48]. A 2009 Scientific Statement reported a Class I, Level of Evidence B recommendation for using telestroke to provide a medical opinion in favor of or against the use of rt-PA in patients with suspected acute ischemic stroke when on-site stroke expertise is not immediately available [6]. This same report concluded that the safety and efficacy of telephone guided rt-PA administration without CT interpretation via teleradiology is not well established (Class IIb, Level of Evidence C) [6].


Inpatient Care and Stroke Units


It is well-established that dedicated stroke units improve the outcome of patients who are hospitalized for acute stroke [49]. The Joint Commission has established requirements for hospitals to attain designations of Primary and Comprehensive Stroke Centers. Many hospitals, especially those in the community or rural setting, do not have ready access to the resources requires for this or similar designations. This includes availability of vascular neurologists, neurosurgeons, and radiologists, therapists with stroke expertise, stroke nurses, and advanced neuroimaging. Such expertise not only facilitates hyperacute management (as described above), but also the management of complications, evaluation and identification of stroke etiology, and initiation of appropriate secondary stroke prophylaxis.

The TEMPiS study reviewed the experience of two comprehensive stroke centers that partnered with 12 regional hospitals without a stroke unit [50]. The telestroke program provided the neurological expertise for new stroke units at the regional hospitals and the study included stroke education programs and additional resource support for therapists at the sites. Patients in telestroke network hospitals had a 38 % lower odds ratio of a poor outcome defined as severe disability, institutional care, or death [51]. A 2009 Scientific Statement reported a Class I, Level of Evidence B recommendation for using telestroke when the lack of local stroke expertise is the only barrier to the implementation of inpatient stroke units [6].


Rehabilitation


Stroke rehabilitation, especially inpatient rehabilitation, constitutes a large portion of the cost attributed to stroke care [52]. Socioeconomic disparities frequently lead to inadequate rehabilitation measures after acute care hospitalization has been completed. Insufficient rehabilitation may contribute further to costs of stroke care in the form of lost earnings, especially in younger stroke patients.

Multiple studies have demonstrated the feasibility of telerehabilitation. This approach involves periodic audio-video sessions to facilitate rehabilitation therapies for various functional modalities, including balance training, motor tasks, range-of-motion tasks, and speech/language therapy. Data that tracks patient progress can be relayed to rehabilitation specialists who can remotely monitor improvement. Traditional scales that measure function can be applied to this information, such as the Fugl-Meyer and Ashworth scales. In several of these series, telerehabilitation has been enhanced with sensors that can detect 3D motion, goniometry, and actigraphy.

Few studies have been done on the efficacy of occupational therapy (OT) assessments via telemedicine; while these were feasible, they were hampered by inadequate video technologies [53] and method related issues [54]. Studies using telemedicine in physical therapy have often centered around interventions involving the completion of tasks in simulated environments [55]. Disability scales can be administered via telemedicine as well, and several small studies have demonstrated reasonable inter-rater reliability in the administration of these scales when done in-person and remotely. HQ-VTC was used in administering the European Stroke Scale (ESS) and the Functional Reach Test (FRT) to 26 stroke patients randomized to be evaluated either in-person or remotely [56]. When directed by the face-to-face therapist, the 2 PTs reported equivalent values in >90 % of patients for the FRT and the components of the ESS (excepting gait and leg position); when directed by the remote PT, both PTs reported equivalent values in >90 % of patients for the FRT as well, but >83 % for all ESS components. There was, however, no statistically significant difference between face-to-face and remote assessments.

In the realm of stroke-related speech therapy (ST), Brennan et al. used HQ-VTC to use the story retelling procedure comparing remote assessment to traditional face-to-face speech evaluation, and found no difference between the ratings acquired [57]. Palsbo used a randomized, double-crossover study to compare remote or in-person administration of parts of the Boston Diagnostic Aphasia Examination as well as speech comprehension, speech expression, and motor speech [58]. Patients were scored in a blinded fashion both in-person and remote, and agreement within the 95 % limits ranged from 92 to 100 % for each measure.

These series suggest that outcome measures during telerehabilitation are similar to traditional rehabilitative approaches, and in some cases motor performance was better in the telerehabilitation groups. While a telemedicine approach to stroke rehabilitation may need to be validated in larger studies, the method shows promise in minimizing outpatient rehabilitation costs while providing therapy that appears to have similar efficacy as traditional therapy. A 2009 Scientific Statement reported a Class I, Level of Evidence B recommendation for the assessment of occupational, physical, or speech disability in stroke patients by allied health professionals via telemedicine systems using specific standardized assessments [6].

A subsequent systematic review of the published literature on post-stroke care telemedicine was performed [59]. Out of 1,405 potential publications, 24 unique manuscripts underwent functionality, application, technology, and evaluative scoring. Results showed that most post-stroke telemedicine rehabilitation studies evaluated only adult populations, were small in sample size, and still preliminary in scope.


Telestroke for Clinical Research


While rt-PA is an available and efficacious therapy for acute stroke, its administration is extremely time-sensitive, and despite its underutilization multiple exclusion criteria further limit its use. Other treatment modalities, including endovascular based intra-arterial treatments, have only limited evidence to support their use. Many stroke neuroprotectant studies have failed to show clinical benefit [60]. Randomized, controlled trials are required to advance the field of stroke therapeutics, but clinical trial enrollment for large stroke studies tends to be very slow. A meta-analysis of large stroke trials found that the enrollment rate was 0.79 subjects per center per month; in North America, that rate was 0.57 subjects per center per month [61].

While selection criteria need to be optimized to assure enrollment of informative patients, increasing the pool of eligible patients via telemedicine may provide a significant boost to trial completion as well. The majority of patient recruitment for such trials tends to occur in large stroke centers, minimizing enrollment in community and rural hospitals. The use of high-quality video teleconferencing has the potential to enlarge the recruitment pool, and thereby expedite clinical trial enrollment. A center in Georgia was able to enhance recruitment into stroke trials by this method; this was limited to screening for eligible patients at the spoke site and then transferring patients to the hub where they were subsequently enrolled and administered the study treatment [62]. Future studies might include spoke sites that have sufficient resources to have patients enrolled into trials on-site with telestroke assistance, and follow-up research visits could be conducted remotely as well. Evidence for telestroke’s use to augment clinical trial enrollments is currently limited. A 2009 Scientific Statement reported a Class IIa, Level of Evidence B recommendation only for using telephone-based contact between emergency medical personnel and stroke specialists for prehospital clinical trial screening and consent for enrollment into hyperacute neuroprotective trials [6].


Telestroke Education


One of the major barriers to better rates of rt-PA use is practitioner unfamiliarity and concern about adverse effects, especially intracranial hemorrhage. While simultaneously providing consultation on acute stroke cases, telestroke allows for education of caregivers on how to time-manage acute stroke evaluation, select patients appropriately, and administer rt-PA. As stroke specialists have a tendency to be located in tertiary health care centers with established stroke systems of care, telestroke provides experience in stroke management in settings where local resources may be limited. Trainees in neurology and vascular neurology further benefit from this same type of exposure. While digital media is available for NIHSS training [63], telehealth modalities may allow for more interactive training sessions, including mock stroke codes. Further, telemedicine discussion with community caregivers as well as the lay public may be a useful mechanism to disseminate knowledge on how to identify stroke symptoms, as well as the importance of early activation of the emergency response system.


Evaluation Times


Years ago, telestroke consultations may have been cumbersome and required significant time to obtain patient consent, set up camera systems, drive to central base stations, and perform examinations and radiographic assessments. Over the years, each of these time requirements has undergone improvements resulting in much more straightforward and efficient care models. For example, obtaining consent for telemedicine in some states has moved from requirement of faxing signed documentations to now obtaining verbal consent. Camera systems are now easily controlled remotely by the consulting telestrokologist, requiring little if any effort by the spoke originating site (where the patient resides). With the development of site-independent, web based, Internet software solutions, providers no longer need to drive to central base stations for point-to-point connectivity. Finally, technology improvements in camera resolution, features such as remote pan-tilt-zoo and on-screen decision support tools have all enabled a higher degree of confidence and improved usability for the telestroke consultant.

In general, telestroke consultations do not take long to perform. The STRokE DOC trial reported average duration of 32 min, though this was notably in the setting of a standardized clinical trial with research investigators [43]. A recent time performance analysis from telestroke software logs from 8 hubs, 24 spokes and 14 physician users (accounting for 203 telestroke encounters) reported a mean consult length of 14.5 min and mean response time to logging into the camera of 76.3 min [64]. Mean consult length was longer in cases where rt-PA was recommended (20.0 vs. 153. Minutes; p = 0.04). Though conclusions from this report are significantly limited by only reporting data based on time actually logged into the camera system, it does highlight that telestroke consult systems are easy to use and can facilitate rapid stroke evaluations. The long response time prior to logging into the system in real practice may be artifactual, as providers may initially use telephone to gather basic information, and subsequently use the telestroke system to perform other critical exam and radiologic assessments. This further highlights the evolving and intriguing use patterns in the high pressure/real time setting of telestroke.


Telestroke Technology and Equipment


The technology related to telestroke has evolved over time. Initial technology was bulky, centrally based, and had limited functionality. Technology has evolved and now enables robust video, audio, and data transmissions to mobile consultants via Internet based and highly compressed data transmission.

Use of reliable hardware and software is crucial when performing telestroke consultations. Technical difficulties may be a major barrier to telestroke implementation in the setting of device or program failure, and technical maintenance is important in avoiding mishaps during hyperacute stroke evaluations. Because high-quality video teleconferencing is integral to the telestroke patient encounter, high bandwidth is crucial to maintaining connectivity. This must be done securely for confidentiality purposes.


Hardware and Software


Telestroke technologies, falling into the category of hardware, generally take the form of hub-side telestroke consultant workstations (such as video codec devices coupled with high resolution cameras) and spoke-side camera systems (such as telemedicine workstations, static carts, or even mobile robot systems). There is no shortage of telemedicine vendors in the healthcare arena today. Systems are usually differentiated by price, support options and features (such as mobility of a robot vs. static cart, or availability of on-screen decision support tools vs. more simple videoconferencing abilities only). It is important to note that not all videoconferencing systems are telestroke systems. Telestroke systems often include high quality of service (QoS) algorithms to mitigate against packet loss and signal errors as well as availability of medical expertise and decision support tools. A general rule of thumb is to know what features and capabilities are needed for a successful workflow before the equipment decision is made. For example, in settings where time is not critical for making a diagnosis, systems that can be accessed at a moment’s notice from mobile locations located throughout the globe may not be imperative. For stroke, where minutes matter, site independent systems are becoming the rule rather than the exception.

Software systems have been now been developed to enable access to telemedicine systems more easily. Today, it is not uncommon to find providers utilizing any number of software solutions for remote camera access from desktop computers, mobile laptops, tablet devices, and even smartphones.


Connectivity and Networks


While there is consensus expert opinion on the minimum quality standards for HQ-VTC use [6], there is no compelling data on the minimum or optimum requirements or settings for HQ-VTC. It is recommended that “transmission rates support >20 frames per second of bidirectional synchronized audio and video, at a resolution capable of being accurately being displayed on monitors of > = 13 in.” Common intermediate format (CIF) and Source Input Format (SIF) are video formats used to standardize visual resolution; the ability to achieve these typically depends on the bandwidth available to transmit data, thus making sufficient bandwidth imperative to conveying high-quality video.

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Apr 21, 2017 | Posted by in NEUROLOGY | Comments Off on Telestroke: Delivery and Design

Full access? Get Clinical Tree

Get Clinical Tree app for offline access