Telestroke Triage Comes of Age

When Dr. Steven R. Levine and Dr. Mark Gorman first proposed that emergency physicians in small rural hospitals link up via bedside video cameras to stroke specialists in large urban hospitals to triage stroke patients, they named the concept "telestroke."

"I never thought it would catch on," Dr. Levine said in a recent interview.

Courtesy SUNY Downstate Medical Center

It was 1999, and while thrombolysis with tissue plasminogen activator (TPA) had by then been approved for 3 years, only about 1% of acute ischemic stroke patients in the United States were receiving it, when as many as half could potentially have been eligible.

One important reason for the underuse of TPA was that smaller and more rural hospitals did not have neurologists available to respond to stroke calls. Patients would have to be transported to larger centers to receive intravenous TPA, and by the time they arrived, many would have missed the time window – which published guidelines have now determined to be up to 4.5 hours after a stroke – when TPA can be safely administered.

Dr. Levine, currently professor of neurology and emergency medicine at SUNY Downstate Medical Center in Brooklyn, N.Y., and a leading researcher and proponent of telestroke, envisioned a spoke-and-hub model of care in which smaller hospitals would be linked by video to larger ones with neurologists available, in active shifts, for two-way consults (Stroke 1999;30:464-9).

TPA could be initiated in the emergency departments of the spokes, and higher-risk patients would be transferred to hub hospitals with dedicated stroke centers once a drip was started.

Early adopters included Massachusetts General Hospital, which began building on the spoke-and-hub model and proving the concept in study after study. "We were the first to demonstrate that the NIH stroke scale could be done with equivalent reliability over telemedicine, and to show that a neurologist could interpret information on a laptop computer as effectively as a radiologist could on expensive imaging systems," said Dr. Lee H. Schwamm, professor of neurology at Harvard University and director of the Partners TeleStroke center, a 30-hospital network run out of Massachusetts General and Brigham and Women’s Hospital, both in Boston.

In 2009, on the strength of evidence showing that telestroke programs increased TPA administration rates, Dr. Schwamm and colleagues, including Dr. Levine, drafted a policy statement for the American Heart Association (Stroke 2009;40:2635-60) advocating the creation of telestroke networks wherever resources were available. More recent evidence has shown telestroke to be cost-effective compared with conventional care (Neurology 2011;77:1590-8; Stroke 2012;43:A3077).

Currently there are at least 27 networks up and running in the United States and Canada, along with 14 in Europe, following spoke-and-hub models. Some comprise a handful of hospitals; others, dozens. Many have existed for only a few years.

North American telestroke programs can now boast TPA rates between 10% and 20% of patients admitted for stroke, even in very rural networks, when the U.S. national average is estimated to be about 5%. In Europe, established programs are reporting TPA rates between 12% and 17%, according to a recent meta-analysis (Curr. Opin. Neurol. 2012;25:5-10).

Dr. Charles H. Tegeler, professor of neurology and head of the telestroke program at Wake Forest Baptist Medical Center in Winston-Salem, N.C., said in an interview that his program, which started in 2009 and now incorporates nine network hospitals, has seen TPA administered in 27% of patients who receive a video stroke consultation. And many patients are receiving TPA in hospitals that, before joining the network, did not administer it.

About 45% of patients receiving a telestroke consultation through one of the network hospitals are transferred to Wake Forest Baptist, or another dedicated stroke center of the patient’s choice, for additional evaluation and management. Depending on local resources and expertise, those treated with TPA may be transferred under a treatment model known as "drip and ship." The rest remain at the network hospital for ongoing care, or "drip and keep."

Rayetta Johnson, R.N., stroke program manager for Wake Forest Baptist, said in an interview that an important factor in its success has been achieving close cooperation, through training seminars and other forms of outreach, with hospital emergency department staff at all levels. The program also reached out extensively to local emergency medical services accustomed to rerouting stroke patients to larger hospitals. "Having EMS on board is extremely important," Ms. Johnson said.

Barriers to Adoption

If telestroke programs are cost effective and improve TPA rates dramatically, why aren’t they more widespread?

Part of the problem, experts say, lies with neurologists’ hesitance to join telestroke networks because reimbursement for telestroke is less straightforward than for a traditional stroke call, and many consider it to be inadequate under current Medicare and Medicaid guidelines.

Dr. Ramesh Madhavan, director of telemedicine for Wayne State University, Detroit, and the Michigan Stroke Network, one of the largest telestroke programs in the country with 36 affiliated hospitals, said in an interview that his program’s eight neurologists – including himself – take stroke calls over 12-hour shifts.

Courtesy Dr. Ramesh Madhavan and Dr. Kumar Rajamani

"We take shifts because there is no direct reimbursement, and we have to do other things during the course of that time. We have to multitask," Dr. Madhavan said, adding that this can make some neurologists feel burned out.

Dr. Tegeler said that, by contrast, the five vascular neurologists in Wake Forest Baptist’s smaller program have incorporated the telestroke coverage as part of their regular stroke attending call duties. Now, neurologists at some of the network hospitals "may not have to disrupt their office schedule for an hour or more to go over to the hospital to see an acute stroke patient," he said.

The presence of telestroke coverage also was used by one network hospital to help recruit a neurologist to a community where there had not been one, since the local neurologist would "not have to take stroke calls 24-7," Dr. Tegeler said.

In addition to the reimbursement issue, Dr. Levine said that telestroke has been hampered by "fears that the technology won’t work or may break down, or of lawsuits, and also doctors’ fear of something more technologically advanced than some are used to doing." Unlike their counterparts in emergency departments, Dr. Levine said, "neurologists aren’t techies, as a rule." However, telestroke experts are increasingly exploring cheaper and more portable options for videoconferencing, which could prove less imposing in terms of both startup costs to networks and in ease of use.

Most telestroke programs currently employ videoconferencing technology using equipment mounted on a mobile cart or purpose-built robots that can be driven to a patient’s bedside.

The robots, Dr. Schwamm said, "are a luxury. You can put a laptop with a specialized camera on top of a cart and roll it to a patient and get everything you need. Technology should not be where the costs are."

One team of researchers recently demonstrated that iPhones could be used in telestroke networks for patient assessment (J. Stroke Cerebrovasc. Dis. 2011 Oct. 24 [doi:10.1016/j.jstrokecerebrovasdis.2011.09.013]).

Dr. Madhavan said that he has been working with several technology companies on applications that would allow more telestroke consults to be carried out on portable devices using 3G or 4G networks. A migration to mobile platforms might help improve door-to-needle times in addition to reducing costs, he said.

Research Priorities

When Dr. Levine and Dr. Gorman first proposed telestroke networks in 1999, they envisioned them not solely as a way to increase TPA uptake but also as a way to facilitate patient entry into clinical trials of new stroke treatments.

This has yet to happen directly, Dr. Levine noted, but it may soon. "Community hospitals are not used to dealing with experimental medicines, or the logistics of randomization, and working with trial coordinators, IRBs, and rigorous and extensive data collection. But now that the systems are being built and there’s some infrastructure over the last 10 years, hospitals will hopefully start to see what clinical trials can bring them in terms of recruitment, state-of-the-art care while testing the most novel and promising treatments, and financially," he said.

Dr. Schwamm, whose program is among the few actively seeking to design clinical trials using telestroke, said that enrolling patients through telemedicine – and obtaining consent through video – is a challenge that, if met, will help broaden trial populations to reach more people living in rural and resource-poor areas. "The big studies tend to be done at the teaching hospitals, in urban areas. Telestroke can help allow people in the community to be enrolled," he said, leading to better representation of the population at large.

More telestroke programs are collecting at least short-term outcome or discharge data, measured in National Institutes of Health Stroke Scale (NIHSS) scores at admission and discharge. Dr. Madhavan’s team recently reported that, over 4 years, patients receiving intravenous TPA through the Michigan Stroke Network experienced a greater than seven-point reduction in NIHSS score by the end of their hospital stay (Stroke 2012;43:A2991).

Wake Forest Baptist tracks as a surrogate measurement whether patients are discharged to home, rehabilitation, or skilled nursing – and 79% are discharged to home or rehab, Dr. Tegeler said.

A few programs are collecting data on 90-day outcomes, an important measurement in clinical trials. For now, though, outcomes are mainly being analyzed with an aim to improve treatment protocols.

Although several studies have demonstrated that outcomes in the "drip and ship" model are comparable with those of patients admitted and treated at dedicated stroke centers, reducing average door-to-needle times and deciding who benefits from "drip and ship" versus "drip and keep" remain key areas of interest.

At the International Stroke Conference in February, neurologist Dr. Shadi Yaghi presented research on a cohort of 562 patients in the Arkansas SAVES program, a telestroke network that links 22 hospitals to stroke specialists at the University of Arkansas Medical Sciences Center in Little Rock. Patients with a worse outcome at 90 days were more likely to have received postthrombolysis care at a spoke hospital (odds ratio, 15.63; P = .019; 95% CI 1.56-166.67), Dr. Yaghi and his colleagues found (Stroke 2012;43:A2885).

"Our study showed that patients with moderate to severe strokes had better outcomes when transferred to a primary stroke center, likely due to more available resources and treatment options, such as neurointerventional treatment and neurocritical care treatment. Those with mild symptoms did not benefit from the transfer," Dr. Yaghi said in an e-mail interview.

Beyond TPA

In Europe and, in particular, Germany, telestroke has developed along a different course, with an emphasis on building multidisciplinary telestroke units in network hospitals.

Instead of concentrating solely on identifying candidates for TPA, network hospitals have teams of nurses, physiotherapists, occupational therapists, speech and language therapists, and physicians trained through seminars held at stroke centers and grand rounds in the affiliated hospitals. Workup of stroke patients is standardized, and treatment includes TPA for qualified patients under a drip-and-keep model.

Neurologist Dr. Peter Müller-Barna of Klinikum Harlaching, Stadtisches Klinikum in Munich, who is project coordinator of the TEMPiS (Telemedic Pilot Project for Integrative Stroke Care) stroke network in Bavaria, said in an interview that dedicated stroke units have proven "very efficient in reducing mortality and dependency in all stroke patients, with a number needed to treat (NNT) of about 5. In comparison to this, intravenous TPA is less effective; the NNT is about 10. Taken together, building up telestroke units is more expensive, but also by far more effective."

The TEMPiS network is one of the most intensively studied telestroke projects in terms of short- and long-term outcomes (Lancet Neurol. 2006;5:742-8; Neurology 2007;69:898-903). Recently, Dr. Müller-Barna and colleagues showed quality of care in TEMPiS network hospitals to be comparable with that seen at conventional stroke units with neurologists (Stroke 2012;43:A2820).

"In my eyes, the main obstacle for not establishing more telestroke programs in the U.S. is the lack of evidence for the telethrombolysis network approach in terms of improved clinical outcome, in contrast to the integrative stroke network approach with its telestroke units," Dr. Müller-Barna said.

Dr. Schwamm, who has collaborated with Dr. Müller-Barna’s group for research, said that much of the difference in the European and American telestroke models can be attributed to the differences in health care systems overall, with European systems better set up to invest in telestroke networks, keep patients in hospitals, and track long-term outcomes.

"In the U.S. the primary driver is the need to provide acute neurological assessment and make sure TPA-eligible patients have a chance to be treated," Dr. Schwamm said. "But telestroke still has lots of collateral benefits. It’s not only drip and keep and drip and ship but evaluate and triage. There’s a lot of that that goes on in telestroke networks even though that’s not the primary driver."

Courtesy Wake Forest Baptist Medical Center

Dr. Tegeler said that the Wake Forest Baptist program is increasingly looking at telestroke consultations as a pathway to other types of interventions.

"For patients who may have missed the window for TPA, telestroke can help facilitate consideration of eligibility for interventional or intraarterial procedures – there’s a whole host of things you could do with the right patients with the right lesions after the 4.5-hour window."

Dr. Tegeler and colleagues at Wake Forest Baptist have successfully pilot tested the use of telestroke technology for postdischarge follow-up visits, allowing patients who were transferred to the medical center for stroke treatment to be followed up, via videoconferencing, at spoke hospitals closer to their homes.

"There are opportunities to use it in all kinds of ways," he said.

Dr. Levine said that he has served on an advisory board and as a consultant for Genentech, which makes TPA (honorarium donated to stroke research), and has received reimbursement from Genentech for travel to a scientific meeting. He also has served as an expert witness in medical-legal cases involving stroke. Dr. Tegeler disclosed serving on the speakers bureau for Genentech. Dr. Schwamm reported being an advisory board member for lifeIMAGE, a company working on technology for the sharing of brain images in telemedicine. Dr. Madhavan said that he is a consultant for Robolytics, Process Proxy Corp., Great Lakes NeuroTechnologies, and MatrixView Ltd. No other sources had any relevant disclosures.

When Dr. Steven R. Levine and Dr. Mark Gorman first proposed that emergency physicians in small rural hospitals link up via bedside video cameras to stroke specialists in large urban hospitals to triage stroke patients, they named the concept "telestroke."

"I never thought it would catch on," Dr. Levine said in a recent interview.

Courtesy SUNY Downstate Medical Center

It was 1999, and while thrombolysis with tissue plasminogen activator (TPA) had by then been approved for 3 years, only about 1% of acute ischemic stroke patients in the United States were receiving it, when as many as half could potentially have been eligible.

One important reason for the underuse of TPA was that smaller and more rural hospitals did not have neurologists available to respond to stroke calls. Patients would have to be transported to larger centers to receive intravenous TPA, and by the time they arrived, many would have missed the time window – which published guidelines have now determined to be up to 4.5 hours after a stroke – when TPA can be safely administered.

Dr. Levine, currently professor of neurology and emergency medicine at SUNY Downstate Medical Center in Brooklyn, N.Y., and a leading researcher and proponent of telestroke, envisioned a spoke-and-hub model of care in which smaller hospitals would be linked by video to larger ones with neurologists available, in active shifts, for two-way consults (Stroke 1999;30:464-9).

TPA could be initiated in the emergency departments of the spokes, and higher-risk patients would be transferred to hub hospitals with dedicated stroke centers once a drip was started.

Early adopters included Massachusetts General Hospital, which began building on the spoke-and-hub model and proving the concept in study after study. "We were the first to demonstrate that the NIH stroke scale could be done with equivalent reliability over telemedicine, and to show that a neurologist could interpret information on a laptop computer as effectively as a radiologist could on expensive imaging systems," said Dr. Lee H. Schwamm, professor of neurology at Harvard University and director of the Partners TeleStroke center, a 30-hospital network run out of Massachusetts General and Brigham and Women’s Hospital, both in Boston.

In 2009, on the strength of evidence showing that telestroke programs increased TPA administration rates, Dr. Schwamm and colleagues, including Dr. Levine, drafted a policy statement for the American Heart Association (Stroke 2009;40:2635-60) advocating the creation of telestroke networks wherever resources were available. More recent evidence has shown telestroke to be cost-effective compared with conventional care (Neurology 2011;77:1590-8; Stroke 2012;43:A3077).

Currently there are at least 27 networks up and running in the United States and Canada, along with 14 in Europe, following spoke-and-hub models. Some comprise a handful of hospitals; others, dozens. Many have existed for only a few years.

North American telestroke programs can now boast TPA rates between 10% and 20% of patients admitted for stroke, even in very rural networks, when the U.S. national average is estimated to be about 5%. In Europe, established programs are reporting TPA rates between 12% and 17%, according to a recent meta-analysis (Curr. Opin. Neurol. 2012;25:5-10).

Dr. Charles H. Tegeler, professor of neurology and head of the telestroke program at Wake Forest Baptist Medical Center in Winston-Salem, N.C., said in an interview that his program, which started in 2009 and now incorporates nine network hospitals, has seen TPA administered in 27% of patients who receive a video stroke consultation. And many patients are receiving TPA in hospitals that, before joining the network, did not administer it.

About 45% of patients receiving a telestroke consultation through one of the network hospitals are transferred to Wake Forest Baptist, or another dedicated stroke center of the patient’s choice, for additional evaluation and management. Depending on local resources and expertise, those treated with TPA may be transferred under a treatment model known as "drip and ship." The rest remain at the network hospital for ongoing care, or "drip and keep."

Rayetta Johnson, R.N., stroke program manager for Wake Forest Baptist, said in an interview that an important factor in its success has been achieving close cooperation, through training seminars and other forms of outreach, with hospital emergency department staff at all levels. The program also reached out extensively to local emergency medical services accustomed to rerouting stroke patients to larger hospitals. "Having EMS on board is extremely important," Ms. Johnson said.

Barriers to Adoption

If telestroke programs are cost effective and improve TPA rates dramatically, why aren’t they more widespread?

Part of the problem, experts say, lies with neurologists’ hesitance to join telestroke networks because reimbursement for telestroke is less straightforward than for a traditional stroke call, and many consider it to be inadequate under current Medicare and Medicaid guidelines.

Dr. Ramesh Madhavan, director of telemedicine for Wayne State University, Detroit, and the Michigan Stroke Network, one of the largest telestroke programs in the country with 36 affiliated hospitals, said in an interview that his program’s eight neurologists – including himself – take stroke calls over 12-hour shifts.

Courtesy Dr. Ramesh Madhavan and Dr. Kumar Rajamani

"We take shifts because there is no direct reimbursement, and we have to do other things during the course of that time. We have to multitask," Dr. Madhavan said, adding that this can make some neurologists feel burned out.

Dr. Tegeler said that, by contrast, the five vascular neurologists in Wake Forest Baptist’s smaller program have incorporated the telestroke coverage as part of their regular stroke attending call duties. Now, neurologists at some of the network hospitals "may not have to disrupt their office schedule for an hour or more to go over to the hospital to see an acute stroke patient," he said.

The presence of telestroke coverage also was used by one network hospital to help recruit a neurologist to a community where there had not been one, since the local neurologist would "not have to take stroke calls 24-7," Dr. Tegeler said.

In addition to the reimbursement issue, Dr. Levine said that telestroke has been hampered by "fears that the technology won’t work or may break down, or of lawsuits, and also doctors’ fear of something more technologically advanced than some are used to doing." Unlike their counterparts in emergency departments, Dr. Levine said, "neurologists aren’t techies, as a rule." However, telestroke experts are increasingly exploring cheaper and more portable options for videoconferencing, which could prove less imposing in terms of both startup costs to networks and in ease of use.

Most telestroke programs currently employ videoconferencing technology using equipment mounted on a mobile cart or purpose-built robots that can be driven to a patient’s bedside.

The robots, Dr. Schwamm said, "are a luxury. You can put a laptop with a specialized camera on top of a cart and roll it to a patient and get everything you need. Technology should not be where the costs are."

One team of researchers recently demonstrated that iPhones could be used in telestroke networks for patient assessment (J. Stroke Cerebrovasc. Dis. 2011 Oct. 24 [doi:10.1016/j.jstrokecerebrovasdis.2011.09.013]).

Dr. Madhavan said that he has been working with several technology companies on applications that would allow more telestroke consults to be carried out on portable devices using 3G or 4G networks. A migration to mobile platforms might help improve door-to-needle times in addition to reducing costs, he said.

Research Priorities

When Dr. Levine and Dr. Gorman first proposed telestroke networks in 1999, they envisioned them not solely as a way to increase TPA uptake but also as a way to facilitate patient entry into clinical trials of new stroke treatments.

This has yet to happen directly, Dr. Levine noted, but it may soon. "Community hospitals are not used to dealing with experimental medicines, or the logistics of randomization, and working with trial coordinators, IRBs, and rigorous and extensive data collection. But now that the systems are being built and there’s some infrastructure over the last 10 years, hospitals will hopefully start to see what clinical trials can bring them in terms of recruitment, state-of-the-art care while testing the most novel and promising treatments, and financially," he said.

Dr. Schwamm, whose program is among the few actively seeking to design clinical trials using telestroke, said that enrolling patients through telemedicine – and obtaining consent through video – is a challenge that, if met, will help broaden trial populations to reach more people living in rural and resource-poor areas. "The big studies tend to be done at the teaching hospitals, in urban areas. Telestroke can help allow people in the community to be enrolled," he said, leading to better representation of the population at large.

More telestroke programs are collecting at least short-term outcome or discharge data, measured in National Institutes of Health Stroke Scale (NIHSS) scores at admission and discharge. Dr. Madhavan’s team recently reported that, over 4 years, patients receiving intravenous TPA through the Michigan Stroke Network experienced a greater than seven-point reduction in NIHSS score by the end of their hospital stay (Stroke 2012;43:A2991).

Wake Forest Baptist tracks as a surrogate measurement whether patients are discharged to home, rehabilitation, or skilled nursing – and 79% are discharged to home or rehab, Dr. Tegeler said.

A few programs are collecting data on 90-day outcomes, an important measurement in clinical trials. For now, though, outcomes are mainly being analyzed with an aim to improve treatment protocols.

Although several studies have demonstrated that outcomes in the "drip and ship" model are comparable with those of patients admitted and treated at dedicated stroke centers, reducing average door-to-needle times and deciding who benefits from "drip and ship" versus "drip and keep" remain key areas of interest.

At the International Stroke Conference in February, neurologist Dr. Shadi Yaghi presented research on a cohort of 562 patients in the Arkansas SAVES program, a telestroke network that links 22 hospitals to stroke specialists at the University of Arkansas Medical Sciences Center in Little Rock. Patients with a worse outcome at 90 days were more likely to have received postthrombolysis care at a spoke hospital (odds ratio, 15.63; P = .019; 95% CI 1.56-166.67), Dr. Yaghi and his colleagues found (Stroke 2012;43:A2885).

"Our study showed that patients with moderate to severe strokes had better outcomes when transferred to a primary stroke center, likely due to more available resources and treatment options, such as neurointerventional treatment and neurocritical care treatment. Those with mild symptoms did not benefit from the transfer," Dr. Yaghi said in an e-mail interview.

Beyond TPA

In Europe and, in particular, Germany, telestroke has developed along a different course, with an emphasis on building multidisciplinary telestroke units in network hospitals.

Instead of concentrating solely on identifying candidates for TPA, network hospitals have teams of nurses, physiotherapists, occupational therapists, speech and language therapists, and physicians trained through seminars held at stroke centers and grand rounds in the affiliated hospitals. Workup of stroke patients is standardized, and treatment includes TPA for qualified patients under a drip-and-keep model.

Neurologist Dr. Peter Müller-Barna of Klinikum Harlaching, Stadtisches Klinikum in Munich, who is project coordinator of the TEMPiS (Telemedic Pilot Project for Integrative Stroke Care) stroke network in Bavaria, said in an interview that dedicated stroke units have proven "very efficient in reducing mortality and dependency in all stroke patients, with a number needed to treat (NNT) of about 5. In comparison to this, intravenous TPA is less effective; the NNT is about 10. Taken together, building up telestroke units is more expensive, but also by far more effective."

The TEMPiS network is one of the most intensively studied telestroke projects in terms of short- and long-term outcomes (Lancet Neurol. 2006;5:742-8; Neurology 2007;69:898-903). Recently, Dr. Müller-Barna and colleagues showed quality of care in TEMPiS network hospitals to be comparable with that seen at conventional stroke units with neurologists (Stroke 2012;43:A2820).

"In my eyes, the main obstacle for not establishing more telestroke programs in the U.S. is the lack of evidence for the telethrombolysis network approach in terms of improved clinical outcome, in contrast to the integrative stroke network approach with its telestroke units," Dr. Müller-Barna said.

Dr. Schwamm, who has collaborated with Dr. Müller-Barna’s group for research, said that much of the difference in the European and American telestroke models can be attributed to the differences in health care systems overall, with European systems better set up to invest in telestroke networks, keep patients in hospitals, and track long-term outcomes.

"In the U.S. the primary driver is the need to provide acute neurological assessment and make sure TPA-eligible patients have a chance to be treated," Dr. Schwamm said. "But telestroke still has lots of collateral benefits. It’s not only drip and keep and drip and ship but evaluate and triage. There’s a lot of that that goes on in telestroke networks even though that’s not the primary driver."

Courtesy Wake Forest Baptist Medical Center

Dr. Tegeler said that the Wake Forest Baptist program is increasingly looking at telestroke consultations as a pathway to other types of interventions.

"For patients who may have missed the window for TPA, telestroke can help facilitate consideration of eligibility for interventional or intraarterial procedures – there’s a whole host of things you could do with the right patients with the right lesions after the 4.5-hour window."

Dr. Tegeler and colleagues at Wake Forest Baptist have successfully pilot tested the use of telestroke technology for postdischarge follow-up visits, allowing patients who were transferred to the medical center for stroke treatment to be followed up, via videoconferencing, at spoke hospitals closer to their homes.

"There are opportunities to use it in all kinds of ways," he said.

Dr. Levine said that he has served on an advisory board and as a consultant for Genentech, which makes TPA (honorarium donated to stroke research), and has received reimbursement from Genentech for travel to a scientific meeting. He also has served as an expert witness in medical-legal cases involving stroke. Dr. Tegeler disclosed serving on the speakers bureau for Genentech. Dr. Schwamm reported being an advisory board member for lifeIMAGE, a company working on technology for the sharing of brain images in telemedicine. Dr. Madhavan said that he is a consultant for Robolytics, Process Proxy Corp., Great Lakes NeuroTechnologies, and MatrixView Ltd. No other sources had any relevant disclosures.

When Dr. Steven R. Levine and Dr. Mark Gorman first proposed that emergency physicians in small rural hospitals link up via bedside video cameras to stroke specialists in large urban hospitals to triage stroke patients, they named the concept "telestroke."

"I never thought it would catch on," Dr. Levine said in a recent interview.

Courtesy SUNY Downstate Medical Center

It was 1999, and while thrombolysis with tissue plasminogen activator (TPA) had by then been approved for 3 years, only about 1% of acute ischemic stroke patients in the United States were receiving it, when as many as half could potentially have been eligible.

One important reason for the underuse of TPA was that smaller and more rural hospitals did not have neurologists available to respond to stroke calls. Patients would have to be transported to larger centers to receive intravenous TPA, and by the time they arrived, many would have missed the time window – which published guidelines have now determined to be up to 4.5 hours after a stroke – when TPA can be safely administered.

Dr. Levine, currently professor of neurology and emergency medicine at SUNY Downstate Medical Center in Brooklyn, N.Y., and a leading researcher and proponent of telestroke, envisioned a spoke-and-hub model of care in which smaller hospitals would be linked by video to larger ones with neurologists available, in active shifts, for two-way consults (Stroke 1999;30:464-9).

TPA could be initiated in the emergency departments of the spokes, and higher-risk patients would be transferred to hub hospitals with dedicated stroke centers once a drip was started.

Early adopters included Massachusetts General Hospital, which began building on the spoke-and-hub model and proving the concept in study after study. "We were the first to demonstrate that the NIH stroke scale could be done with equivalent reliability over telemedicine, and to show that a neurologist could interpret information on a laptop computer as effectively as a radiologist could on expensive imaging systems," said Dr. Lee H. Schwamm, professor of neurology at Harvard University and director of the Partners TeleStroke center, a 30-hospital network run out of Massachusetts General and Brigham and Women’s Hospital, both in Boston.

In 2009, on the strength of evidence showing that telestroke programs increased TPA administration rates, Dr. Schwamm and colleagues, including Dr. Levine, drafted a policy statement for the American Heart Association (Stroke 2009;40:2635-60) advocating the creation of telestroke networks wherever resources were available. More recent evidence has shown telestroke to be cost-effective compared with conventional care (Neurology 2011;77:1590-8; Stroke 2012;43:A3077).

Currently there are at least 27 networks up and running in the United States and Canada, along with 14 in Europe, following spoke-and-hub models. Some comprise a handful of hospitals; others, dozens. Many have existed for only a few years.

North American telestroke programs can now boast TPA rates between 10% and 20% of patients admitted for stroke, even in very rural networks, when the U.S. national average is estimated to be about 5%. In Europe, established programs are reporting TPA rates between 12% and 17%, according to a recent meta-analysis (Curr. Opin. Neurol. 2012;25:5-10).

Dr. Charles H. Tegeler, professor of neurology and head of the telestroke program at Wake Forest Baptist Medical Center in Winston-Salem, N.C., said in an interview that his program, which started in 2009 and now incorporates nine network hospitals, has seen TPA administered in 27% of patients who receive a video stroke consultation. And many patients are receiving TPA in hospitals that, before joining the network, did not administer it.

About 45% of patients receiving a telestroke consultation through one of the network hospitals are transferred to Wake Forest Baptist, or another dedicated stroke center of the patient’s choice, for additional evaluation and management. Depending on local resources and expertise, those treated with TPA may be transferred under a treatment model known as "drip and ship." The rest remain at the network hospital for ongoing care, or "drip and keep."

Rayetta Johnson, R.N., stroke program manager for Wake Forest Baptist, said in an interview that an important factor in its success has been achieving close cooperation, through training seminars and other forms of outreach, with hospital emergency department staff at all levels. The program also reached out extensively to local emergency medical services accustomed to rerouting stroke patients to larger hospitals. "Having EMS on board is extremely important," Ms. Johnson said.

Barriers to Adoption

If telestroke programs are cost effective and improve TPA rates dramatically, why aren’t they more widespread?

Part of the problem, experts say, lies with neurologists’ hesitance to join telestroke networks because reimbursement for telestroke is less straightforward than for a traditional stroke call, and many consider it to be inadequate under current Medicare and Medicaid guidelines.

Dr. Ramesh Madhavan, director of telemedicine for Wayne State University, Detroit, and the Michigan Stroke Network, one of the largest telestroke programs in the country with 36 affiliated hospitals, said in an interview that his program’s eight neurologists – including himself – take stroke calls over 12-hour shifts.

Courtesy Dr. Ramesh Madhavan and Dr. Kumar Rajamani

"We take shifts because there is no direct reimbursement, and we have to do other things during the course of that time. We have to multitask," Dr. Madhavan said, adding that this can make some neurologists feel burned out.

Dr. Tegeler said that, by contrast, the five vascular neurologists in Wake Forest Baptist’s smaller program have incorporated the telestroke coverage as part of their regular stroke attending call duties. Now, neurologists at some of the network hospitals "may not have to disrupt their office schedule for an hour or more to go over to the hospital to see an acute stroke patient," he said.

The presence of telestroke coverage also was used by one network hospital to help recruit a neurologist to a community where there had not been one, since the local neurologist would "not have to take stroke calls 24-7," Dr. Tegeler said.

In addition to the reimbursement issue, Dr. Levine said that telestroke has been hampered by "fears that the technology won’t work or may break down, or of lawsuits, and also doctors’ fear of something more technologically advanced than some are used to doing." Unlike their counterparts in emergency departments, Dr. Levine said, "neurologists aren’t techies, as a rule." However, telestroke experts are increasingly exploring cheaper and more portable options for videoconferencing, which could prove less imposing in terms of both startup costs to networks and in ease of use.

Most telestroke programs currently employ videoconferencing technology using equipment mounted on a mobile cart or purpose-built robots that can be driven to a patient’s bedside.

The robots, Dr. Schwamm said, "are a luxury. You can put a laptop with a specialized camera on top of a cart and roll it to a patient and get everything you need. Technology should not be where the costs are."

One team of researchers recently demonstrated that iPhones could be used in telestroke networks for patient assessment (J. Stroke Cerebrovasc. Dis. 2011 Oct. 24 [doi:10.1016/j.jstrokecerebrovasdis.2011.09.013]).

Dr. Madhavan said that he has been working with several technology companies on applications that would allow more telestroke consults to be carried out on portable devices using 3G or 4G networks. A migration to mobile platforms might help improve door-to-needle times in addition to reducing costs, he said.

Research Priorities

When Dr. Levine and Dr. Gorman first proposed telestroke networks in 1999, they envisioned them not solely as a way to increase TPA uptake but also as a way to facilitate patient entry into clinical trials of new stroke treatments.

This has yet to happen directly, Dr. Levine noted, but it may soon. "Community hospitals are not used to dealing with experimental medicines, or the logistics of randomization, and working with trial coordinators, IRBs, and rigorous and extensive data collection. But now that the systems are being built and there’s some infrastructure over the last 10 years, hospitals will hopefully start to see what clinical trials can bring them in terms of recruitment, state-of-the-art care while testing the most novel and promising treatments, and financially," he said.

Dr. Schwamm, whose program is among the few actively seeking to design clinical trials using telestroke, said that enrolling patients through telemedicine – and obtaining consent through video – is a challenge that, if met, will help broaden trial populations to reach more people living in rural and resource-poor areas. "The big studies tend to be done at the teaching hospitals, in urban areas. Telestroke can help allow people in the community to be enrolled," he said, leading to better representation of the population at large.

More telestroke programs are collecting at least short-term outcome or discharge data, measured in National Institutes of Health Stroke Scale (NIHSS) scores at admission and discharge. Dr. Madhavan’s team recently reported that, over 4 years, patients receiving intravenous TPA through the Michigan Stroke Network experienced a greater than seven-point reduction in NIHSS score by the end of their hospital stay (Stroke 2012;43:A2991).

Wake Forest Baptist tracks as a surrogate measurement whether patients are discharged to home, rehabilitation, or skilled nursing – and 79% are discharged to home or rehab, Dr. Tegeler said.

A few programs are collecting data on 90-day outcomes, an important measurement in clinical trials. For now, though, outcomes are mainly being analyzed with an aim to improve treatment protocols.

Although several studies have demonstrated that outcomes in the "drip and ship" model are comparable with those of patients admitted and treated at dedicated stroke centers, reducing average door-to-needle times and deciding who benefits from "drip and ship" versus "drip and keep" remain key areas of interest.

At the International Stroke Conference in February, neurologist Dr. Shadi Yaghi presented research on a cohort of 562 patients in the Arkansas SAVES program, a telestroke network that links 22 hospitals to stroke specialists at the University of Arkansas Medical Sciences Center in Little Rock. Patients with a worse outcome at 90 days were more likely to have received postthrombolysis care at a spoke hospital (odds ratio, 15.63; P = .019; 95% CI 1.56-166.67), Dr. Yaghi and his colleagues found (Stroke 2012;43:A2885).

"Our study showed that patients with moderate to severe strokes had better outcomes when transferred to a primary stroke center, likely due to more available resources and treatment options, such as neurointerventional treatment and neurocritical care treatment. Those with mild symptoms did not benefit from the transfer," Dr. Yaghi said in an e-mail interview.

Beyond TPA

In Europe and, in particular, Germany, telestroke has developed along a different course, with an emphasis on building multidisciplinary telestroke units in network hospitals.

Instead of concentrating solely on identifying candidates for TPA, network hospitals have teams of nurses, physiotherapists, occupational therapists, speech and language therapists, and physicians trained through seminars held at stroke centers and grand rounds in the affiliated hospitals. Workup of stroke patients is standardized, and treatment includes TPA for qualified patients under a drip-and-keep model.

Neurologist Dr. Peter Müller-Barna of Klinikum Harlaching, Stadtisches Klinikum in Munich, who is project coordinator of the TEMPiS (Telemedic Pilot Project for Integrative Stroke Care) stroke network in Bavaria, said in an interview that dedicated stroke units have proven "very efficient in reducing mortality and dependency in all stroke patients, with a number needed to treat (NNT) of about 5. In comparison to this, intravenous TPA is less effective; the NNT is about 10. Taken together, building up telestroke units is more expensive, but also by far more effective."

The TEMPiS network is one of the most intensively studied telestroke projects in terms of short- and long-term outcomes (Lancet Neurol. 2006;5:742-8; Neurology 2007;69:898-903). Recently, Dr. Müller-Barna and colleagues showed quality of care in TEMPiS network hospitals to be comparable with that seen at conventional stroke units with neurologists (Stroke 2012;43:A2820).

"In my eyes, the main obstacle for not establishing more telestroke programs in the U.S. is the lack of evidence for the telethrombolysis network approach in terms of improved clinical outcome, in contrast to the integrative stroke network approach with its telestroke units," Dr. Müller-Barna said.

Dr. Schwamm, who has collaborated with Dr. Müller-Barna’s group for research, said that much of the difference in the European and American telestroke models can be attributed to the differences in health care systems overall, with European systems better set up to invest in telestroke networks, keep patients in hospitals, and track long-term outcomes.

"In the U.S. the primary driver is the need to provide acute neurological assessment and make sure TPA-eligible patients have a chance to be treated," Dr. Schwamm said. "But telestroke still has lots of collateral benefits. It’s not only drip and keep and drip and ship but evaluate and triage. There’s a lot of that that goes on in telestroke networks even though that’s not the primary driver."

Courtesy Wake Forest Baptist Medical Center

Dr. Tegeler said that the Wake Forest Baptist program is increasingly looking at telestroke consultations as a pathway to other types of interventions.

"For patients who may have missed the window for TPA, telestroke can help facilitate consideration of eligibility for interventional or intraarterial procedures – there’s a whole host of things you could do with the right patients with the right lesions after the 4.5-hour window."

Dr. Tegeler and colleagues at Wake Forest Baptist have successfully pilot tested the use of telestroke technology for postdischarge follow-up visits, allowing patients who were transferred to the medical center for stroke treatment to be followed up, via videoconferencing, at spoke hospitals closer to their homes.

"There are opportunities to use it in all kinds of ways," he said.

Dr. Levine said that he has served on an advisory board and as a consultant for Genentech, which makes TPA (honorarium donated to stroke research), and has received reimbursement from Genentech for travel to a scientific meeting. He also has served as an expert witness in medical-legal cases involving stroke. Dr. Tegeler disclosed serving on the speakers bureau for Genentech. Dr. Schwamm reported being an advisory board member for lifeIMAGE, a company working on technology for the sharing of brain images in telemedicine. Dr. Madhavan said that he is a consultant for Robolytics, Process Proxy Corp., Great Lakes NeuroTechnologies, and MatrixView Ltd. No other sources had any relevant disclosures.