Telemedicine versus face-to-face evaluation in the delivery of thrombolysis for acute ischaemic stroke: a single centre experience
- Correspondence to Dr Ajay Bhalla, Department of Ageing & Health, Guy's and St Thomas' Hospitals, 9th Floor, North Wing, St Thomas' Hospital, London SE1 7EH, UK;
Contributors AB: original idea for the study, data analysis and co-wrote the paper. MC: data collection, analysis and co-wrote the paper. JB: contributed original work used in the writing of the paper and edited the paper. AR: contributed to discussion and content of Methods and Discussion sections, and edited the paper.
- Received 23 March 2011
- Accepted 16 December 2011
- Published Online First 19 January 2012
Background Telemedicine is increasingly used in the UK to deliver thrombolysis. It is primarily used to enable assessment of people presenting with an acute stroke by a remote specialist in stroke care, and to determine eligibility for thrombolysis with alteplase (recombinant tissue plasminogen activator). This study aims to evaluate the process of acute stroke care, safety and outcome profiles when comparing face-to-face evaluation and telemedicine in the delivery of thrombolysis.
Methods This was a retrospective single centre cohort study, evaluating patients thrombolysed from July 2007 to December 2009 inclusive. All patients were given treatment within a 3-hour window from onset of symptoms. Of the 97 patients thrombolysed, 45 (46%) were evaluated by telemedicine. Process times of the steps taken to deliver thrombolysis for the two groups were compared. The authors include the rates of symptomatic intracranial haemorrhage (SICH). Outcome data include 3-month mortality and functional status.
Results Process times were significantly better in face-to-face: Admission to CT (p=0.001), CT to treatment (p≤0.001) and admission to treatment (p≤0.001). SICH occurred in four patients (7.7%) in the face-to-face group compared with the two patients (4.4%) in the telemedicine group (p=0.7). Favourable outcome: a modified Rankin score of 0–2 was observed in 19 patients (36.5%) in the face-to-face group compared with 19 patients (42%) in the telemedicine group (p=0.9).
Conclusions This analysis shows that the use of telemedicine compared with face-to-face evaluation is feasible in the delivery of thrombolysis during out of hours. There are several areas of our emergency process of hyper-acute stroke care that need improving when using telemedicine.
Thrombolysis using recombinant tissue plasminogen activator (rtPA) is one of the very few proven evidence-based interventions in salvaging brain tissue and improving outcome in acute ischaemic stroke patients. The benefit of using alteplase has now been seen up to 4.5 h after the onset of symptoms, but for the purposes of our study we aimed to treat within 3 h of the onset of symptoms. This is the time given with the license for alteplase.1
Access to thrombolysis still varies between hospitals in the UK and outside of normal working hours due to the lack of availability of in-hospital stroke specialist expertise. From the National Sentinel Stroke Clinical Audit 2010 data, 14% of all audit patients would have been considered eligible for thrombolysis, and only 25% of those eligible received thrombolysis.2
The sooner rtPA is given to stroke patients, the greater the benefit, especially if started within 90 min.3
To improve access and minimise delay in delivering thrombolysis, there has been an investment in telemedicine. At an NHS Stroke Improvement Programme meeting in February 2011, telemedicine had its own platform: Improving access and outcomes for acute stroke patients and expanding opportunities along the pathway.
In February 2011, Damian Jenkinson, National Clinical Lead NHS Stroke Improvement Programme, explained how telemedicine offers potential financial savings citing a quality, innovation, productivity and prevention business case. The business case supposed 10% stroke patients might benefit from thrombolysis and that half would be seen out of hours using telemedicine. Increased access to thrombolysis might mean more patients with no significant symptoms or disability, and fewer patients with a severe disability.
A telemedicine system consists of a digital network including a two-way video and audio conference facility, plus brain scan image transfer using a high-speed data transmission. Audio and video telecommunications are used to ensure stroke patients receive evidence-based interventions whenever and wherever they present.
A prospective, blinded, randomised controlled trial in North America suggested that telemedicine was efficacious in acute medical decision making and its use led to higher uptake rates of rtPA compared with telephone consultation.4 Meyer et al showed that the correct decision around thrombolysis was made 98% of the time using telemedicine. The thrombolysis rate was 28% in the telemedicine arm. Technical observations were noted in 12 (19%) tele-consultations and only 1 (0.9%) could not be performed due to technical failure.
Limited data exist in the UK describing the process of stroke care with this model. This study was conducted to describe the process of a hyper-acute emergency care pathway in delivering thrombolysis, and its safety and outcome. We retrospectively compare the process of delivering rtPA and its outcomes in patients evaluated using telemedicine compared with face-to-face evaluation in a central London Teaching Hospital.
Patients were mainly assessed via the emergency department at St Thomas' Hospital, and to a lesser extent as inpatients in other medical or surgical specialty wards while being treated for completely separate conditions. During normal working hours (9:00–17:00, Monday to Friday), a stroke specialist would accompany a senior stroke nurse to where ever the patient was located in the hospital. In the emergency department, the stroke team would be supported by an emergency department doctor and nurse. Outside of these hours, Monday to Friday and including the weekend, and with support from an on call stroke specialist, a senior stroke nurse unit provides on site expertise and would assist in the set up of telemedicine.
Between July 2007 and December 2009, patients with an acute ischaemic stroke and who after evaluation by telemedicine received thrombolysis with rtPA within a 3-hour time frame from the onset of symptoms were compared with patients who underwent face-to-face assessment.
For both telemedicine and face-to-face evaluation, a senior stroke nurse, trained in standardised assessment of the patient, was always present. Treatment pathways had been predetermined and written with the emergency department. Stroke and emergency department staff had organised three monthly training sessions to ensure there is an understanding of the hyper-acute management of stroke patients, and identify individual roles and responsibility when receiving a patient presenting to the emergency department. On arrival in the emergency department, the CT radiographer is alerted. The CT room in the emergency department is manned 24 h a day. Direct communication with the radiographers from either the stroke or the emergency room staff would ensure an urgent unenhanced CT brain sequence. Imaging was reviewed by the stroke physician either on site, as in the case with face-to-face consultation, or remotely if using telemedicine.
Decision to treat was made by the stroke specialist expert, and informed consent was taken where applicable and documented in both processes. Where there was a decision made via telemedicine, this was documented and transferred via email to the onsite stroke nurse, and then a paper copy was filed in the patient notes. Patients were only then transferred to the stroke unit when the senior stroke nurse felt the patient was safe to transfer, and with the appropriate monitoring.
Baseline characteristics such as age, sex, glucose, smoking status, admission blood pressure, prior history of diabetes, hypertension, atrial fibrillation, stroke, high cholesterol(>5 mmol/l) and handicap (modified Rankin scale (mRS))5 were recorded. Data were collected from patient medical notes, and centre specific data from the Safe Implementation of Thrombolysis in Stroke registry.
Stroke subtype was defined as total anterior circulatory stroke, partial anterior circulatory stroke, posterior circulatory stroke and lacunar stroke.6 This form of subclassification was designed so that we could demonstrate that the intervention was being delivered across all subtypes of an ischaemic stroke. Neurological deficits on admission prior to the use of rtPA were recorded using the National Institute Health Stroke Scale (NIHSS). Process measures included: stroke onset to admission time, admission to CT time, CT to rtPA treatment time, stroke onset to rtPA treatment time and admission to rtPA treatment time. Safety analysis included the incidence of symptomatic intracranial haemorrhage (SICH) which was defined as: ‘any remote or local parenchymal haemorrhage with an NIHSS greater than or equal to 4 at the time of the 24–36-hour follow-up CT scan’.7 Outcome included 3-month mortality and functional status. Functional status was assessed using the mRS. The mRS is sensitive to healthcare interventions after stroke and may yet be used as a national outcome indicator when looking at recovery from a stroke. In this study, the mRS was divided into favourable outcome (mRS 0–2) and unfavourable outcome (mRS 3–6) (box 1).
The modified Rankin scale is a commonly used scale for measuring the degree of disability or dependence in the daily activities of people who have suffered a stroke
The scale runs from 0 to 6, running from good health without symptoms to death.
0: No symptoms.
1: No significant disability. Able to carry out all usual activities, despite some symptoms.
2: Slight disability. Able to look after own affairs without assistance, but unable to carry out all previous activities.
3: Moderate disability. Requires some help, but able to walk unassisted.
4: Moderately severe disability. Unable to attend to own bodily needs without assistance, and unable to walk unassisted.
5: Severe disability. Requires constant nursing care and attention, bedridden, incontinent.
All the analyses were performed using Stata V.10.0. The Mann–Whitney test was used for continuous and ordinal variables, and the χ2 test for dichotomous variables or Fisher's exact test (where patient observations were five or less). Mean and SD have been used to summarise continuous variables. All p values are two sided and significance values have been set at p<0.05.
Between July 2007 and December 2009, 97 patients with acute ischaemic stroke were thrombolysed at St Thomas' Hospital, London. Of the 97 patients, 52 (54%) were assessed face-to-face while 45 (46%) were assessed by telemedicine. All patients received treatment within 3 h of onset.
Baseline characteristics of both groups are shown in table 1, revealing no significant differences in case mix apart from higher smoking status in the telemedicine group (p=0.03). Thirty-eight patients (73%) in the face-to-face group were independent (mRS 0–2) prior to stroke compared with 21 patients (47%) in the telemedicine group (p=0.13). The median NIHSS was higher in the face-to-face group (13, inter-quartile range 9–21 vs 12, inter-quartile range 2–24), but this did not reach statistical significance (p=0.3).
The onset to admission time was longer in the face-to-face group (p=0.06). Despite this, the admission to CT (p=0.001), CT to treatment (p<0.001) and onset to treatment (p=0.001) times were all longer in the telemedicine group. The overall admission to treatment time was longer in the telemedicine group: 61 min (inter-quartile range: 43–106) compared with face-to-face group: 33 min (inter-quartile range: 23–47), p<0.001 (table 2).
Outcomes and adverse events
SICH occurred in four patients (7.7%) in the face-to-face group compared with the two patients (4.4%) in the telemedicine group (p=0.7). Three-month mortality was eight patients (15.5%) in the face-to-face group compared with telemedicine group with five patients (11%) (p=0.6). Favourable outcome was observed in 19 patients (36.5%) in the face-to-face group compared with 19 patients (42%) in the telemedicine group (p=0.9).
This is the first UK study to describe the processes of emergency thrombolysis management for ischaemic stroke using a telemedicine model in a single centre.
The data from this present study suggest telemedicine could have a vital role in providing specialist stroke advice and expertise where there is a shortage of ‘in-hospital expertise’ out of hours. In our hospital, almost half of all thrombolysed cases were reviewed by telemedicine, reiterating its value in improving access to rtPA therapy. Another UK study suggests that telemedicine achieves similar safety and outcome measures to bedside evaluation for ischaemic stroke but no emergency process measures were reported.8
The data do show there is inequity in the quality of thrombolysis management when comparing face-to-face evaluation with telemedicine assessment. There were significant delays in the time from admission to CT in the telemedicine group. The prealert systems for the stroke team are the same out of hours as daytime and the emergency department CT scanner at St Thomas' Hospital is manned and running 24 h, suggesting this delay may reflect a difficulty in injecting urgency into the emergency pathway, without onsite stroke specialists.9 The mean CT to treatment times were also longer in the telemedicine group (36 min vs face-to-face 19 min). This may reflect the additional time taken to complete telemedicine consultations due to set up, assessment, obtaining consent or communication difficulties. Technological failure, which was not recorded in this study, may have contributed to delay. It would have been interesting to look at the staffing levels in the departments involved, for example, CT radiographers and emergency room staff out of hours, to see if reduced staffing levels directly affect process times. These factors might also explain the longer admission to treatment time in the telemedicine group.
The admission to treatment time in the telemedicine group was greater than the London target of <30 min from arrival to hospital. The median delay of approximately 1 h observed is in keeping with international data.10 11 In terms of complications, the SICH rates observed in the telemedicine group were within the range of the NINDS trial1 (4.4% vs 6.4%) reflecting that this model did not pose a major safety concern, although the definition of SICH may vary across studies.
The outcomes observed in this study were at least equivalent to those observed in an international monitoring study, implying that telemedicine compared with face-to-face evaluation was effective in routine clinical use.7
This study is limited by its observational nature and small sample size. There was no measurement of the duration of tele-consultation and no note of any technological failure limiting assessment of the process pathway.
If this use of telemedicine were replicated elsewhere, close attention to the timing of the process, identifying delays and then dealing with them are necessary so that there is no disadvantage to patient care.
For telemedicine to work successfully in hyper-acute stroke management, it is vital that the process is integrated within the emergency stroke care pathway, with collaboration between stroke specialists, emergency medicine specialists, nursing staff and radiology departments.6 This is important to ensure that the quality of acute stroke care is maintained through appropriate training, assessment and adopting standardised operating protocols for telemedicine.12
Teams adopting telemedicine in the assessment of acute stroke should be aware of the new real-time online audit of the first 72 h of process of stroke care, Stroke Improvement National Audit Project.13 There also needs to be an ongoing review of work-based competencies for those delivering assessment with telemedicine, for example, stroke specialty nurses and stroke specialty junior doctors. Teams embarking on using telemedicine should ensure there is a governance framework for its use.7 This should promote analysis of outcomes, particularly for those patients who have a major complication as a result of treatment.
There was inequity in the emergency processes en route to thrombolysis when using telemedicine.
Telemedicine improves accessibility to a 24/7 stroke service in this single centre experience, when out of hours specialist teams are not available face to face.
Telemedicine has similar safety data to face-to-face consultation outcomes.
Each telemedicine method of assessment should to operate within a clinical governance framework.
Current research questions
The use of telemedicine in the pre-hospital assessment with paramedical staff.
The application of telemedicine with hyper-acute stroke clinical research trials.
The cost effectiveness of telemedicine for both short and long outcomes.
The integration of telemedicine with other aspects of stroke care: rehabilitation, secondary prevention and stroke education.
Competing interests None.
Patient consent This article is related to a process (time data) and patient characteristics that are non-identifiable to specific individuals.
Ethics approval All data regarding patients are non-identifiable, and data were collected retrospectively from patient notes and the stroke database.
Provenance and peer review Not commissioned; externally peer reviewed.