Stroke is a common cause of death, disability and institutionalisation in New Zealand. Around one half of those that suffer a stroke will remain permanently disabled.1 For 16 years evidence has accumulated that confirms that recombinant tissue plasminogen activator, alteplase (rt-PA) can reduce the combined outcome of death and disability if given within three hours of stroke onset.2Recently published New Zealand Guidelines recommend the use of rt-PA for selected individuals up to four and a half hours from stroke onset.3 It has been demonstrated that rt-PA can be safely delivered within the New Zealand Health system.4 Despite these evidence-based guidelines, the burden of disability from stroke and the evidence that rt-PA can be safely administered, there remains disparity in access to thrombolysis treatment for stroke between hospitals in New Zealand.5A twenty-four hour service to deliver thrombolysis treatment for stroke was introduced at Wellington Regional Hospital in November 2009. This clinical audit reports the safety and efficiency of this model and examines whether there is a disparity between in-hours and out-of-hours treatment times. The purpose is to inform and stimulate discussion around the development of equitable, comprehensive stroke thrombolysis service models within New Zealand.Method The medical records of all patients treated with rt-PA between 1 November 2009 and 31 October 2010 were examined. Data collected included time of stroke onset to ambulance call; arrival in the emergency department (ED); ED assessment; CT scan and administration of rt-PA. The delivery of rt-PA is driven by a strict protocol and clinical eligibility criteria agreed by the Stroke Physician Leads at Capital and Coast, Hutt and Wairarapa District Health Boards (Table 1). All patients undergo CT scan prior to treatment and have a further scan between 24 and 36 hours after treatment. The National Institute of Health Stroke Scale (NIHSS) scores are calculated pre-treatment and 24 hours post treatment. Final place of discharge is recorded for all patients. Symptomatic intracranial haemorrhage is classified as any blood on the second CT scan and deterioration in NIHSS of greater or equal to 4 or leading to death. Asymptomatic haemorrhage is classified as any blood on the second CT scan with or without a deterioration in NIHSS score <4.6 In-hours care is considered as weekdays Monday to Friday 08:00 to 16:00, excluding public holidays, during which period rt-PA is delivered by the onsite stroke team comprising of a stroke specialist nurse, stroke registrar and stroke physician. All other periods are categorised as out-of-hours and rt-PA is delivered by the oncall medical team comprising two medical registrars, a house surgeon and an offsite stroke specialist. Statistical analysis was performed using unpaired two-tailed Student t-tests for parametric data. A p value of <0.05 was considered significant. Table 1. Clinical eligibility criteria for treatment with alteplase (rt-PA) All patients must have: - Clinical diagnosis of ischaemic stroke with measurable deficit - Onset of symptoms well established to be <4.0 hours Absolute clinical contraindications to treatment: - Profound stroke with obtundation, fixed eye deviation AND complete hemiplegia - Clinical presentation consistent with SAH, even if CT normal - Any evidence of haemorrhage on initial CT - CT/MRI - infarct >1/3 MCA territory - Active internal bleeding - Stroke, intracranial surgery or head trauma (last 3 months) - Any history of intracranial haemorrhage, AVM, aneurysm - Recent major surgery or organ biopsy (14 days) - Recent transmural MI or post-MI pericarditis (3 weeks) - Platelet count <100 00d7109/L - INR >1.5 (if patient not on warfarin, treatment can begin pending INR result) - IV heparin AND have an APTT >40 sec - Any LMWH in last 48 hours - Pregnancy or parturition last 30 days - Despite treatment, systolic BP >185 mmHg or Diastolic BP >110 mmHg Relative contraindications (presence of two or more absolutely contraindicate treatment): - Minor or rapidly improving symptoms - Recent lumbar puncture or arterial puncture at non-compressible site (14 days) - Blood sugar <3 or >22 mmol/L - INR between 1.3-1.5 - History of GI/GU or other internal bleeding within 21 days - Any seizure - Obvious small vessel disease on initial CT - Age > 80 years - Not ambulatory and/or dependent on others for ADLs and/or frailty Results Thirty-one people received rt-PA over the 12-month period, 16 of whom received treatment in-hours and 23 of whom were male. The average age of patients was 69 years, with range 43-87 years. Demographic details are shown in Table 2. Table 2. Demographic details for those treated in-hours and out-of-hours with alteplase (rt-PA) Variables In-hours (n=16) Out-of-hours (n=15) Overall Male, number Average age, years (range) Pre-treatment NIHSS, median (range) Post-treatment NIHSS,median (range) 11 68 (43-87) 12 (4-26) 6 (0-23) 12 69 (44-86) 10 (4-22) 3 (0-14) 23 69 (43-87) 10 (4-26) 5 (0-23) There was no significant difference in age, gender or NIHSS scores between those who presented in-hours compared with out-of-hours. Treatment times for the overall group are presented in Table 3. Table 3. Treatment times for all those given alteplase (rt-PA) Variables Average time in minutes (range) Symptom onset to ambulance Ambulance to ED ED Assessment ED to CT scan CT scan to rt-PA Average time to treatment 21 (0-83) 43 (0-89) 20 (0-51) 35 (1-53) 48 (7-150) 168 (88-240) Table 4 compares treatment times in-hours versus out-of-hours. Table 4. Comparison between treatment times in-hours and out-of-hours for those given alteplase (rt-PA) Variables Average time in minutes (range) in-hours (n=16) Average time in minutes (range) out-of-hours (n=15) Significance (p value) Symptom onset to ambulance Ambulance to ED ED assessment ED to CT scan CT scan to rt-PA Average time to treatment 22 (0-83) 42 (0-82) 21 (0-50) 31 (1-63) 34 (8-71) 152 (88-225) 20 (0-68) 43 (9-71) 19 (2-51) 38 (17-59) 62 (7-150) 185 (115-240) 0.77 NS 0.89 NS 0.72 NS 0.26 NS 0.01** S 0.03** S Safety 2014Two of the 31 patients died (6.5%), both from intracranial haemorrhage. Both suffered haemorrhagic transformations within large embolic infarcts; one in anterior circulation and the other middle cerebral artery territories. There were no other symptomatic intracranial haemorrhages. Three patients had small petechial intracranial haemorrhage without any deterioration in NIHSS score. Three further patients had minor gastrointestinal bleeds that required no specific treatment. All patients were treated within the clinical eligibility criteria with no protocol violations. Outcome 2014Median NIHSS pre-treatment was 10 (range: 4-26). Median post treatment NIHSS was 5 (range: 0-23) (Figure 1); 24 patients were able to return to their own homes, 4 patients went to live with relatives, 1 patient needed nursing home care and 2 died. Figure 1. Individual changes in NIHSS scores pre- and post-treatment with alteplase (rt-PA) Note: Two patients had NIHSS change 4 to 0; two patients 10 to 2; two patients 10 to 3 and two 5 to 5. Discussion Wellington delivers rt-PA efficiently and within agreed clinical eligibility criteria. The symptomatic intracranial haemorrhage rate (6.5%) is consistent with other published data from other centres in New Zealand and internationally though the absolute numbers of events are too low to draw robust conclusions.2,4,7 To assist those not familiar with the NIHSS scale a score <5 could be considered a 'minor' stroke, 5-15 a 'moderate to severe' stroke and >15 a 'severe to devastating' stroke. Of interest is the dramatic improvement in some individuals who presented with NIHSS scores >23 whom may have been assessed previously as not eligible for thrombolysis (Figure 1).7,8 There is growing evidence that rt-PA can be safely administered to people who would have been excluded from treatment in the original trials, for example those aged over 80 years.8 Stroke Physicians from Wellington, Hutt and Wairarapa District Health Boards met to review the current evidence and developed clinical eligibility criteria that allow discretion in decision-making (Table 1). The criteria are reviewed annually. There are other examples of centres using this approach which supports treatment within a changing evidence base.8,9 This collaborative approach drove the development of stroke thrombolysis services across the three District Health Boards. We acknowledge a lack of conformity in eligibility criteria for thrombolysis across New Zealand. Based on our experience a national process for review and revision of rt-PA eligibility criteria could become a catalyst to ensure this treatment is made available throughout the country. It is an approach we would recommend. Overall assessment and treatment times were similar to those reported previously by Christchurch Hospital.4Although comparisons between hospitals can be problematic, Christchurch reported a time from stroke onset to rt-PA bolus of 150 minutes for those treated in-hours which is remarkably consistent with our experience of 152 minutes (Table 4). In both centres potential candidates for rt-PA are given priority within the ambulance, ED and CT scan services and results show this process works well. The surprising result in Wellington is the prolonged time taken from CT scan to rt-PA treatment particularly out-of-hours. The audit demonstrated an important disadvantage in treatment times for rt-PA given out-of-hours (Table 4). This is of relevance to those planning stroke service models of care. The number needed to benefit for one additional patient to have a good outcome for rt-PA is 3.6 if treated within 90 minutes, 4.3 if treated within 90-180 minutes and 5.9 if treated within 180-270 minutes.10 Our numbers treated were too small to show differences in outcome, however the evidence is that thirty minutes delay has notable implications and so this disparity in time to treatment is clinically significant. Internationally there is evidence to show increased mortality amongst those admitted as medical emergencies out-of-hours and at weekends.11,12 Disadvantage in outcome has also been found specifically in those admitted with stroke at weekends and out-of-hours.13-15 Possible explanations such as reduced clinical staffing, availability of on-site senior medical officers and access to diagnostics have been suggested.11-15 This audit demonstrates that treatment delay after CT scan is a highly relevant consideration and accounted for the disparity in treatment times. Our delay was not due to CT scan reporting rather the availability of medical staff. Out-of-hours the medical registrar was more likely to be diverted to other tasks rather than accompany the patient to the CT scan. Subsequently when the patient returned to ED for treatment the registrar had to be re-called. In-hours the stroke registrar remains with the patient from ED to CT scan to rt-PA treatment. To address this disadvantage we propose that when a stroke patient arrives in ED out-of-hours a doctor or senior nurse remain with the patient until either the drug is administered or deemed contraindicated. This clinician will drive the speed of the process and alert the stroke consultant as soon as the scan is arranged. The stroke consultant can review the CT scan immediately and make the treatment decision. This should reduce time from CT scan to rt-PA treatment. As this time period accounts for the disadvantage in out-of-hours care this should be reduced by this recommendation. Stroke is common and rt-PA treatment substantially reduces disability. rt-PA should be available to all who suffer a stroke and fulfil eligibility criteria. This is currently not the situation in New Zealand. This clinical audit has demonstrated an inequitable delivery of treatment out-of-hours. We will address this disadvantage in Wellington and hope other Stroke services are inspired to do the same.

Stroke is a common cause of death, disability and institutionalisation in New Zealand. Around one half of those that suffer a stroke will remain permanently disabled.1 For 16 years evidence has accumulated that confirms that recombinant tissue plasminogen activator, alteplase (rt-PA) can reduce the combined outcome of death and disability if given within three hours of stroke onset.2Recently published New Zealand Guidelines recommend the use of rt-PA for selected individuals up to four and a half hours from stroke onset.3 It has been demonstrated that rt-PA can be safely delivered within the New Zealand Health system.4 Despite these evidence-based guidelines, the burden of disability from stroke and the evidence that rt-PA can be safely administered, there remains disparity in access to thrombolysis treatment for stroke between hospitals in New Zealand.5A twenty-four hour service to deliver thrombolysis treatment for stroke was introduced at Wellington Regional Hospital in November 2009. This clinical audit reports the safety and efficiency of this model and examines whether there is a disparity between in-hours and out-of-hours treatment times. The purpose is to inform and stimulate discussion around the development of equitable, comprehensive stroke thrombolysis service models within New Zealand.Method The medical records of all patients treated with rt-PA between 1 November 2009 and 31 October 2010 were examined. Data collected included time of stroke onset to ambulance call; arrival in the emergency department (ED); ED assessment; CT scan and administration of rt-PA. The delivery of rt-PA is driven by a strict protocol and clinical eligibility criteria agreed by the Stroke Physician Leads at Capital and Coast, Hutt and Wairarapa District Health Boards (Table 1). All patients undergo CT scan prior to treatment and have a further scan between 24 and 36 hours after treatment. The National Institute of Health Stroke Scale (NIHSS) scores are calculated pre-treatment and 24 hours post treatment. Final place of discharge is recorded for all patients. Symptomatic intracranial haemorrhage is classified as any blood on the second CT scan and deterioration in NIHSS of greater or equal to 4 or leading to death. Asymptomatic haemorrhage is classified as any blood on the second CT scan with or without a deterioration in NIHSS score <4.6 In-hours care is considered as weekdays Monday to Friday 08:00 to 16:00, excluding public holidays, during which period rt-PA is delivered by the onsite stroke team comprising of a stroke specialist nurse, stroke registrar and stroke physician. All other periods are categorised as out-of-hours and rt-PA is delivered by the oncall medical team comprising two medical registrars, a house surgeon and an offsite stroke specialist. Statistical analysis was performed using unpaired two-tailed Student t-tests for parametric data. A p value of <0.05 was considered significant. Table 1. Clinical eligibility criteria for treatment with alteplase (rt-PA) All patients must have: - Clinical diagnosis of ischaemic stroke with measurable deficit - Onset of symptoms well established to be <4.0 hours Absolute clinical contraindications to treatment: - Profound stroke with obtundation, fixed eye deviation AND complete hemiplegia - Clinical presentation consistent with SAH, even if CT normal - Any evidence of haemorrhage on initial CT - CT/MRI - infarct >1/3 MCA territory - Active internal bleeding - Stroke, intracranial surgery or head trauma (last 3 months) - Any history of intracranial haemorrhage, AVM, aneurysm - Recent major surgery or organ biopsy (14 days) - Recent transmural MI or post-MI pericarditis (3 weeks) - Platelet count <100 00d7109/L - INR >1.5 (if patient not on warfarin, treatment can begin pending INR result) - IV heparin AND have an APTT >40 sec - Any LMWH in last 48 hours - Pregnancy or parturition last 30 days - Despite treatment, systolic BP >185 mmHg or Diastolic BP >110 mmHg Relative contraindications (presence of two or more absolutely contraindicate treatment): - Minor or rapidly improving symptoms - Recent lumbar puncture or arterial puncture at non-compressible site (14 days) - Blood sugar <3 or >22 mmol/L - INR between 1.3-1.5 - History of GI/GU or other internal bleeding within 21 days - Any seizure - Obvious small vessel disease on initial CT - Age > 80 years - Not ambulatory and/or dependent on others for ADLs and/or frailty Results Thirty-one people received rt-PA over the 12-month period, 16 of whom received treatment in-hours and 23 of whom were male. The average age of patients was 69 years, with range 43-87 years. Demographic details are shown in Table 2. Table 2. Demographic details for those treated in-hours and out-of-hours with alteplase (rt-PA) Variables In-hours (n=16) Out-of-hours (n=15) Overall Male, number Average age, years (range) Pre-treatment NIHSS, median (range) Post-treatment NIHSS,median (range) 11 68 (43-87) 12 (4-26) 6 (0-23) 12 69 (44-86) 10 (4-22) 3 (0-14) 23 69 (43-87) 10 (4-26) 5 (0-23) There was no significant difference in age, gender or NIHSS scores between those who presented in-hours compared with out-of-hours. Treatment times for the overall group are presented in Table 3. Table 3. Treatment times for all those given alteplase (rt-PA) Variables Average time in minutes (range) Symptom onset to ambulance Ambulance to ED ED Assessment ED to CT scan CT scan to rt-PA Average time to treatment 21 (0-83) 43 (0-89) 20 (0-51) 35 (1-53) 48 (7-150) 168 (88-240) Table 4 compares treatment times in-hours versus out-of-hours. Table 4. Comparison between treatment times in-hours and out-of-hours for those given alteplase (rt-PA) Variables Average time in minutes (range) in-hours (n=16) Average time in minutes (range) out-of-hours (n=15) Significance (p value) Symptom onset to ambulance Ambulance to ED ED assessment ED to CT scan CT scan to rt-PA Average time to treatment 22 (0-83) 42 (0-82) 21 (0-50) 31 (1-63) 34 (8-71) 152 (88-225) 20 (0-68) 43 (9-71) 19 (2-51) 38 (17-59) 62 (7-150) 185 (115-240) 0.77 NS 0.89 NS 0.72 NS 0.26 NS 0.01** S 0.03** S Safety 2014Two of the 31 patients died (6.5%), both from intracranial haemorrhage. Both suffered haemorrhagic transformations within large embolic infarcts; one in anterior circulation and the other middle cerebral artery territories. There were no other symptomatic intracranial haemorrhages. Three patients had small petechial intracranial haemorrhage without any deterioration in NIHSS score. Three further patients had minor gastrointestinal bleeds that required no specific treatment. All patients were treated within the clinical eligibility criteria with no protocol violations. Outcome 2014Median NIHSS pre-treatment was 10 (range: 4-26). Median post treatment NIHSS was 5 (range: 0-23) (Figure 1); 24 patients were able to return to their own homes, 4 patients went to live with relatives, 1 patient needed nursing home care and 2 died. Figure 1. Individual changes in NIHSS scores pre- and post-treatment with alteplase (rt-PA) Note: Two patients had NIHSS change 4 to 0; two patients 10 to 2; two patients 10 to 3 and two 5 to 5. Discussion Wellington delivers rt-PA efficiently and within agreed clinical eligibility criteria. The symptomatic intracranial haemorrhage rate (6.5%) is consistent with other published data from other centres in New Zealand and internationally though the absolute numbers of events are too low to draw robust conclusions.2,4,7 To assist those not familiar with the NIHSS scale a score <5 could be considered a 'minor' stroke, 5-15 a 'moderate to severe' stroke and >15 a 'severe to devastating' stroke. Of interest is the dramatic improvement in some individuals who presented with NIHSS scores >23 whom may have been assessed previously as not eligible for thrombolysis (Figure 1).7,8 There is growing evidence that rt-PA can be safely administered to people who would have been excluded from treatment in the original trials, for example those aged over 80 years.8 Stroke Physicians from Wellington, Hutt and Wairarapa District Health Boards met to review the current evidence and developed clinical eligibility criteria that allow discretion in decision-making (Table 1). The criteria are reviewed annually. There are other examples of centres using this approach which supports treatment within a changing evidence base.8,9 This collaborative approach drove the development of stroke thrombolysis services across the three District Health Boards. We acknowledge a lack of conformity in eligibility criteria for thrombolysis across New Zealand. Based on our experience a national process for review and revision of rt-PA eligibility criteria could become a catalyst to ensure this treatment is made available throughout the country. It is an approach we would recommend. Overall assessment and treatment times were similar to those reported previously by Christchurch Hospital.4Although comparisons between hospitals can be problematic, Christchurch reported a time from stroke onset to rt-PA bolus of 150 minutes for those treated in-hours which is remarkably consistent with our experience of 152 minutes (Table 4). In both centres potential candidates for rt-PA are given priority within the ambulance, ED and CT scan services and results show this process works well. The surprising result in Wellington is the prolonged time taken from CT scan to rt-PA treatment particularly out-of-hours. The audit demonstrated an important disadvantage in treatment times for rt-PA given out-of-hours (Table 4). This is of relevance to those planning stroke service models of care. The number needed to benefit for one additional patient to have a good outcome for rt-PA is 3.6 if treated within 90 minutes, 4.3 if treated within 90-180 minutes and 5.9 if treated within 180-270 minutes.10 Our numbers treated were too small to show differences in outcome, however the evidence is that thirty minutes delay has notable implications and so this disparity in time to treatment is clinically significant. Internationally there is evidence to show increased mortality amongst those admitted as medical emergencies out-of-hours and at weekends.11,12 Disadvantage in outcome has also been found specifically in those admitted with stroke at weekends and out-of-hours.13-15 Possible explanations such as reduced clinical staffing, availability of on-site senior medical officers and access to diagnostics have been suggested.11-15 This audit demonstrates that treatment delay after CT scan is a highly relevant consideration and accounted for the disparity in treatment times. Our delay was not due to CT scan reporting rather the availability of medical staff. Out-of-hours the medical registrar was more likely to be diverted to other tasks rather than accompany the patient to the CT scan. Subsequently when the patient returned to ED for treatment the registrar had to be re-called. In-hours the stroke registrar remains with the patient from ED to CT scan to rt-PA treatment. To address this disadvantage we propose that when a stroke patient arrives in ED out-of-hours a doctor or senior nurse remain with the patient until either the drug is administered or deemed contraindicated. This clinician will drive the speed of the process and alert the stroke consultant as soon as the scan is arranged. The stroke consultant can review the CT scan immediately and make the treatment decision. This should reduce time from CT scan to rt-PA treatment. As this time period accounts for the disadvantage in out-of-hours care this should be reduced by this recommendation. Stroke is common and rt-PA treatment substantially reduces disability. rt-PA should be available to all who suffer a stroke and fulfil eligibility criteria. This is currently not the situation in New Zealand. This clinical audit has demonstrated an inequitable delivery of treatment out-of-hours. We will address this disadvantage in Wellington and hope other Stroke services are inspired to do the same.