The objective of this study is to investigate the effectiveness of prehospital diagnosis and, when appropriate, of intravenous thrombolytic treatment of ischemic stroke. At the same time, we will take the opportunity to do an explorative study with the aim to further improve the intervention by biomarkers, and outcome measures volumetric measured by MRI images.

The intervention study aim to:

• Determine the time from symptom onset to thrombolytic treatment in the Mobila Stroke Unit (MSU) compared to the conventional model
• Determine the number of patients receiving thrombolytic treatment within the 4.5 hrs window in the MSU compared to the conventional model
• Determine if thrombolytic treatment in the MSU, when adjusted for time, results in better mRS- and Barthel outcome compared to treatment in the conventional model The explorative study aim to
• Determine if final IS infarction volume estimated by MRI, is independently correlated with time from symptom onset to thrombolytic treatment
• Define cut-off values for GFAP and RBP4 combined that with sufficient specificity and sensitivity can distinguish ICH from IS
• Determine the influence of time to treatment on pro-inflammatory markers after stroke

Hypothesis Intervention study I. The Treat - NASPP MSU model is feasible and reduces onset to treatment time less than 15 min II. The number of patient treated with thrombolysis within 4.5 hrs of symptom onset is significantly increased in the Treat-NASPP MSU model III. Treatment in the Treat-NASPP MSU model does not result in increased day 90 mRS and Barthel as compared to the conventional model when adjusted for time IV. Prehospital thrombolytic treatment of stroke does not increase the risk of secondary cerebral bleeding as compared to in-hospital thrombolytic treatment of stroke (cerebral bleeding worsening within 36 hrs less than 4 per cent, Norsk hjerneslagregister) Explorative study V. The final infarct volume, estimated by MRI, is significantly reduced when thrombolytic treatment is initiated already in the MSU VI. Biomarkers is a valid tool in the hyper acute phase of cerebral illness to exclude contraindication to thrombolysis VII. Reduced onset to treatment time results in lower levels of selected pro-inflammatory molecules

Treat-NASPP is a prospective controlled intervention study. The main aim of this study is to prospectively compare patients with ischemic stroke (IS) who are diagnosed and treated prehospitally in the MSU (intervention), with those who receive conventional pre- and inhospital diagnostics and treatment (control). At the same time we will do an explorative study with the aim to further improve diagnostic- (the biomarker study) and outcome measures (the biomarker study and MRI infarction volume study). The MSU will be staffed with an anaesthesiologist, a paramedic-nurse and a paramedic while on call. On the weeks the MSU is not operating, the data collection will take place from the conventional ambulance on weekdays from 8 am to 8 pm. Only ambulances staffed with the paramedic-nurse and paramedic that work as staff on the MSU will participate in the study. All emergency calls to the central EMS dispatch center (AMK 113) from the catchment region of Ostfold county will be screened for stroke symptoms by the EMS dispatcher according to normal procedures: they will use the FAST scale (one or more of the following neurological deficits: paralysis of arm or leg, facial paralysis, aphasia or dysarthria) and in regard to the inclusion criteria (listed below) the MSU will be dispatched when the inclusion criteria are fulfilled on weeks when the MSU is on call. The same inclusion criteria apply for the ordinary ambulance on weeks when the MSU is not on call. The intervention (prehospital CT and thrombolytic treatment) can only be administered in the MSU. The control (in-hospital CT and thrombolytic treatment) can only be administered in the hospital.

MSU procedure On site the anesthesiologist will take the actual medical history and do a rapid screening using the ABCDE's of trauma care. If the patient is stable and further investigations can proceed including NIHSS scoring. The patient will get two venous lines, and blood samples will be collected. Blood samples for biomarkers (se details below) will be stored and delivered to the laboratory at the hospital for further analyses and storage in bio bank (related to REK 2014/1161). The patient will be taken into the stroke ambulance, where the CT scan is performed and blood tests are run in the point of care laboratory (POC). After completing the CT scan examination, the anesthesiologist will immediately get in contact with the on call neurologist (stroke team) at Ostfold Hospital. The anesthesiologist will provide the stroke team with clinical history, POC blood tests, NIHSS score, time of symptom onset and any known clinical contraindication of thrombolysis. The stroke team and the on call radiologist will interpret the CT scan by teleradiology and a treatment decision will eventually be made. If indication of thrombolytic treatment the stroke ambulance nurse will prepare and initiate iv. rtPa (Actilyse) as per protocol (appendix). The prehospital study data will be completed in an electronic study form by the anesthesiologist.

Conventional ambulance management Conventional ambulance data collection will only take place when the MSU is not on duty. The conventional ambulance is then staffed with the same personnel (except the anaesthesiologis) as the MSU. After the paramedic or nurse-paramedic has taken the patient's actual medical history, performed a physical examination including NIHSS, established a venous line and eventually given the emergency treatment needed, the patient will be transported to the Østfold hospital. The paramedic will contact EMS dispatcher to inform patient inclusion in the study and the EMS dispatcher will contact the hospital stroke team. The paramedic will withdraw a blood sample for the biomarker study, and an additional blood sample will be taken at the hospital for standard analysis.

Therapeutic decision In both the conventional and the MSU pathway a cerebral CT scan will be conducted as soon as possible after symptom onset. Images will in both pathways be registered in the hospital PACS system, and interpret by the on call stroke team (neurologist and radiologist). In the MSU pathway the anesthesiologist will decide which patients are eligible for CT scan, but in the conventional pathway the neurologist on call in the emergency department will make this decision. Prehospital clinical assessment will include the actual medical history, a stroke scale score, and ABCDE evaluation. The clinical information and scores will be completed in designated study forms, and in the specialized prehospital patient record system AMIS. Both paramedic/nurse in the conventional pathway and the anesthesiologist in the MSU will do a stroke scale score, and scores will be analyzed for research purposes.

If a thrombolytic treated patient shows signs of clinical detoriation, expressed as an increase of 4 or more points on the NIHSS scale, an intracranial hemorrhage is to be suspected and the thrombolytic infusion must be stopped immediately. If in the MSU, a new cerebral CT scan should be conducted if the driving distance to the hospital exceeds 20 minutes, and the CT findings should immediately be transferred and reported to the hospital. After identification of location and distribution of damage based on the CT image, treatment will either be initiated on site in accordance with in-hospital procedures, or the patient will be transported directly to location for neurosurgery (see attachment 3).

Prehospital use of stroke scales: Stroke scales will be conducted in the MSU and the regular ambulance. All participants (anesthesiologists, paramedics and nurses) will attend a two-day course in stroke clinics, stroke treatment and the use of stroke scales. An online certification in NIHSS will be mandatory for participation.

MR volumetric: MRI will be completed at day 1 in all included patients treated with thrombolysis, prehospital or in hospital. Final infarct volume will be estimated by using image-series according to standard MRI protocol at Ostfold hospital.

Biomarker All serum- and plasma samples will be stored in the bio bank related to REK 2014/2261. Venous blood for measurements of biomarkers will be drawn at very early time points after symptom onset (i.e. after arrival of the paramedics/anesthesiologists at the scene), and at different time points up to weeks after arrival at hospital from all patients enrolled. The total volume of blood drawn will not exceed 100 ml on any day. All samples will be drawn by personnel certified by Østfold hospital. One standard serum tube, one EDTA plasma tube and one citrate serum tube will be used for blood collection. Blood samples will be centrifuged within 2 hrs of blood collection using a standard centrifuge (10 min at 1500-2000g). Serum/plasma will be transferred immediately in aliquots to Eppendorf tubes (each containing 0.5 ml) and stored at minus 80 degrees celcius. The Eppendorf tubes should be labeled appropriately indicating the patient number and the tubes collected per patient. Biomarkers will be analyzed in Professor Sandip Kanses lab, UiO, or shipped on dry ice to Professor Christian Forch, Department of Neurology, Schleusenweg 2-16, 60528 Frankfurt am Main, Germany. We will compare biomarker levels (index test) with CT findings (reference standard) and optimize cut-off points by using Receiver Operating Characteristics (ROC) analysis. Sensitivity and specificity will be calculated based on cross tabulations. The broad range of time points will be needed to monitor the rise and fall of biomarker concentrations, inflammatory response and diagnostic and prognostic windows in the wake of stroke.

Data monitoring, harms and auditing An independent safety committee consisting of two experienced stroke neurologists will review all safety data after 10, 20, 50, 100, 150 and 200 patients treated with thrombolysis. The committee will stop the study if they find evidence for an unacceptable increase of symptomatic cerebral bleedings (more than 4%) or deaths. Symptomatic cerebral haemorrhage is assessed as local or remote parenchymal haemorrhage combined with a neurological deterioration of four points or more on the NIHSS from baseline or from the lowest NIHSS value between baseline and 24 h, or a significant clinical worsening linked to the bleeding, or a bleeding leading to death.

The over-all rate of cerebral bleeding complications and the mortality rate at seven days will be compared with data from the Norwegian stroke register.

The main safety issues in the acute phase of stroke are linked to respiratory failure, cardiac arrhythmias and cerebral bleeding secondary to thrombolytic treatment.

The anaestesiologists working in the MSU are highly qualified to take care of acute respiratory and cardiac failure. The MSU is technically equipped as an air ambulance helicopter.

The main aim of our study is to provide early thrombolytic treatment to patients with cerebral infarction. The most serious complication of thrombolytic treatment is cerebral bleeding which may be fatal. Up to 10 % of all patients treated with thrombolysis will have a cerebral bleeding confirmed by CT, whereas only 2-4% of all patients will die or have a worsened outcome due to bleeding (called symptomatic bleeding). Cerebral bleedings following thrombolytic therapy will show up during the first few days, sometimes even in the very acute phase. Studies with thrombolytic therapy in MSU models have, however, not shown an increased risk of cerebral bleedings.

In-hospital cerebral bleedings will be diagnosed and treated according to the standard routines. Patients are monitored both in the MSU and in the stroke unit with NIHSS scoring at close intervals for 24-36 hours. At the hospital a MRI cerebral scan will be performed as a routine at approximately 24 hours after symptom start.

Statistical analysis The Treat-NASPP is designed in accordance with the Standards of Reporting of Diagnostic Accuracy initiative guidelines. For the prospective controlled intervention study our primary outcome will be i) OTT time, and ii) number of patients treated within 4.5 hours. Our secondary outcome will be mRS and Barthel at day 90 adjusted for OTT time.

For the primary outcomes we will use the Mann-Whitney U test. For power calculation, we want to compare two continuous variables in two groups, or compare two means. 2-sample, 2-sided equality. If we estimate time saved by MSU-treatment (intervention) as compared to the conventional pathway (control) we find Mean group 1: 210 minutes (29) Mean group 2: 180 minutes (we estimate 30 minutes saved in MSU) SD: 70 minutes Sampling ratio: 1 This gives us a number n=86 patients with thrombolytic treatment in each group. As we observed a mean reduction of time from onset of symptoms to diagnostics in the referred to NASPP study (unpublished data, REK 2013/2298) of 100 minutes, we consider this to be achievable.

Biomarkers As we expect concentrations of biomarkers to have a skewed distribution (21, 30), we will use the Mann-Whitney U-test when comparing concentrations in patients with ischemic stroke and ICH. The ROC-curve analysis will be used to calculate the diagnostic accuracy of the biomarkers in distinguishing between IS and ICH.

Sample size calculation for logistic regression is a complex problem, but based on the work of Peduzzi et al. (31) the following guideline for a minimum number of cases to include in the study can be suggested. Let p be the smallest of the proportions of negative or positive cases in the population and k the number of covariates (the number of independent variables), then the minimum number of cases to include is: N = 10 k / p. A statistician (Jo Røislien, see working group below) will be consulted for correct data analysis.

Data storage Clinical data will be registered and stored at Ostfold Hospital Kalnes. Study data will be un-identified and registered using the secure server provided by Teknograd https://www.teknograd.no

Inclusion Criteria:

• All patients with suspected stroke met by emergency services within 4 hours after symptom onset Stroke symptoms: sudden weakness of leg or arm, especially on one side, facial asymmetry and/or sudden trouble walking, and speech disturbance (Norwegian Index of medical emergencies 27.03-27.06).

Exclusion Criteria:

• Age under 18 years Pregnancy Female < 50 years and uncertainty of pregnancy Uncertainty regarding symptom onset time

Clinical tests and recordings will not be performed by investigator invested in the study.