EVT is the gold standard treatment in patients with AIS and LVO (1). However, currently, the probability of severe functional dependence or death at 3 months remains between 14-31% according to the results of the five large clinical trials that demonstrated efficacy and safety of EVT (2). There are two fundamental problems that need to be overcome to bring this treatment to more patients. First, patients need to seek help more often and sooner after their symptoms start. Second, once the alert occurs, the patients who may benefit from this treatment need to be identified and triaged to the appropriate facility. Unmet needs are the development of a portable, accurate, user-friendly device to quickly determine in real-time along the stroke journey if there is salvageable brain. This would be especially useful at prehospital levels to accelerate the transfer to comprehensive stroke centers and in the angio-room to monitor potential complications of the EVT.
Biomarkers could be useful to determine a priori patients who could most likely benefit from treatment from those in whom EVT would probably be futile. Multiple predictive factors of clinical response have been described (3). Clinical variables include age, baseline NIHSS score, systolic blood pressure, and hyperglycemia (3-7). Relevant radiological parameters are early ischemic signs in non-contrast CT (NCCT) assessed by the ASPECT score (3,8), thrombus extension and location,9 collateral status on CT or /MR angiography(10) and perfusion parameters such as infarct core and penumbra volume (11). Most of these factors require evaluation in a stroke center and are not available at previous levels of care, such as the pre-hospital setting or local hospitals.
SEP, and in particular the N20 response to peripheral nerve stimulation, reflect the electrical activity of different areas and tracts of the nervous system and correlate with cortical cerebral perfusion(12). SEP recording is a fast, non-invasive and reproducible technique that can be performed at the patient's bedside and provides real-time data on the functional status and perfusion of brain even when the patients are anesthetized or in an induced coma(13).
The PROMISE (Prognostic Accuracy of N20 Somatosensory Potential in Patients With Acute Ischemic Stroke and Endovascular Thrombectomy) trial (Stroke: Vascular and Interventional Neurology. 2023;3:e000735) investigated the presence and amplitude of the N20 response of SEP recorded before EVT in the ischemic hemisphere of 223 patients with anterior LVO undergoing EVT. N20 predicted functional independence at day 7 with a sensitivity of 93% (95%CI, 78-98%) and negative predictive value of 93% (80-98%), which increased to 100% after the EVT. Indeed, its predictive performance before treatment was superior to that of common factors used in clinical routine and added 0.07 points to the predictive model combining clinical variables, NIHSS score and ASPECT score (AUC of 0.84 versus AUC 0.77). Furthermore, N20+ had a higher predictive value of good outcome than advanced imaging variables recorded at baseline, such as the volume of infarct core, volume of hypoperfusion and collateral circulation. Taken together, these findings indicate that N20 might be a neurophysiological surrogate of salvageable brain and highlight the concept that ischemic brain tissue may remain viable from the perspective of cortical electrical function. Therefore, N20 SEP monitoring is a novel neurophysiological biomarker in acute stroke with the potential of having a major impact on the way stroke patients are managed throughout the world.
A major limitation of the PROMISE study was that N20 recording in the angio-room was interfered by electromagnetic artifacts precluding the blinded correct evaluation of the SEP in 25% of patients. These artifacts occurred in available commercial multifunction equipment not designed for the particular detection of N20 avoiding potential artifacts such as movements, electromagnetic interferences, noise, etc.
In this study, we aimed to validate a newly developed medical device, BraiN20®, specifically designed for the automatic recording and reading of the N20 SEP by not specialist or specifically trained physicians compared to the evaluation from two blind expert physicians.
The primary aim is to prove that the percentage of patients with optimal or good reliability of the BraiN20® Medical Device automatic recording of N20 is higher than 75% (i.e., the lower limit of the one-sided 95% confidence interval is higher or equal to 75%) assuming a true proportion equal to 87.5%, according to the classification by two expert neurophysiologists blind to BraiN20® reading results.
The principal secondary goal is to obtain a predictive capacity of functional outcome at day 7 of an algorithm based on the presence and characteristics of the N20 response recorded through the BraiN20® device with an AUC equal or higher than 0.80 in patients with AIS and anterior LVO undergoing EVT.
PROMISE20 is a prospective, multi-center, single arm, observational study with blinded evaluation of the primary endpoint of a cohort of patients with AIS and LVO undergoing endovascular treatment.
At admission at the emergency unit, acute stroke patients with suspected large vessel occlusion (Rapid Arterial oCclusion Evaluation, RACE \>4) will be immediately attended by neurologists and transferred to the CT/MR room for neuroimaging evaluation. If an ischemic stroke is confirmed and the patient is eligible for intravenous (IV) tissue plasminogen activator (tPA) or tenecteplase, IV thrombolysis will be administered without delay. CT angiography (CTA) or MR angiography (MRA) will be performed to confirm anterior LVO. Serum glucose, body temperature and blood pressure will be controlled following the American Heart Association (AHA)/American Stroke Association (ASA) guidelines.
ASPECT score will be measured by local investigators on CT or DWI but it will not preclude EVT if this is indicated according to local protocols. Perfusion CT or MR is advisable but not mandatory.
The patient will be transferred to the angio-suit as soon as possible for catheterization and EVT. Prior to puncture, BraiN20® electrodes will be set up, and NIHSS score and the result of the first BraiN20® reading will be recorded in the CRF. Informed consent will be obtained from the patients or family members.
Patients will be studied once AIS and LVO are confirmed by CT/CTA or MR/MRA and when, at the physicians' criteria, endovascular treatment is indicated. SEP monitoring with the BraiN20® medical device will be continuously performed from prior to arterial puncture to the end of the EVT while the patients remain in the angio-room. Patients' cohort will be followed up to 90 days after inclusion. The principal secondary outcome variable will be determined at day 7 by a local investigator blind to the BraiN20® recordings. Furthermore, functional capacity will be evaluated at 90 days by telephone interview performed by local blinded certified rater. Prospective, multi-center, single arm, observational study with blinded evaluation of the primary endpoint of a cohort of patients with AIS and LVO undergoing endovascular treatment. Patients will be studied once AIS and LVO are confirmed by CT/CTA or MR/MRA and when, at the physicians' criteria, endovascular treatment is indicated. SEP monitoring with the BraiN20® medical device will be continuously performed from prior to arterial puncture to the end of the EVT while the patients remain in the angio-room. Patients' cohort will be followed up to 90 days after inclusion. The principal secondary outcome variable will be determined at day 7 by a local investigator blind to the BraiN20® recordings. Furthermore, functional capacity will be evaluated at 90d by telephone interview performed by local blinded certified rater.
SEP recording will start in the emergency room or in the angiography room before puncture and monitored thereafter continuously during the procedure. Because of neurological damage can also occur after reperfusion, SEP monitoring will be extended until the patient is transferred from the angiography room to the acute stroke unit or the intensive care unit. SEP of both median nerves will be recorded, transferred and stored for their evaluation. BraiN20® Medical Device provides an automatic reading of the presence and amplitude of an N20 response ipsilateral to the cerebral hemisphere affected by the stroke. BraiN20® measures one N20 wave per 33 seconds. N20 is defined as the appearance of a negative peak with latency of ± 5 msec with respect to the expected normal N20 latency value (19.5 msec), amplitude greater than 0.1 microV and reproducible for a minimum of 2 successive recordings. The target variable is the reliability of the BraiN20® recording of N20 evaluated by two experts in neurophysiological recordings.
Registries will be recorded in a SD memory card that it is included inside the device. Data will be downloaded from the SD card weekly and transmitted to the neurophysiologist. Data evaluated by the experts will be registered into the eCRF. Reliability will be classified as optimal (100% confidence in the presence or absence of N20), good (≥80% confidence in the presence or absence of a N20 amplitude), low (\<80% confidence in the presence or absence of a N20 amplitude) and non-readable recording because the presence of artifacts. Optimal or good classification will be considered in agreement. Other differences in classification between examiners will be considered disagreements and handle as the worst classification.
A device validation will be performed in the first patients (in up to 6 patients, 2 patients by center). If re-adjustments in the device are required, a protocol (and Investigator's brochure) amendment will be submitted to the regulatory authority and ethics committee. These 6 patients will not be considered for the statistical analysis.
SEP expert examiners will evaluate the reliability of the N20 registration along the monitoring time (four potential N20 responses per minute). They will be blind to the results of SEP monitoring during the acute phase, neuroimaging variables and to the clinical outcome. N20 registration will be stored in a restricted field of electronic case report form (eCRF) to these examiners.
DATA COLLECTION METHODS Data collection will be done based on current registries already implemented in Catalonia. Data required for the PROMISE20 study is currently included by local investigators in the ICTUS process registry, which covers in-hospital data. This is an Institut Català de la Salut (ICS) mandated, hospital-based registry of all stroke admissions in the public Stroke Centers in Catalonia. However, the PROMISE20 study will generate a particular eCRF with the variables of interest.
A screening log form together with a prospective, government-mandated, population based registry of stroke code patients in Catalonia (CICAT: Stroke Code Registry of Catalunya) will be used for controlling an adequate consecutive enrollment of candidates in the PROMISE20 during the study period.
DATA MONITORING The Steering Committee has assigned a clinical research organization (CRO) (Anagram-Esic) to this study whose duties are to aid the principal investigator and the Steering Committee members in the maintenance of complete, legible, well organized, and easily retrievable data. Personnel from CRO will ensure that the study complies with relevant Good Clinical Practices (GCPs).
Periodic monitoring visits will be made throughout the investigation to assure that the investigator's obligations are being fulfilled. The PROMISE20 database will be keep confidentially at the CRO for centralized monitoring of the electronic records. Any data discrepancies may be queried during ongoing review of data by the Study Coordinating Center or may be identified and queried during routine monitoring visits.
Monitoring visits will be performed to verify data accuracy and ensure queries are resolved.
STATISTICAL METHODS
Primary Analyses - The primary endpoint is to estimate the percentage of patients with optimal or good reliability of the BraiN20® Medical Device automatic recording of N20 according to the classification by two expert physicians blind to BraiN20® reading results. Response categories from expert physicians (optimal, good, low and non readable) will be summarize by frequencies and percentages. In addition, for the combination optimal/good the one-sided 95% confidence interval according to Wald asymptotic confidence limits based on the normal approximation to the binomial distribution will be estimate.
Secondary Analyses
\- The principal secondary endpoint is to obtain a prognostic accuracy of an algorithm based on the presence and characteristics of the N20 response of SEP recorded through the BraiN20® equal or higher than 80% as a novel biomarker of functional outcome at day 7 in patients with AIS and anterior LVO undergoing EVT. Prognostic accuracy will be evaluated through of functional independence at 7 days (gold standard) according to the N20 response after the EVT by area under the receiver operating characteristic curve (AUC) and the 95% of confidence interval. Frequencies, percentages and 95% of confidence intervals will be also estimated for other characteristic of the test as sensibility and specificity and indicators of clinical relevance as predictive values. In order to avoid major problems of the device malfunction or its misuse, an early interim analysis at 25% of the final sample is recommended. If we assume a real proportion of good/optimal assessments of 0.875, 65 patients will be needed (17 of them will be analyzed at the interim).
No risk or benefit for patients is expected in the PROMISE20 trial since it is an observational study without any specific decision-making based on the BraiN20® results.
