Prediction of Functional Recovery in Patients With Intracerebral Hemorrhage

• OBJECTIVES:

1.1. Primary objective: To prospectively validate, in a multicenter, real-world clinical practice setting, the predictive value of a novel, simple, and easy-to-apply prognostic scale for determining functional recovery at 90 days, assessed using the modified Rankin Scale (mRS), in patients with intracerebral hemorrhage who received maximal treatment.

1.2. Secondary objectives:

1.2.1. To prospectively validate, in a multicenter, real-world clinical practice setting, the predictive value of a simple and easy-to-apply prognostic scale for determining long-term functional recovery (at 6 months and 1 year), assessed using the mRS, in patients with intracerebral hemorrhage who received maximal treatment.

1.2.2. To prospectively validate, in a multicenter, real-world clinical practice setting, the predictive value of a simple and easy-to-apply prognostic scale for determining mortality and "mortality or severe disability" (defined as mRS 5 or 6) at 90 days, 6 months, and 1 year in patients with intracerebral hemorrhage who received maximal treatment.

1.2.3. To assess the performance of the scale in predicting quality of life at 90 days, 6 months, and 1 year in patients with intracerebral hemorrhage who received maximal treatment.

1.2.4. To compare the prognostic performance of the new scale with that of the oICH and Max-ICH prognostic scales at 90 days.

1.2.5. To determine whether baseline serum concentrations of novel biomarkers or the increase in peripheral transferrin saturation are associated with functional outcome at 90 days, assessed using the mRS, in patients with intracerebral hemorrhage who received maximal treatment, and to evaluate whether they improve the predictive value of the clinical scale.

• TYPE OF STUDY: The study aims to validate a prognostic clinical scale for ICH designed by researchers from the Department of Neurology at the Germans Trias i Pujol Hospital, which, as a novelty, offers easy applicability and will be validated not only in the short term (3 months) but also in the long term (6 and 12 months) - an important aspect in a condition as devastating as Intracerebral Hemorrhage (ICH). In addition, new serum biomarkers will be analyzed that may potentially improve the scale's predictive capacity for both mortality and the probability of functional recovery among affected patients.
• DESIGN: Prospective, multicenter, observational study of consecutive patients diagnosed with spontaneous intracerebral hemorrhage, admitted to 11 reference stroke hospitals within the national health system and forming part of RICORS (Redes de Investigación Cooperativa Orientadas a Resultados en Salud [Results-Oriented Cooperative Research Networks]) in Cerebrovascular Diseases.
• VARIABLE COLLECTION:

• Baseline clinical, laboratory, and radiological variables: age, sex, vascular risk factors, comorbidities, use of antiplatelet/anticoagulant agents, time from onset to emergency department arrival, time from emergency department to brain CT, NIHSS, Glasgow Coma Scale, capillary glycemia, blood glucose, blood pressure, urea, CRP, white blood cell count, platelet count, baseline hematoma volume, ICH location and laterality. PRO-ICH scale score, oICH, and Max-ICH.

• Follow-up variables: Site of admission (emergency department, general ward, Stroke Unit, Critical Care Unit). Treatments: anticoagulation reversal treatment, advanced neurosurgical treatments (intracranial pressure monitor, external ventricular drain, hematoma evacuation, minimally invasive surgery, decompressive craniectomy, embolization). Therapeutic limits: adequacy of therapeutic measures (ATM) within the first 24 or 72 hours, previously established advance directives. Medical or neurological complications. Etiology. Discharge destination, length of hospital stay, NIHSS at discharge, mRS at discharge, in-hospital death, cause of death.

• STUDY EVALUATIONS: Patients admitted to the study will follow the usual clinical and therapeutic management protocol for ICH patients, based on current clinical guidelines. On admission, patients will undergo systemic and neurological evaluation, blood samples will be drawn according to standard practice, and the pre-established neuroimaging protocol will be performed (plain cranial CT +/- CT angiography). Additionally, for the study, two 10 ml blood tubes will be drawn - one at baseline and one at 1h ± 15 minutes after baseline - which will be processed and stored for biomarker analysis. Decisions on adequacy of therapeutic measures, standard treatment, and the indication of specific treatments (orotracheal intubation, treatment of complications, neurosurgical treatments) will be made by the treating physician, independently of the subject's participation in the study, and taking into account only the recommendations of clinical guidelines as well as the wishes of the patient and family. The variables necessary to calculate the scale under study (PRO-ICH) and the oICH and max-ICH scales will be collected. All data collected form part of routine clinical practice. At 90 days, 6 months, and 1 year, a centralized assessment of functional status (mRS) will be carried out through a structured telephone interview conducted by expert assessors blinded to the clinical situation, treatments received, and complications. The EuroQoL-5D health questionnaire will also be administered at the same time points.

• SAMPLE SIZE: According to Riley et al. (BMJ. 2024; 384:e074821; Stat Med. 2021; 40:4230), the external validation of a binary predictive model requires a sample large enough to assess discrimination - via the area under the ROC curve (AUC) - and calibration (observed/expected [O/E] ratio and slope). Adequate calibration is assumed (O/E = 1, slope = 1) and AUC = 0.835, based on data from the HIC-CAT registry (Catalonia), with 40% of patients having mRS ≤ 3. A 95% confidence interval (CI) width of 0.2 is assumed for O/E, and a skewed normal distribution with mean = -1.2, standard deviation = 1.5, skewness = -0.5, and kurtosis = 2.3 for the values of the model's linear predictor on the log-odds scale, used to estimate the calibration slope. For the AUC, the 95% CI must be ≤ 0.1. Thus, the minimum sample size required to accurately estimate the discrimination and calibration of the scale in this external validation study is 1,688 patients, with at least 676 events.

• STATISTICAL ANALYSIS: A descriptive analysis of the sample characteristics will be performed: categorical variables as absolute and relative frequencies; continuous variables as mean, standard deviation, median, interquartile range, minimum, and maximum. Validation of the scale will be carried out through discrimination and calibration analysis at the predefined follow-up time points (3, 6, and 12 months). Discrimination will be assessed using the AUC and the concordance statistic. Sensitivity, specificity, positive predictive value, and negative predictive value will be calculated. Calibration will be analyzed using the calibration curve, comparing expected and observed risk, the calibration slope, and the Brier score. Baseline concentrations of eight biomarkers on admission will be determined, and their association with 3-month prognosis (mRS 0-3 as good prognosis) will be assessed. For transferrin saturation, its variation between the baseline value and the value determined one hour after admission will also be analyzed. The inferential analysis will include multivariate logistic models to assess the impact of the biomarkers on functional prognosis. Those with significant effects (p<0.05) will be considered for inclusion in the scale. The predictive performance of the original scale versus the optimized scale will be compared using the DeLong test for AUC and the net reclassification improvement test. In addition, the influence of these biomarkers on model calibration will be explored using the Hosmer-Lemeshow statistic and the calibration slope. A sensitivity analysis will be performed to assess the robustness of the findings, including stratified analyses and adjustment for potential confounding factors.

• DATA SOURCE, MANAGEMENT AND QUALITY CONTROL: This study is an observational, multicenter, prospective study of patients with ICH. As part of the study, personal data (age, sex), health data (medical history, information on the current condition), biometric data (vital signs), and neuroimaging data (hemorrhage volume) will be collected from participants. This data will be obtained directly from participants after obtaining informed consent, in accordance with the provisions of Articles 6.1(a) and 9.2(a) of the General Data Protection Regulation (GDPR). In addition, biological serum samples will be obtained. Data will be handled in coded form, such that each patient will be assigned a code upon inclusion in the study, which will also be used to identify their biological samples. The form linking these codes to personal information (medical record number) will be stored at each site, separate from the study database, and kept by the principal investigator of each site, so that only investigators at each site will have access to their patients' personal data.

The project database containing the remaining data, without personal data, will be hosted on the REDCap platform (Research Electronic Data Capture, https.//www.project-redcap.org), for which the IGTP has controlled user access. IGTP's REDCap meets the following security requirements. Data access to REDCap is controlled through robust permissions management, data encryption, access auditing, and secure authentication; periodic security updates are performed, and it complies with data privacy and security regulations such as HIPAA (Health Insurance Portability and Accountability Act) in the United States and the GDPR (General Data Protection Regulation) in the European Union. A backup and restoration system is in place, with a daily backup of the database, allowing administrators to restore data in the event of loss or corruption. Thus, users participating in the study will only be able to enter data through REDCap access managed by the IGTP. Investigators from other sites will sign a confidentiality agreement provided by the IGTP, with restricted access to the data of patients from their own site.

• ETHICAL AND LEGAL ASPECTS: The principal investigator undertakes to submit the study for evaluation by the Clinical Research Ethics Committee (CEIC) of the Germans Trias i Pujol Hospital, and to comply with the fundamental ethical principles set out in the Declaration of Helsinki (Helsinki, October 2024). Furthermore, in accordance with Law 14/2007 on Biomedical Research and Royal Decree 1716/2011 regulating Biobanks, samples will be collected directly from the data subject. Samples will be kept at each site until the end of the study, when they will be sent to the Neuroscience Laboratory of the Germans Trias i Pujol Research Institute for analysis. After the study, any remaining samples will be stored as part of the collection (B.0000643) registered in the ISCIII National Biobank Registry. They will only be retained if specific consent is obtained, for a maximum of 15 years; otherwise, at the end of the study, any remaining biological samples will be destroyed. Study procedures will not begin until the study has been approved by the ethics committee of each participating site. The Ethics Committee and the Legal Department of the Germans Trias i Pujol Research Institute will review the relevant documentation at the start of the project, as well as any relevant amendments. They will act as advisory bodies on data protection matters.

Benefit-risk assessment for research subjects: Although patients with ICH who receive maximal treatment within the first 24 hours will be included, investigatorscannot rule out that the bias of adequacy of therapeutic measures, or self-fulfilling prophecy, will not be fully corrected, since, although therapeutic decisions will not be based on the new scale, the prognostic factors (age, NIHSS, volume, and time since onset) will nonetheless be known to the treating physician. Therefore, even though decisions are not made based on the scale score, they could be influenced by these factors, which are components of the scale, as occurs in routine clinical practice. Patients participating in this study should not expect any health benefit from their participation, but they may help improve care for future patients with cerebral hemorrhage.

Information to subjects and informed consent: The principal investigator and collaborating investigators undertake to inform patients and/or their legal representatives about the specifics of the study and to obtain their informed consent.

Confidentiality and data protection: The study will comply with Spanish and European data protection regulations: Organic Law 3/2018, of December 5, on the Protection of Personal Data and the Guarantee of Digital Rights, and Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data. Data will be obtained directly from participants after obtaining informed consent, in accordance with the provisions of Articles 6.1(a) and 9.2(a) of the General Data Protection Regulation (GDPR). Data will be handled in coded form, such that each patient will be assigned a code upon inclusion in the study, which will also be used to identify their biological samples. The form linking these codes to personal information (medical record number) will be stored at each site, separate from the study database, and kept by the principal investigator of each site, so that only investigators at each site will have access to their patients' personal data.

The project database containing the remaining data, without personal data, will be hosted on the REDCap platform (Research Electronic Data Capture, https.//www.project-redcap.org), for which the IGTP has controlled user access. IGTP's REDCap meets the following security requirements. Data access to REDCap is controlled through robust permissions management, data encryption, access auditing, and secure authentication; periodic security updates are performed, and it complies with data privacy and security regulations such as HIPAA (Health Insurance Portability and Accountability Act) in the United States and the GDPR (General Data Protection Regulation) in the European Union. A backup and restoration system is in place, with a daily backup of the database, allowing administrators to restore data in the event of loss or corruption. Thus, users participating in the study will only be able to enter data through REDCap access managed by the IGTP. Investigators from other sites will sign a confidentiality agreement provided by the IGTP, with restricted access to the data of patients from their own site.

The Germans Trias i Pujol Hospital, the Germans Trias i Pujol Research Institute (IGTP), and the study sponsor (the principal investigator) will act as data controllers within the framework of this observational study. No international data transfers are anticipated. The Germans Trias i Pujol Hospital belongs to the Catalan Health Institute (Institut Català de la Salut), which will be the final data controller.

• OBTAINING AND MANAGING BIOLOGICAL SAMPLES: following diagnosis of cerebral hemorrhage, two 10 ml blood tubes will be drawn from the patient - one at baseline and another at 1h ± 15 minutes after baseline - which will be processed and stored for biomarker analysis. The IGTP Cellular and Molecular Neurobiology Laboratory will determine the concentrations of biomarkers related to iron metabolism: transferrin saturation (TSAT) by TBE-Urea gel electrophoresis (Thermo Fisher Scientific) followed by western blotting and detection of the bands corresponding to its isoforms using a specific antibody against human transferrin; ferritin by ELISA (Meso Scale Discovery); free iron by a colorimetric assay (Abcam); soluble transferrin receptor by ELISA (Meso Scale Discovery); and hepcidin by ELISA (R&D Systems), as well as IL-6 by ELISA (Thermo Fisher Scientific), glial fibrillary acidic protein (GFAP) by ELISA (Thermo Fisher Scientific), and copeptin by ELISA (RayBiotech). In accordance with Law 14/2007 on Biomedical Research and Royal Decree 1716/2011 regulating Biobanks, samples will be collected directly from the data subject. Samples will be kept at each site until the end of the study, when they will be sent to the Neuroscience Laboratory of the Germans Trias i Pujol Research Institute for analysis. After the study, any remaining samples will be stored as part of the collection (B.0000643) registered in the ISCIII National Biobank Registry, in the -80°C ultra-freezer of the Department of Neuroscience located at the Germans Trias i Pujol Hospital. They will only be retained if specific consent is obtained, for a maximum of 15 years; otherwise, at the end of the study, any remaining biological samples will be destroyed. Study procedures will not begin until the study has been approved by the ethics committee of each participating site.

Inclusion criteria

• Patients ≥ 18 years diagnosed with spontaneous intracranial hemorrhage (ICH) by CT scan and treated maximally within the first 24 hours.
• Prior functional independence defined as modified Rankin scale (mRS) ≤ 3.
• Time from symptom onset or last known seen well ≤ 24 hours.
• Signed informed consent by the patient or their representative.

Exclusion criteria

• Age < 18 years
• Intracranial hemorrhage secondary to trauma, ischemic stroke, or fibrinolytic or endovascular treatment.
• Spontaneous ICH with limitation of therapeutic effort (LTE) within the first 24 hours. LTE is defined as the withholding or withdrawal of potentially life-sustaining treatment and/or the initiation of comfort care measures with the expectation that the patient may die as a result. Potentially life-sustaining treatments include: orotracheal intubation and mechanical ventilation, cardiopulmonary resuscitation, and neurosurgical treatments (intracranial pressure monitoring, external ventricular drainage, hematoma evacuation, decompressive craniectomy). LTE is not considered to apply if only a do-not-resuscitate (DNR) order was established, without limitation of any other treatment.
• Prior functional dependence, defined as mRS > 3
• Time from symptom onset or last known seen well >24 hours