Intracerebral hemorrhage (ICH) is one of the common fatal types of cerebral apoplexy with high mortality and disability rates. Hematoma volume and complications of intracerebral hemorrhage are major predictors of early death and poor prognosis. The hematoma and its metabolites are key therapeutic targets. At present, in order to improve the prognosis of patients, cerebrospinal fluid(CSF) replacement with normal saline(NS) is commonly used in clinical practice to clear the bloody components, which shows a good clinical effect. However, due to the large difference between NS and CSF composition, it is easy to cause secondary injury of brain tissue. Therefore, the replacement of artificial CSF with similar CSF composition will be more effective in reducing the incidence of complications and improve the prognosis of neurological function.

The Magnesium-rich Artificial Cerebrospinal Fluid(MACSF) was designed and developed in the early stage of this project which has similar physical and chemical properties to physiological CSF, such as ion species, concentration, the potential of hydrogen (pH) value, and osmotic pressure. Animal experiments had confirmed its safety and effectiveness. In this study, patients with basal ganglia intracerebral hemorrhage ruptured into the ventricle or subarachnoid hemorrhage were stratified randomly divided into MACSF group and NS group. MACSF and NS were used as replacement fluid for lumbar puncture CSF replacement, respectively. By observing and comparing two groups of patients of the Modified Rankin Scale (mRS) on the days14, 30, 60 and 90 after onset; hematoma absorption rate, hemorrhagic CSF removal rate; changes of cerebral autoregulation； incidence of complications, such as acute obstructive hydrocephalus (AOH) and cerebral vasospasm (CVS); the changes of scores and scales about imaging; assessment of neurological function recovery, such as the National Institutes of Health Stroke Scale (NIHSS) and the Glasgow Coma Score (GCS) during hospitalization, headache duration and the Visual Analogue Scale (VAS), vomiting duration, duration of meningeal irritant, ICU hospitalization duration, total hospitalization duration; change of CSF and peripheral blood biochemical indicators. The objective is to evaluate MACSF replacement therapy in patients with basal ganglia cerebral hemorrhage broken into ventricles and nonaneurysmal subarachnoid hemorrhage of the influence of absorption rate and prognosis.

This study is a prospective, interventional and double-blind randomized controlled clinical trial. After approval of the clinical registry, patients who meet the standards of inclusion and exclusion criteria will be recruited to this study, if they and their relatives are willing to join the study. According to the clinical manifestations of the patients at admission and the results of cranial imaging, the patients' conditions were evaluated using the modified Fisher score, the World Federation of Neurological Surgeons Scale (WFNSS), the max-ICH score (max-ICH), the IVH score (IVHs), the Hijdra Sum Score (Hijdra), the Modified Graeb Score (mGS) and other grading criteria. Two groups (NS group and MACSF group) were divided into groups using a completely stratified randomization method. This study was a double-blind clinical study, and the grouping protocol was managed by the project research assistant. After the subject or his/her family members sign the informed consent, the project research assistant completed the grouping according to the randomly assigned sequence given by the clinical research center of the First Affiliated Hospital of Xi'an Jiaotong University. The project research assistant records the group and informs the specialist of the type of replacement fluid required. The replacement fluid was sent into the ward in a special container, and then lumbar puncture and CSF replacement were performed by the operator. The replacement liquid used in both groups was identical in appearance, color, temperature, and smell. During the whole study, the investigator did not participate in the preparation of the replacement liquid, and the subjects and their family members were also unaware of the grouping. After the case collection and electronic database records are completed, the research assistant will inform the researcher of the group name under the supervision of the third party to complete the unblinding, and the researcher shall inform the subjects of the group name in written form. For the control group, patients have treated the normal saline (0.9% Sodium Chloride Injection) as the replacement fluid. For the intervention group, MACSF will be used. The composition and concentration of MACSF is as follows:Na+: 146.2 mEq/L, K+: 2.7 mEq/L, Mg2+: 4.2 mEq/L, Cl-: 123 mEq/L, HCO3-:23.2 mEq/L, Glucose: 0.75 g/L. MACSF is prepared by the Pharmacy Intravenous Admixture Services staff of First Affiliated Hospital of Xi'an Jiaotong University, according to a specific formula in the specific working area. The whole procedure follows the aseptic principles strictly. Fresh-made MACSF will be placed in a specific container and sent to the patient ward through specific channels. Intracranial pressure (ICP) was measured by a non-invasive ICP test before lumbar puncture. If it is higher than 350cmH2O, a lumbar puncture can be performed 30 minutes after an intravenous drip of 100ml 20% mannitol. Then, for the patients who signed the informed consent, a CSF sample (3ml) was taken before each lumbar puncture CSF replacement and synchronous blood samples (6ml) were taken. The blood samples should be drawn into a non-anticoagulant tube and a heparin anticoagulant tube respectively. All the samples will be transferred to BioBank of the First Affiliated Hospital of Xi'an Jiaotong University in a liquid nitrogen tank. After re-melted in a water bath of 37°, the blood samples will be centrifuged at 4° for 15min at 3000r/min, the CSF samples will be centrifuged at 4° for 10min at 1000r/min and then the supernatant will be collected and stored at -80°. At the end of this study, the contents of oxyhemoglobin (oxy-Hb), S100 calcium-binding protein β (S100β), neuron-specific enolase(NSE), D-dimer (DD), matrix metalloproteinases-9 (MMP-9), Mg2+, Ca2+ in the samples will be detected by enzyme-linked immunosorbent assay (ELISA) or mass spectrometry. The total amount of replacement was generally 30ml, and the replacement was performed once every 2 days. When the red cell (RBC) count in CSF is less than 100×10^6/L, the CSF is considered to be cleared. CSF replacement is performed up to four times. NIHSS and GCS scores were evaluated daily after admission. Cerebral computerized tomography (CT) was performed at least every week after admission. If there are new low-density lesions, delayed cerebral infarction (DCI) will be diagnosed. A transcranial Doppler(TCD) examination was performed the first three days after admission and after each replacement. Cerebral autoregulation (CA) was evaluated by transient hyperemic response test (THRT), a transcranial Doppler (TCD)-based CA evaluation method, which describes the changes in cerebral blood fow velocity after a brief compression of the ipsilateral common carotid artery. The specific content is described later. CVS will be diagnosed by TCD, the diagnostic criteria will be described later. All subjects were followed up on the 14, 30, 60, and 90 days after onset, and the mRS was evaluated. Meanwhile, some indicators will be used to evaluate the safety of MACSF application, which include secondary infection, fluctuation of intracranial pressure, and hypermagnesemia. For patients or their family members who do not agree to do the CSF replacement, the sample collection can be used as a supplementary analysis for this study after they sign the informed consent for sample collection.

The diagnostic criteria and the solutions will be described in the next part.

Inclusion Criteria:

• Patient is between 18 and 80 years of old;
• Patients with ICH rupture into the ventricle in the primary basal ganglia region or non-aneurysmal subarachnoid hemorrhage confirmed by skull imaging examination;
• Patient is admitted to hospital within 72 hours after onset;
• Informed consent of the patients or their family members, and signed informed consent for CSF replacement treatment.

Exclusion Criteria:

• Non-spontaneous intracranial hemorrhage;
• Time from onset to admission is longer than 72 hours;
• The patient who needs surgical treatment;
• Contraindication of lumbar puncture, such as: cerebral hernia, severe intracranial hypertension, puncture site inflammation, blood system diseases, etc.;
• Patient has other serious diseases, such as heart failure, kidney failure, liver failure, etc;
• Patient or his/her relatives refuse to accept the above research plan.