Multiple preclinical and clinical studies, including the investigators' published work and the ongoing SOLUTION series, have consistently demonstrated that intracalvariosseous (ICO) injection can markedly increase drug exposure within the central nervous system with acceptable safety. This trial is designed to further evaluate the efficacy and safety of antibiotic delivery via the ICO route in the treatment of bacterial meningitis, particularly in patients with moderate-to-severe disease who have shown an inadequate response to standard therapy.

Bacterial meningitis remains a major cause of death and neurological disability despite advances in antimicrobial therapy, vaccination, and critical care. Moderate-to-severe disease, particularly healthcare-associated infection and cases caused by multidrug-resistant pathogens, continues to pose substantial therapeutic challenges because intravenous antibiotic therapy alone may not achieve sufficiently rapid or sustained drug exposure in cerebrospinal fluid and at the meningeal surface. Although intrathecal and intraventricular administration can increase local drug concentrations, these approaches are invasive and have limited diffusion.

Anatomical and physiological studies have demonstrated communication among calvarial bone marrow, dura, cerebrospinal fluid spaces, and glymphatic pathways, supporting the rationale for intracalvariosseous (ICO) injection as a regional route for drug delivery to the central nervous system. Preclinical and clinical studies, including the investigators' published work and the ongoing SOLUTION series, suggest that ICO injection can enhance local central nervous system drug exposure with an acceptable safety profile. In addition, an exploratory study conducted by the investigators, using vancomycin administered via ICO injection in an experimental animal model of bacterial meningitis, further supported the potential of this route to improve anti-infective efficacy.

This trial is a multicenter, prospective, randomized, open-label, blinded-endpoint study comparing ICO injection plus intravenous antibiotic therapy with intravenous antibiotic therapy alone.

The study aims to evaluate differences in efficacy and safety between combined treatment with ICO injection and intravenous antibiotics versus intravenous antibiotics alone in participants with moderate-to-severe bacterial meningitis who have shown a suboptimal response to treatment.

Participants with moderate-to-severe bacterial meningitis and inadequate improvement after 48-72 hours of initial intravenous treatment will be randomized in a 1:1 ratio.

In the intervention group, bilateral parietal ICO access devices will be placed, and the selected antibiotic will be continuously administered through the calvarial bone marrow route for 7 days while the same antibiotic is also given intravenously.

In the control group, treatment will consist of intravenous antibiotic therapy alone.

Guideline-based supportive care, including intracranial pressure management, seizure control, organ support, and other standard measures, will be provided in both groups.

Face-to-face visits will be conducted at randomization (baseline), 48-72 hours, Day 5, Day 8, Day 10, Day 14, Day 30 (±3 days), or on the day of discharge. A telephone follow-up visit will be conducted at Month 3 (±7 days).

A Data and Safety Monitoring Board (DSMB) will regularly monitor safety throughout the study. The trial has been approved by the Institutional Review Board (IRB) and Ethics Committee (EC) of Beijing Tiantan Hospital, Capital Medical University.

Inclusion Criteria:

• 1. Age 18-75 years, gender not limited;

• 2. Meeting the diagnostic criteria for moderate to severe bacterial meningitis:

  1. Meeting the diagnostic criteria for bacterial meningitis, i.e., meeting at least item i:

     i. Abnormal body temperature (>38℃ or <36℃), turbid or purulent Cerebrospinal fluid (CSF) , CSF leukocytosis (>500×10⁶/L), CSF glucose/serum glucose concentration <0.4, CSF protein concentration >50mg/dL, meeting the clinical diagnosis; ii. In addition to item i, positive microbial tests or cultures of specimen smears, drainage tube heads, implants, and CSF (excluding contamination and colonization), meeting the etiological diagnosis;

  2. GCS score ≤12 points or a decrease of 2 points from the previous score;

  3. At least one of the following conditions is present: altered consciousness, seizures, brain parenchymal involvement, severe manifestations such as mechanical ventilation or circulatory support.

• 3. After 48-72 hours of antibiotic treatment, if the investigator determines that the patient's condition shows no significant improvement or continues to worsen, or if at least one of the following conditions is present:

  1. Symptoms and signs related to intracranial infection show no trend of relief or worsen;
  2. CSF white blood cell count shows no trend of decrease, or increases after decreasing;
  3. CSF protein concentration shows no trend of decrease, or increases after decreasing;
  4. CSF/serum glucose concentration shows no trend of increase, or decreases after increasing;
  5. CSF bacterial culture remains positive or becomes positive again after initially being negative.

• 4. Based on the patient's condition, treatment with polymyxin B, tigecycline, or vancomycin may be necessary.

• 5. Obtain informed consent.

Exclusion Criteria:

• 1. History of allergy to polymyxin B, tigecycline, or vancomycin;
• 2. Received intrathecal or intraventricular antibiotic treatment before randomization;
• 3. Patients with severe pulmonary infection/acute respiratory distress syndrome (PaO₂/FiO₂ < 150 mmHg, FiO₂ ≥ 0.6, PEEP ≥ 5 cmH₂O) or whose primary cause of mechanical ventilation is not determined to be intracranial infection;
• 4. Patients with a primary extracranial infection focus (e.g., lungs, abdomen, urinary tract, etc.) who, after adequate fluid resuscitation, require vasoactive drugs to maintain MAP ≥ 65 mmHg (e.g., norepinephrine ≥ 0.25 μg/kg/min) and lactate > 2 mmol/L, or whose primary cause of critical illness is not determined to be intracranial infection;
• 5. Patients with contraindications to transcranial administration, such as severe skull fracture, poor visualization of the skull diploic, or planned decompressive craniectomy, which may affect transcranial administration.
• 6. Clinical signs of brain herniation, such as unilateral or bilateral pupillary dilation and fixation; or loss of other brainstem reflexes determined by the investigator to be caused by meningitis or brain herniation; or other uncontrollable signs of unstable vital signs;
• 7. Bleeding tendency deemed unfavorable for the procedure by the investigator: abnormal coagulation function (e.g., platelet count <50×10⁹/L; prothrombin time [PT]>3s), patients with a previous diagnosis of hemophilia or other coagulation disorders;
• 8. Patients with severe hepatic or renal insufficiency (where severe hepatic insufficiency is defined as alanine aminotransferase (ALT) value ≥3 times the upper limit of normal (ULN) or aspartate aminotransferase (AST) value ≥3 times the ULN; severe renal insufficiency is defined as serum creatinine (CRE) ≥1.5 times the ULN or glomerular filtration rate (eGFR) <40 mL/min/1.73m²;
• 9. Within the past 3 months, the patient has experienced an acute ST-segment elevation myocardial infarction and/or decompensated heart failure (meeting New York Heart Association [NYHA] functional class III or IV).
• 10. Patients with severe or extremely severe anemia (hemoglobin < 60 g/L) at the time of randomization;
• 11. Patients with active hepatitis B infection (positive hepatitis B surface antigen and/or positive serum HBV DNA or serum HBV DNA > 2 × 10⁸ IU/ml);
• 12. Patients with positive hepatitis C virus antibody or a history of positive testing;
• 13. Patients with positive HIV test or a history of positive testing;
• 14. Pregnant, lactating, or potentially pregnant patients, or patients planning to become pregnant;
• 15. Patients currently participating in other interventional trials or who have used other investigational drugs within one month or five drug half-lives;
• 16. Patients deemed unsuitable for participation in this study by the investigator.