The investigators aim to determine if oral antibiotics are clinically acceptable as treatment of brain abscess. Following 2 weeks of standard intravenous antibiotic therapy, half of patients will continue with this treatment for another 4 weeks or longer while the other half will be assigned to oral antibiotics for the remaining duration of treatment.

Treatment of brain abscess remains a considerable challenge due to the precarious location of the infection and the impenetrability of the blood-brain-barrier for most drugs. Thus, cure usually requires a combination of neurosurgical evacuation of abscess material and 6-8 weeks of high-dose intravenous (IV) antibiotic therapy to ensure eradication of bacteria within the abscess cavity. Disadvantages include risks of nosocomial infections and line-associated complications (e.g. bleeding, venous thrombosis, or need for replacement due to malfunction) in addition to the considerable costs of such long-term admission. However, improved insights into the pharmacokinetic properties and favourable bioavailability of some oral antibiotics may allow such treatment at an early stage. To date, there are no randomised controlled trials to guide treatment of bacterial brain abscess.

The investigators wish to determine whether a treatment strategy of transition to oral antibiotics after two weeks of treatment is non-inferior to continued IV antibiotics in clinically stable brain abscess patients assessed by the proportion with a favourable outcome at six months since randomisation.

Inclusion Criteria:

1. A clinical presentation (e.g. headache, neurological deficit or fever) and cranial imaging (CT or MRI) consistent with brain abscess AND
2. The physician responsible for the patient decides to treat the patient for bacterial brain abscess AND
3. Ability to take and absorb oral medications (including by nasogastric tube) AND
4. To have received relevant antibiotic therapy for bacterial brain abscess for 14 consecutive days before randomisation AND
5. Expected to be treated with antibiotic therapy for at least another 14 days after time of randomisation AND
6. No progression in symptom intensity or occurrence of new-onset neurological symptoms (excluding seizures) within five days before time of randomisation.

Exclusion Criteria (patients fulfilling either criteria):

1. Expected substantially reduced compliance with treatment (e.g. IV drug abuse)
2. Pregnancy (proven by positive urine or plasma human chorionic gonadotropin test in fertile women <50 years of age)
3. Concomitant (empirical) brain abscess treatment for tuberculosis, nocardiosis, Pseudomonas spp., fungi, toxoplasmosis or other CNS parasites
4. Device related brain abscesses (e.g. deep brain stimulators, ventriculo-peritoneal shunts)
5. Severe immuno-compromise defined as ongoing need for biological- or chemotherapy, prednisolone >20 mg/day for 14 days or longer, uncontrolled HIV/AIDS, haematological malignancies, and organ transplant recipients
6. Concomitant or unrelated infections necessitating IV antibiotics beyond seven days of duration after time of randomisation
7. Previous enrolment into this trial

Beginning six months and ending three years after publication, an anonymized dataset can be shared with qualified researchers who provide a methodologically sound proposal for a post-hoc study assessed by the members of the steering committee. The anonymized dataset will include baseline demographics of patients, laboratory results, treatments and primary outcome at six months since randomisation. In addition, the study protocol, statistical analysis plan, informed consent form, clinical study report and analytic code can also be shared. Proposals should be directed to the trial manager. To gain access, data requestors will need to sign a data access agreement. Data will be deposited at Mendeley Data (https://data.mendeley.com/).

6 months until 3 years after publication

Qualified researchers who provide a methodologically sound proposal for a post-hoc study assessed by the members of the steering committee may gain access to data after proper governmental approvals have been obtained.

An independent blinded endpoint committee will assess the primary outcome.