This prospective randomized, controlled interventional pilot trial, aims to compare the achievement of the optimal target concentration with continuously administered flucloxacillin (FLU) or cefazolin (CZO) coupled with TDM and subsequent dose adjustment versus standard of care (intermittent bolus application without TDM-guidance) in patients with complicated Staphylococcus aureus (S. aureus) infections (CSAI). The overall goal is to individualize and optimize antibiotic treatment in a very vulnerable group of patients overcoming the standard strategy of "one-dose-fits-all".

Therapeutic drug monitoring (TDM) has recently been established as one of the cornerstones to individualize treatment of β-lactam antibiotics. It is particularly useful in patients hospitalized in the intensive care unit (ICU) being at risk to not achieve optimal antibiotic plasma concentrations due to a strongly altered metabolism. Along the same lines, continuous administration of β-lactam antibiotics instead of standard intermittent bolus administration may maintain drug concentrations in the target range throughout the dosing interval, and even contribute to a decrease in mortality. This prospective randomized, controlled interventional pilot trial, aims to compare the achievement of the optimal target concentration with continuously administered flucloxacillin (FLU) or cefazolin (CZO) coupled with TDM and subsequent dose adjustment versus standard of care (intermittent bolus application without TDM-guidance) in patients with complicated Staphylococcus aureus (S. aureus) infections (CSAI). The overall goal is to individualize and optimize antibiotic treatment in a very vulnerable group of patients overcoming the standard strategy of "one-dose-fits-all". The primary objective of this trial is to evaluate the achievement of the optimal pharmacological target concentration (100% fT 2 to 12 mg/L) in blood on day 3 after inclusion with continuous infusion FLU and CZO in combination with real-time TDM and subsequent dose adjustment, versus the current standard of care in patients with CSAI, and to estimate the effect size for future trials. To evaluate the PKPD of unbound FLU or CZO in the intervention versus the control group as measured by the following:

• Drug concentration at second or later TDM including the incidence of high (e.g. 100% fT>12 mg/L) and low concentrations (e.g. 100% fT<2mg/L)
• Incidence of potential drug-related toxicity during the course of treatment (e.g. nephrotoxicity, hepatotoxicity, neurotoxicity)
• Time to optimal target attainment and percentage of days with optimal target attainment in relation to total study drug treatment duration
• Intra-individual variability in FLU und CZO total and unbound plasma concentrations
• Factors associated with pharmacological target attainment (e.g. kidney function, protein, albumin, age, sex, weight, concomitant medication)

Sub-study I:

- Evaluation of the pharmacological profile of penicillin in patients in whom treatment was changed from FLU or CZO to penicillin due to a penicillin susceptible S. aureus strain.

Sub-study II:

- Assessment of patient satisfaction, rest-activity rhythms and sleep quality by actigraphy, sleep diaries and questionnaire in patients admitted to a general ward.

Patients will be randomized in two parallel groups stratified to the use of FLU or CZO in a 1:1 ratio to be treated either by continuous infusion plus TDM and dose adjustment or by standard intermittent bolus application.

Drug concentrations will be measured at day 1, 3, 5, 7, 10 and thereafter two times weekly until treatment period is completed (i.e. up to 6 weeks) or until discharge. After discharge, TDM will be performed 1x/week if the patient is treated in the outpatient parenteral antibiotic treatment (OPAT) program but without any dose adjustments. Dose adjustments of FLU and CZO in the intervention group will be performed according to a pharmacokinetic modelling application that is based on the data of our previous studies TARGET [2] and TARGET II (unpublished data). In the control group, blood samples will be drawn and analysed directly, but the results will not be communicated to the study team or a physician involved in the treatment of the patient. No TDM-guided dose adjustment will be performed in the control group

Inclusion Criteria:

• Informed Consent as documented by signature. For patients, who are not able to sign consent, a physician not involved in the current study has to confirm that patient's interest and rights are guaranteed during participation in the current study. Subsequently, informed consent will be obtained as soon as possible from the patient or his/her legally authorised representative.
• Age ≥ 18 years
• CSAI which is defined as (i) blood stream infection (BSI) with S. aureus or (ii) deep-seated infections caused by S. aureus (e.g. osteoarticular infections, deep-seated abscesses) without BSI.
• Intended or active (less than 24 hours) treatment with FLU or CZO

Exclusion Criteria:

• Patients on hemodialysis or eGFR<10 ml/min as these patients have a special pharmacokinetic
• Patients on Cytosorb® therapy
• Patients with liver cirrhosis CHILD B and C
• Patients who are very likely to stop treatment with FLU or CZO in the next 48 hours as per treating physician (because of treatment failure, switch to oral medication, palliative care, allergy etc.) or who are very likely to be discharged or transferred to another hospital in the next 48 hours as per treating physician.
• Polymicrobial infection except concomitant isolation of a likely contaminant (e.g. Staphylococcus epidermidis or Cutibacterium acnes). If an additional pathogen is identified after inclusion of the patient into the study, the patient will remain in the study.
• CSAI caused by methicillin-resistant S. aureus (MRSA)
• Participation in another study with investigational drug within the 30 days preceding and during the present study
• Previous enrolment into the current study
• Any uncontrolled or significant concurrent illness that would put the patient at a greater risk or limit compliance with the study requirements at the discretion of the investigator

There will be no blinding in this study. The antibiotic concentrations in the control group will be measured but not communicated to the study team or the treating physician.