Infectious endocarditis (IE) is associated with mortality rates of 10-12%. Adequate antibiotic therapy is crucial for survival and is administered in high doses due to the severity of the disease. In most cases, beta-lactam antibiotics (e.g. ampicillin, penicillin G, cefotaxime or cloxacillin) are employed. A number of patient characteristics, such as age, body weight, and renal function) influence the pharmacokinetics of these drugs. Yet, the interindividual variability is poorly understood meaning that a large proportion of patients are at risk of subtherapeutic or excessive drug concentrations that might result in treatment failure or side effects, respectively. In the present study, data will be collected on antibiotic concentrations in patients treated with beta-lactams for infectious endocarditis as well as patient characteristics and treatment outcomes. A mathematical model will be developed to determine which patient factors determine drug pharmacokinetics. Based on this model, predictions will be made by mathematical simulations on which dosing regimens are optimal for individual patients to ensure therapeutic and non-toxic drug concentrations. In total, 150 patients will be included at four University Hospitals in Sweden; Uppsala University Hospital, Sahlgrenska University Hospital in Gothenburg, Skåne University Hospital in Lund and Karolinska University Hospital in Stockholm. Following informed consent to participate blood samples will be collected at 6 time-points during a dose interval and then at 3 time-points weekly during the full treatment episode (maximum 6 weeks).