BACKGROUND: The spread of multidrug-resistant bacteria represents a well-known problem, which must be face up by optimizing antibiotic therapy both in terms of choosing the most appropriate drug and of an adequate treatment duration. The method of administration is also a critical element. There are data relating to the maximization of the efficacy of Carbapenems and Piperacillin/Tazobactam by continuous infusion, able to constantly maintain adequate drug concentrations. Several studies, conducted comparing a standard administration of Carbapenem or Piperacillin/Tazobactam to an extended administration or continuous perfusion to evaluate safety and efficacy in terms of mortality reduction, have been documented. The achievement of optimal serum concentrations during continuous infusion has been documented both for Carbapenems and Piperacillin/Tazobactam, and for other types of antibiotics such as Cefepime and Vancomycin. The duration of antibiotic treatment is a critical factor for the prevention of relapses as well as the onset of resistance. The recommended duration of antibiotic treatment varies according to the site of infection and the type of pathogen and is generally between 7 and 14 days, however, in particular cases it is possible that the administration of antibiotics must be longer than one month. In general, the length of hospitalization is associated with a greater likelihood of complications for patients, with a substantial increase in the risk of developing multiple types of complications, such nosocomial infections, bed rest, bedsores, falls, malnutrition and disorientation. It should also be noted that prolonged hospitalization leads to a substantial increase in care costs. The advantage in terms of greater therapeutic success linked to the continuous infusion of some types of antibiotics has been used for the development of home infusion protocols for antibiotic therapy. There are numerous studies that show the feasibility and effectiveness of home infusion antibiotic therapy using elastomeric devices, documenting its substantial equivalence with respect to hospital treatment. The antibiotics for which there is evidence of feasibility are various, including Cefepime, Vancomycin and Piperacillin/Tazobactam.

STUDY DESIGN: The study is aimed to patients with severe bacterial infections who have started an antibiotic treatment and are benefiting from such therapy. The purpose of the study is to move the continuation of antibiotic therapy to the home setting once its efficacy and tolerability during hospitalization have been documented, in order to allow the patient a potentially eradicating treatment, of adequate duration without the need of hospitalization. Patients are enrolled when the efficacy and tolerability of the ongoing antibiotic treatment based on Cefepime, Meropenem, Piperacillin/Tazobactam or Vancomycin has been documented. The protocol requires that the patient go to the hospital every morning to replace the elastomer and to carry out a medical examination. Blood chemistry tests, which include at least blood counts, electrolytes, renal function, liver function and inflammation indices are scheduled to be performed at least once a week. Exams can also be done more frequently based on clinical needs. Antibiotic therapy should be carried out until the infection is cured, as per current clinical practice. For the purposes of the study, the patient remains under observation for 30 days after enrollment. Blood samples for the assessment of antibiotic concentrations will be performed in correspondence with the blood chemistry tests performed routinely for patient assessment. In any case, for each patient, a sample is taken for the dosage of the antibiotic used, which will be a single sample in the event of a continuous infusion, or a downstream sample (within one hour of the new antibiotic administration) and peak (30 minutes after the end of the infusion). The pharmacokinetic sampling relating to the outpatient phase will be carried out on the third or fourth day of continuous infusion therapy at the time of the elastomer change.

STATISTICAL ANALYSIS: This is an interventional monocentric study. The primary endpoint of the study is a reduction in hospitalization days. The study population is divided into 4 treatment arms: assuming that the average length of hospitalization for the standard treatment of a severe infection, for each of the types of antibiotic studied, is 14 days, with the treatment envisaged by the study it is assumed to reduce this duration,for all 4 treatment arms, to a number equal to or less than 12 days. With these premises, applying the One sample t-test design method, assuming α = 0.05 and β = 0.1, the number of patients expected is 9 for each treatment arm. Considering a 30% dropout, the total population to be enrolled for each type of antibiotic is 12 patients. In total, the study will enroll 48 patients. The study provides for an observation period of 30 days from enrollment. During this period, efficacy, safety and pharmacokinetic data will be collected for each treatment arm.

Inclusion Criteria:

• Patients aged 18 to 90 years.
• Ability to understand and sign informed consent.
• Treatment infection responsive to Cefepime, Meropenem or Piperacillin / Tazobactam, Vancomycin.
• Absence of allergies to Cefepime, Meropenem or Piperacillin / Tazobactam, Vancomycin.
• Evidence of clinical response to treatment with Cefepime, Meropenem or Piperacillin / Tazobactam, Vancomycin and at least 3 days of apyrexia with current infusion antibiotic therapy.
• Good tolerance to ongoing antibiotic treatment with Cefepime, Meropenem or Piperacillin / Tazobactam, Vancomycin.

Exclusion Criteria:

• Inability for any reason (e.g. absence of caregiver) to manage the elastomeric pump and access the hospital on a daily basis.
• Lack of adequate venous access.
• Inability to hydrate themselves properly orally.
• Infection involving the central nervous system.
• Creatinine> 2 mg / dL.
• Neutrophil granulocytes ≤ 1000 / µL.
• Platelets ≤ 20000 / µL.
• "aspartate amino transferase" and "alanine amino transferase" > 100 U / L.
• Bilirubin> 3 mg / dL.
• Presence of any comorbidities that, in the opinion of the physician, could compromise the safe execution of antibiotic home continuous infusion.