Sepsis is one of the main causes of mortality and morbidity in an ICU setting, while the responsible microorganisms most frequently isolated are multidrug-resistant gram-negative bacteria. Aminoglycoseides (AG) seem to be particularly effective in dealing with these microbes, however their potential toxicity, especially nephrotoxicity, often makes them an unsuitable treatment option. This becomes particularly evident in patients with already impaired renal function, a common occurrence in septic patients requiring ICU treatment. AG are bacteriocidal antibiotics the efficiency of which depends on the maximum concentration in patients' serum (Cpeak).

Pathophysiological changes in critically ill patients, result in significant distribution of the drug extravascullary resulting in a decreased concentration of the biologically active component. On the other hand, impaired renal clearance results in high serum drug levels (C trough) making the desired once-daily administration not always achieved.

The purpose of this study is to test the hypothesis of successful clearance of AG after achieving satisfactory serum levels and therefore their maximum effect minimizing potential toxicity, by using continuous veno-venous haemodiafiltration in patients with sepsis or septic shock and impaired renal function. This way, the aforementioned antibiotics could become a more frequent and potentially earlier choice for physicians in the treatment of sepsis and septic shock patients from multidrug-resistant microbes.

Sepsis is one of the leading causes of ICU mortality and morbidity. Globally, the first place in the list of responsible microorganisms appears to be occupied by multidrug-resistant gram-negative microorganisms (Escherichia Coli, Klebsiella pneumoniae, Pseudomonas aeruginosa και Acinetobacter baumannii). In the quiver of antibiotics we have against them, aminoglycosides (amikacin, gentamicin, tobramycin) have been particularly effective, either as monotherapy or in combination with other antibiotics, e.g. β-lactams. These are bactericidal antibiotics whose action depends on the concentration they achieve in the patients' plasma. As a measure based on which their effectiveness is predicted, the ratio of their maximum plasma concentration (Cpeak) against their minimum inhibitory concentration (MIC) is used, i.e. Cpeak/MIC and when this ratio is >8.

Pathophysiological changes that occur in the critically ill patient with sepsis involve the change in the volume of distribution (Vd) of the antibiotics, either due to increased renal clearance (increased GFR) or reduced renal clearance. In addition, studies have demonstrated a significant transcapillary escape of albumin extravascularly which increases the free fraction of the circulating antibiotic leading, especially in the case of hydrophilic drugs, to a greater and faster distribution extravascularly and thus to a lower concentration of the biologically active drug intravascularly (subtherapeutic concentrations).

Regarding aminoglycoside nephrotoxicity, this appears to occur in 10-20% of patients. Aminoglycosides are drugs that are not metabolized and are removed mainly by glomerular filtration.

Risk factors that have been associated with the occurrence of nephrotoxicity are increased serum trough levels, duration and frequency of administration, pre-existing kidney damage, hypoalbuminemia, liver dysfunction and the co-administration of nephrotoxic drugs, especially vancomycin.

Aminoglycosides are antibiotics with bactericidal activity based on reaching the maximum concentration of the antibiotic. Based on the above, it is understood that the prescription of aminoglycosides is a particular challenge, in terms of administering the effective dose. In practice this means that when the recommended daily dose of aminoglycoside is given, in patients with reduced renal clearance the minimum concentration (Ctrough measured 1 hour before the next administration) is at toxic levels (>5mg/dL for amikacin and >2mg/dL for tobramycin-gentamicin). Thus, administration of the planned next dose becomes impossible, resulting in possibly reduced clinical efficacy - response to treatment.

Given this fact, it is considered reasonable to artificially remove the antibiotic after reaching its maximum effective concentration in the serum and therefore its desired bactericidal effect, thus minimizing the risk of toxicity. Such a practice would make it possible to administer the next dose of antibiotic in 24 hours or at least earlier than usual (48-72 hours) increasing the likelihood of clinical efficacy.

There are few corresponding data in the literature. Taccone et al, showed that it is possible to administer the next dose (trough levels <5mg/L) at an interval of 36 hours in patients who received amikacin under CVVHDF at a dosage of 25mg/Kgr. Roger et al concluded that it is possible to re-administer amikacin at a dosage of 25 mg/Kgr after 48 hours in critically ill patients under CVVHDF in order to achieve pharmacokinetic-pharmacodynamic goals while simultaneously minimizing toxicity while Boyer et al, showed that when amikacin is administered at a dosage of 20mg/kgr with simultaneous use of CVVHDF, the infusion interval is reduced to 30 hours. The study by Brasseur et al 2016, included 15 ICU patients with confirmed MDR gram negative bacteria infection and demonstrated that it is possible to administer higher doses of aminoglycoside (25mg/Kgr amikacin) once daily in combination with CVVHDF thus achieving a cure rate 40% of patients. However, it failed to achieve a satisfactory Cpeak in 1/3 of patients only after 3 days of administration and after increasing the daily dosage.

Objectives of the study

Primary

• Possibility of more frequent administration of aminoglycosides with a clearance protocol using CVVHDF

• Cpeaks evaluation

• Evaluation of clinical effectiveness based on the SOFA score during days 1, 4, 8, 11 Secondary

• 28 day mortality

  • Renal function
  • CVVHDF-free days
  • WBC, CRP, Procalcitonin, Presepsin
  • ICU length of stay Study design Prospective study Study population ICU patients with sepsis or septic shock due to either confirmed Gram-negative infection susceptible to aminoglycosides, or probable Gram-negative infection (i.e. pending blood cultures or negative result) with impaired GFR.

Duration of study for each patient included 28 days or equal to the length of stay in the ICU if this is shorter RESEARCH METHODOLOGY 1. Study design-protocol In the present study, the possibility of removing aminoglycosides through CVVHDF is to be tested so that they can be re-administered in a shorter time than usual (48-72 hours) compared to the usual practice (not using CVVHDF to remove the antibiotic in question), in patients with sepsis or septic shock and impaired renal function.

1.1 The following will be performed on the patients who are going to participate:

• Biological fluid cultures

• GFR calculation (24-hour urine collection) before the start of aminoglycoside administration and 4 days after.

• Dosing: amikacin-25mg/Kgr, tobramycin and gentamicin 7mg/Kgr. The infusion will last 30 minutes

• TDM 30 min after the end of the infusion; before starting CVVHDF (approximately 4-6 hours after infusion) and before the next scheduled administration.

• administration of the daily dose, when Ctrough is: for amikacin <5mg/L and for tobramycin and gentamicin <2mg/L)

• monitoring of acute response markers and GFR.

• SOFA score on Days 1, 4, 8, 11 1.2 Participants will join one of three study groups:

  1. Initial CVVHDF. In this group, patients will receive CVVHDF for 96 hours, then it will be interrupted for 48 hours and resumed for at least another 48 (until the 8th day of the study).
  2. Late CVVHDF. Patients who do not require immediate initiation of CVVHDF but who may have impaired renal function (ClCr <40) or have increased Ctrough 24 hours after administration of the aminoglycoside. Therefore, in this study group, patients will receive the aminoglycoside initially without CVVHDF for 72 hours followed by initiation of CVVHDF and continued until day 8.
  3. Patients without CVVHDF. 1.3 Special notes: Patients who are going to receive the aminoglycoside and start CVVHDF, they will start CVVHDF after 4-6 hours of drug administration, if there is no reason for its earlier application. Fluid removal protocol will not be prescribed, but, this is left at the discretion of the attending physician.

Filter size: 1.5m2-1.7m2 (based on availability), changed every72 hours Anticoagulation: heparin/citrates.

Flow: 20-25ml/kgr/h. 2. Inclusion criteria

-age >18 years

• diagnosis of sepsis or septic shock based on established criteria (Sepsis-3)
• patients with GFR <40 ml/min
• Blood Stream Infection from a Gram-negative microorganism and for which the attending physician decides to receive an aminoglycoside or sepsis/septic shock for which it is decided to administer it
• signed consent of the same or 1st degree relative 3. Exclusion criteria
• absence of consent
• known allergic reaction to aminoglycosides
• infection from strains resistant to aminoglycosides. As long as the patient has received an aminoglycoside (empirical regimen) and the antibiogram follows it, the treatment can be modified based on the judgment of the attending physician, but the patient's data for the time he received the treatment under consideration are recorded and evaluated
• impossible placement of a central venous line with the possibility of continuous venovenous hemodialysis 4. Ethical issues The clinical study will be carried out in accordance with the principles of the Declaration of Helsinki and will receive the approval of the Scientific-Ethical Committee of the Larissa University Hospital.

Data protection: The confidentiality of all participants'; data will be guaranteed. The anonymity of the study participants will be preserved in any case. Only the principal investigator and study co-investigators will have access to participants' personal data. Consent will be obtained from each patient or their next of kin (in case of an affected level of consciousness) for the use of information from their medical record.

Inclusion Criteria:

• age >18 years
• diagnosis of sepsis or septic shock based on established criteria (Sepsis-3) 29
• patients with GFR <40 ml/min
• microbiemia from a Gram-negative microorganism and for which the attending physician decides to receive an aminoglycoside or sepsis/septic shock for which it is decided to administer it
• signed consent of the patients' next of kin

Exclusion Criteria:

• absence of consent
• known allergic reaction to aminoglycosides
• infection from strains resistant to aminoglycosides. As long as the patient has received an aminoglycoside (empirical regimen) and the antibiogram follows it, the treatment can be modified based on the judgment of the attending physician, but the patient's data for the time he received the treatment under consideration are recorded and evaluated
• unattainable placement of a central venous line for renal replacement therapy