Acute kidney injury (AKI) is a common complication among patient admitted in the hospital worldwide, with estimates of prevalence ranging from less than 1% to 66%.1, 2 In critically ill patients approximately 49% were acute kidney injury network (AKIN) stage 3 which required intensive care unit (ICU) admission, kidney replacement therapy (KRT), and is associated with higher mortality rate.3 Sepsis-associated acute kidney injury (S-AKI) is a frequent complication in critically ill patient and is associated with high morbidity and mortality. S-AKI is defined as AKI in presence of sepsis without other significant contributing factors or simultaneous presence of both Sepsis-3 definition and Kidney Disease Improving Global Outcomes (KDIGO) criteria for AKI. 4, 5 Multicentre studies show that 30-60% of critically ill patient having AKI, and approximately 10-15% require KRT.6 Both Online-hemodiafiltration (OL-HDF) which is convection-based hemodiafiltration and conventional intermittent hemodialysis (IHD) which is diffusion-based hemodialysis are modalities of KRT that can be used to treat AKI in sepsis.7 Conventional intermittent hemodialysis (IHD) involves the removal of waste products and excess fluids from the blood by using a semipermeable membrane that acts as an artificial kidney. However, IHD has limitations in removing certain larger solutes, such as cytokines, which are involved in the inflammatory response associated with sepsis. In contrast, OL-HDF is a more advanced form of hemodialysis (HD) that combines convective clearance with diffusive clearance, resulting in more efficient removal of larger solutes.8 Several studies have suggested that OL-HDF may have advantages over conventional HD in the management of sepsis-associated AKI.9 Some studies found that OL-HDF was associated with improved patient survival and lower incidence of dialysis dependence compared to conventional IHD. Additionally, some studies have suggested that OL-HDF may have anti-inflammatory effects, which could be beneficial in sepsis.9-11 Some observational studies have shown that OL-HDF provide benefit over IHD including, a reduction in the length of ICU stay and a decrease in inflammatory surrogate markers. However, the effect of OL-HDF in improving survival has not yet been established. 1, 7, 9 Some studies have shown that sustained low-efficiency dialysis (SLED) may has advantage in hospital survival in over the continuous veno-venous hemofiltration (CVVH) modality.12 While there are several extracorporeal treatment modalities for AKI in critically ill patients but no randomized study has yet demonstrated a survival benefit over another.

A prospective and comparative study between IHD and OL-HDF groups showed no significant difference in mortality between the groups. However, a significant benefit in terms of a reduced length of ICU stays and vasopressor free day was found in the OL-HDF group.13 The past studies show that CRP and IL-6 levels increase in patients treated with IHD and remain stable in patients treated with OL-HDF, with a statistically significant difference. 14 In AKI patient, our aim is to remove small molecules such as uremic toxin or metabolic abnormalities. In chronic kidney disease, the benefit of high flux dialyser or OL-HDF to remove middle to large molecule such as ß2-microglobulin and others chronic inflammation molecules and cytokines through a combination of diffusion-based and convection-based techniques are well-known. These cytokines including C-reactive protein (CRP), interleukin 6 (IL-6), interleukin 10 (IL-10), procalcitonin (PCT), which are commonly elevated in AKI patient, have been shown to be significantly reduced by using hemodiafiltration techniques. 9, 15 However, the benefits of removing middle molecule and inflammatory makers in the acute setting such as AKI in critically ill patient remain controversial.11, 16, 17 Therefore, this study aims to verify the benefit of convection-based treatment in reducing inflammatory molecule such as CRP over diffusion-based treatment in both critically-ill and standard AKI patients.

However, it is important to note that OL-HDF requires specialized equipment and may be more complex to administer compared to conventional IHD. Additionally, it is generally more expensive. Therefore, the choice of KRT modality in sepsis-associated AKI should be based on careful consideration of the individual patient's clinical status and available resources.7 It is important to note that every patient's condition is unique and requires individualized treatment, so the specific choice of KRT modality should be made in consultation with a healthcare professional.

Study design: Stratified randomized controlled trial, parallel group Population: Sepsis-associated AKI requiring KRT patient Hospital: Ramathibodi hospital and Chakri Naruebodindra Medical Institute (CNMI) Randomization: Stratified Randomization based on hemodynamics of the patient (envelope technique)

• In hemodynamic unstable group (allocation 1:1) Sustained low-efficiency diafiltration (SLED-f) vs. Sustained low-efficiency dialysis (SLED) SLED-f is representative of convection-based KRT in critically ill patient. Dialysis prescription: Nikkiso DBB-05 (Tokyo Japan) Hemodiafiltration machine, High-flux polysulfone (Elisio 190HR) dialyzer (Nipro, Japan), minimum blood flow rate (BFR) 200-250 ml/min, minimum dialysate flow rate (DFR) 300 ml/min, dialysate Ca 2.5-3.5, Na 135-140, HCO3 35-38 mmol/L, dialysate temperature 35.5-36.5oC, dialysis duration 4-8 hrs, heparin 500 unit/hr if no contraindication, replacement fluid (online pre-dilution technique) with minimum flow rate 100 ml/min SLED is representative of diffusion-based KRT in critically ill patient which is currently a standard of dialysis care in hemodynamic unstable AKI patient.

Dialysis prescription: Fresenius 4008 (Bad homburg, Germany) hemodialysis machine, low-flux polysulfone (Elisio 170L) dialyzer (Nipro, Japan), minimum BFR 150-250 ml/min, minimum DFR 300 ml/min, dialysate Ca 2.5-3.5, Na 135-140, HCO3 35-38 mmol/L, dialysate temperature 35.5-36.5 oC, dialysis duration 4-8 hrs, heparin 500 unit/hr if no contraindication

• In hemodynamic stable group (allocation 1:1) Online hemodiafiltration (OL HDF) vs. Intermittent hemodialysis (IHD) OL HDF is representative of convection-based KRT in hemodynamic stable AKI patient Dialysis prescription: Nikkiso DBB-05 (Tokyo, Japan) or Fresenius 5008 (Bad homburg, Germany) online-hemodialysis machine, high-flux polysulfone (Elisio 190HR) dialyzer (Nipro, Japan), minimum BFR 200-300 ml/min, DFR 500 ml/min, dialysate Ca 2.5-3.5, Na 135-140, HCO3 35-38, dialysate temperature 35.5-36.5oC, dialysis duration 4 hrs, heparin 500 unit/hr if no contraindication, replacement fluid (online pre-dilution technique) with minimum flow rate 100 ml/min IHD is representative of diffusion-based dialysis treatment which is currently a standard of dialysis care in hemodynamic stable AKI patient.

Dialysis prescription: Fresenius 4008 (Bad homburg, Germany) hemodialysis machine, low-flux polysulfone (Elisiio 170L) dialyzer (Nipro, Japan), minimum BFR 200-300 ml/min, DFR 500 ml/min, dialysate Ca 2.5-3.5, Na 135-140, HCO3 35-38, dialysate temperature 35.5-36.5 oC, dialysis duration 4 hrs, heparin 500 unit/hr if no contraindication

Inclusion Criteria:

• Sepsis-associated AKI requiring KRT
• Age >18 years old

Exclusion Criteria:

• Pregnancy
• Patients with end stage kidney disease who are currently receiving kidney replacement therapy (chronic hemodialysis, peritoneal dialysis, kidney transplantation)
• Patient who has refused to consent
• Patient who dead or aim to dead within 24 hours after randomization
• Patient who is receiving very high dose of vasopressor (norepinephrine > 1 mcg/kg/min)
• Patient who is receiving CRRT at first randomization

Participants don't know which kidney replacement therapy they would receive.