Effect of continuous high cutoff hemodialysis on vascular permeability of septic patients with acute kidney injury

Dear Editor, The endothelial dysfunction and glycocalyx shedding that are observed during severe sepsis and septic shock lead to albumin loss and fluid shift in the interstitial space, with subsequent hypoalbuminemia, hypovolemia and generalized edema. Microalbuminuria and increased urinary albumin/creatinine ratio (MACR) are considered biomarkers of altered glomerular filtration barrier and indirect markers of injury to glycocalyx and increased endothelial permeability [1]. Mediators of systemic inflammation may be responsible for these effects [2]. Continuous venovenous hemodialysis with high cutoff membranes (HCO-CVVHD) effectively remove circulating inflammatory mediators and improve hemodynamics and respiratory functions in septic patients with acute kidney injury (AKI) [3]. To evaluate the effect of this extracorporeal treatment on glycocalyx function and glomerular endothelial permeability, we prospectively measured MACR in 10 patients with septic shock and AKI undergoing HCO-CVVHD. The study was approved by the local ethical committee. All patients were subsequently admitted to a mixed medical-surgical intensive care unit with microbiologically confirmed septic shock and AKI (Kidney Disease: Improving Global Outcomes stage II [4]), which indicated renal replacement therapy. Diagnosis and management of septic shock were based on up-to-date clinical practice. Of 10 patients, 5 had secondary peritonitis, 3 had respiratory infection, and 2 had soft tissue infections. Continuous venovenous hemodialysis with high cutoff membranes was performed with polyarylethersulfone membranes with pore size less than 0.01 μm (SepteX; Gambro Lundia AB, Lund, Sweden) and started within 12 hours of the diagnosis of septic AKI. Procalcitonin (PCT) and C-reactive protein (CRP) were dosed at the initiation of HCO-CVVHD and after 12, 24, and 48 hours of treatment. At the same time points, MACR was evaluated in the residual urinary output and Sequential Organ Failure Assessment score was calculated. All variables were expressed as median (I-III interquartile range) values. Wilcoxon signed rank test was used to compare median values of PCT, CRP, urinary and serum albumin and creatinine, and MACR at each time point, with respect to baseline value (P < .05).