EARLY INITIATION OF CONTINUOUS VENO-VENOUS HAEMOFILTRATION IN SEPTIC PATIENTS IS ASSOCIATED WITH IMPROVED SURVIVAL: A SIX-YEAR ONE-CENTRE RETROSPECTIVE STUDY

It is currently unclear whether continuous veno-venous haemofiltration (CVVH) affects the course of sepsis. The aim of this study was to investigate if early-initiated CVVH treatment is associated with septic patient survival. We retrospectively analysed 158 septic patients treated with CVVH at Riga East University Hospital between 2014 and 2019. Based on outcome, the patients were divided into two groups: survivors (n = 66) and non-survivors (n = 92). All patients were treated with post-dilutional CVVH with a substitution fluid flow of 15-25 mL/kg/h. Late indications for CVVH were a blood urea concentration of 30 mmol/L or higher, severe metabolic acidosis with pH < 7.15, loop diuretic unresponsive fluid overload with oligo-anuria for six hours and severe hyperkalaemia (potassium concentration of 7 mEq/L or higher). The association of all categorical and continuous independent variables with in-hospital short-term and long-term mortality was calculated using Cox regression analysis. The hazard ratio (HR) was also calculated. On the 7th and 28th day of hospital stay, the septic patients' mortality rates were 43/158 (27.2%) and 85/158 (53.8%), respectively. The log rank test revealed that both short-term and long-term survival were significantly different (p = 0.018 and p = 0.001, respectively). Following adjustment for Sequential Organ Failure Assessment score, surgical treatment and mechanical lung ventilation, the association between early CVVH initiation and short-term survival lost its statistical power (HR = 0.655, p = 0.126); however, early CVVH initiation and long-term survival remained significantly associated (HR = 0.440, p = 0.001). Our main finding is that early CVVH initiation among septic shock patients is associated with improved survival.