#2722 Hemodialysis vascular access flow assessment by ultrasound dilution method (NephroFlow™) and ultrafiltration method with NIKKISO DBB-EXA™

Abstract

Abstract Background and Aims A functioning vascular access (VA) is needed to ensure optimal dialysis. Reduced or falling levels of VA flow volume (Qa) indicate VA dysfunction and predict thrombosis. Qa can be measured by a number of different techniques, most of all using dilution techniques such as ultrasound dilution to measure the amount of recirculation induced by reversal of blood lines. The NephroFlow™ hemodialysis monitor estimates Qa using ultrasound velocity measurements of flowing blood and their dilution with 0.9% NaCl. Nikisso DBB-EXATM dialysis machine has the capacity to measure the VA recirculation rate by photometry performing a controlled rise on ultrafiltration rate; Qa is subsequently estimated using the Krivitski formula. There is little literature on the comparison of the different methodologies, but it was described that the DBB-EXATM method underestimates Qa compared to others techniques [1]. Our study aims to compare the Qa between the NephroFlow™ ultrasound dilution technique and the Nikisso DBB-EXATM ultrafiltration method. Method We performed an observational study that included 45 hemodialysis patients followed at our unit. Measurements of VA flow using the NephroFlow® device and the Nikisso-EXATM Ultrafiltration method were performed in patients with a native arteriovenous fistula or a graft. They were hemodynamically stable without acute clinical conditions. Both studies were undertaken during the first 90 minutes of a mid-week dialysis session, under a constant dialysis blood pump flow with the ultrafiltration interrupted during the procedure after having reversed the dialysis lines. For each patient we calculated the average of two Qa evaluations conducted at Nikisso-EXATM, spaced 15 days apart, with an additional evaluation using NephroFlow® conducted between these assessments. Results Forty-five patients (male: 71.1%) with a mean age of 73.2 ± 8 years were selected. The main CKD causes were diabetic kidney disease (n = 13, 28.9%) and uncertain etiology (n = 13, 28.9%). The dialysis vintage of these patients was 142.6 months (minimum 19, maximum 168). An arteriovenous fistula was present in 38 patients (84.4%) and an arteriovenous graft in 7 patients (15.6%). In the entire group, the mean flow with NephroFlow™ was 986.9 ± 515 ml/min and with DBB-EXATM 859.7 ± 522 ml/min. The measured values of the two methods presented a significant positive correlation (by Pearson) (R = 0.822, P < 0.01) (Fig. 1). In the group of patients with arteriovenous fistula (n = 38, 84.4%), the mean flow with NephroFlow™ was 961.8 ± 395 ml/min and with DBB-EXATM 839.1 ± 577 ml/min. The measured values of the two methods presented a significant positive correlation (by Pearson) (R = 0.844, P < 0.01). In the group of patients with arteriovenous graft (n = 7, 15.6%), the mean flow with NephroFlow™ was 1017.3 ± 401 ml/min and with DBB-EXATM 909.2 ± 604 ml/min. The measured values of the two methods presented a significant positive correlation (by Pearson) (R = 0.906, P < 0.01). Conclusion We found a significantly positive correlation between the Qa measurements obtained using both methods. The measurement of Qa using the ultrafiltration method on Nikisso DBB-EXATM seems to be a valuable and easily implementable tool for VA surveillance in both arteriovenous fistula and grafts. However, a bigger number of studies are required in this field in order to clarify this correlation with an additional comparative analysis to assess the reproducibility of measurements.