Background Patients with kidney failure are at risk for lethal complications from hyperkalemia. Resuscitation, medications, and hemodialysis are used to mitigate increased potassium (K+) levels in circulating blood; however, these approaches may not always be readily available or effective, especially in a resource limited environment. We tested a sorbent cartridge (KC, K+ontrol CytoSorbents Medical Inc., Monmouth Junction, New Jersey) which contains a resin adsorber for K+. The objective of this study was to test the utility of KC in an ex vivo circulation system. We hypothesized that KC reduces K+ levels in extracorporeal circulation of donor swine whole blood infused with KCl. Methods A six-hour circulation study was carried out using KC, a NxStage (NxStage Medical, Inc., Lawrence, MA) membrane, blood bag containing heparinized whole blood with KCl infusion, 3/16-inch ID tubing, a peristaltic pump, and flow sensors. The NxStage permeate line was connected back to the main circuit in the Control group ( n = 6), creating a recirculation loop. For KC group ( n = 6), KC was added to the recirculation loop, and a continuous infusion of KCl at 10 mEq/hour was administered for two hours. Blood samples were acquired at baseline and every hour for 6 h. Results In the control group, K+ levels remained at ∼9 mmol/L; 9.1 ± 0.4 mmol/L at 6 h. In the KC group, significant decreases in K+ at hour 1 (4.3 ± 0.3 mmol/L) and were sustained for the experiment duration equilibrating at 4.6 ± 0.4 mmol/L after 6 h ( p = 0.042). Main loop blood flow was maintained under 400 mL/min; recirculation loop flow varied between 60 and 70 mL/min in the control group and 45–55 mL/min in the KC group. Decreases in recirculation loop flow in KC group required 7% increase of pump RPM. Conclusions During ex-vivo extracorporeal circulation using donor swine blood, KC removed approximately 50% of K+, normalizing circulating levels.

Background: Numerous biomaterials are developed to mitigate untoward effects of blood exposure to foreign surfaces on blood-contacting medical devices, such as extracorporeal life support (ECLS); however, limited testing is performed using clinical blood flow conditions and device components. A liquid infused nitric oxide-releasing (LiNORel) material was developed to prevent surface protein adsorption while eluting nitric oxide (NO) to minimize platelet activation and aggregation. The study objective was to assess LiNORel applied to full-scale ECLS circuit tubing during 6 h ex vivo circulation of porcine blood at a clinically relevant flow rate. We hypothesized that LiNORel reduces thrombus deposition and platelet sequestration versus unmodified tubing. Methods: Heparinized blood (0.75 U/mL) was collected from mechanically ventilated swine (n=9, 45-55kg) with ARDS following polytrauma in an unrelated multi-day ICU study. Blood was divided into circuits with the following (500 mL/circuit): CTRL -unmodified tubing; LI -tubing with liquid infused surface modification; NO -tubing with NOrelease modification; LiNORel -tubing with liquid infused and NO-release modifications. Circuits included a blood reservoir, centrifugal pump, and 2 tubing segments (1 m, 3/8" ID). Circulation was performed for 6 h at 1.5 L/min flow rate. Heparin was sparingly administered to maintain activated clotting times (ACT) of 125-160s. Data included: heparin administered, blood gas, methemoglobin (MetHb) as an indicator of NO toxicity, pump revolutions per minute. At baseline, 3-and 6 h, a blood panel was performed to assess platelet count/activation, thromboelastography, PT, aPTT, fibrinogen, D-dimer, von Willebrand factor (vWF) activity and plasma free hemoglobin. Post-circulation thrombus deposition was analyzed. Data were analyzed for within-group time-dependent changes and between group differences. All tests were two-sided (p<0.05 for significance). Results: Circuits remained patent in all groups with ACT in range. Numerically (n.s.), CTRL received more heparin (100±49 U) to maintain ACT (NO=50±32 U, LI=33±27, LiNORel=47±33 U). There was no group difference in blood gases and MetHb. Numerically, reduction in platelet count at 6 h was greatest in CTRL and least in LiNORel (n.s., Fig 1A). Activated platelets were elevated at 6 h in CTRL vs NO (p=0.013) and LiNORel (p=0.020) (Fig 1B). Numerically, procoagulant platelets were elevated in CTRL (Fig 1C). No group differences in thromboelastography, PT, aPTT and fibrinogen were observed. D-dimer was elevated at 6 h in CTRL (p=0.007) and LI (p=0.027) groups. vWF was elevated at 6 h in CTRL (p=0.001). Clot deposition was minimal and not different between groups. Conclusions: LiNORel reduced platelet activation in this model which utilized a clinically relevant blood flow rate and full-scale circulation tubing for partial lung support. Additional testing in vivo for multi-day duration is ongoing in our laboratory. Figure 1. Platelet count (A), concentration of activated platelets indicated by exp...