Evaluation of Combined Extracorporeal Life Support and Continuous Renal Replacement Therapy on Hemodynamic Performance and Gaseous Microemboli Handling Ability in a Simulated Neonatal ECLS System

Kaitlyn R. Shank, Elizabeth Profeta, Shigang Wang, Christian O'Connor, Allen Kunselman, Karl Woitas, John Myers, Akif Undar

Research output: Contribution to journalArticle

2 Scopus citations


The objective of this study was to evaluate the hemodynamic performance and gaseous microemboli (GME) handling ability of a simulated neonatal extracorporeal life support (ECLS) circuit with an in-line continuous renal replacement therapy (CRRT) device. The circuit consisted of a Maquet RotaFlow centrifugal pump or HL20 roller pump, Quadrox-iD Pediatric diffusion membrane oxygenator, 8-Fr arterial cannula, 10-Fr venous cannula, and Better-Bladder (BB) with “Y” connector. A second Quadrox-I Adult oxygenator was added postarterial cannula for GME experiments. The circuit and pseudo-patient were primed with lactated Ringer's solution and packed human red blood cells (hematocrit 40%). All hemodynamic trials were conducted at ECLS flow rates ranging from 200 to 600 mL/min and CRRT flow rate of 75 mL/min at 36°C. Real-time pressure and flow data were recorded with a data acquisition system and GME were detected and characterized using the Emboli Detection and Classification Quantifier System. CRRT was added at distinct locations such that blood entered CRRT between the pump and oxygenator (A), recirculated through the pump (B), or bypassed the pump (C). With the centrifugal pump, all CRRT positions had similar flow rates, mean arterial pressure (MAP), and total hemodynamic energy (THE) loss. With the roller pump, C demonstrated increased flow rates (293.2–686.4 mL/min) and increased MAP (59.4–75.5 mm Hg) (P < 0.01); B had decreased flow rates (129.7–529.7 mL/min), and MAP (34.2–45.0 mm Hg) (P < 0.01); A maintained the same when compared to without CRRT. At 600 mL/min C lost more THE (81.4%) (P < 0.01) with a larger pressure drop across the oxygenator (95.6 mm Hg) (P < 0.01) than without CRRT (78.3%; 49.1 mm Hg) (P < 0.01). C also demonstrated a poorer GME handling ability using the roller pump, with 87.1% volume and 17.8% count reduction across the circuit, compared to A and B with 99.9% volume and 65.8–72.3% count reduction. These findings suggest that, in contrast to A and B, adding CRRT at position C is unsafe and not advised for clinical use.

Original languageEnglish (US)
Pages (from-to)365-376
Number of pages12
JournalArtificial organs
Issue number4
StatePublished - Apr 2018


All Science Journal Classification (ASJC) codes

  • Bioengineering
  • Medicine (miscellaneous)
  • Biomaterials
  • Biomedical Engineering

Cite this