(M1330-03-13) Artificial Extracorporeal Replicate for Drugs (AER_x): A Low-Cost Extracorporeal Membrane Oxygenator (ECMO) Mimic for High Throughput Drug Screening

Purpose: Extracorporeal membrane oxygenation (ECMO) is a medical device that supports critically ill patients’ respiratory and cardiac systems. However, multiple drugs are known to adsorb on the components of these systems [1, 2]. For research, well-established ex-vivo circuit setups are used to determine adsorption profiles of drugs that are commonly used in ECMO [3, 4]. One drawback however, of the ex vivo ECMO circuit is its high cost which limits the high throughput testing of multiple drugs. Our goal is to build a mimic of the ECMO oxygenator to offset the cost of testing multiple drugs while also achieving high throughput screening to select lead candidates for testing in the actual ex vivo ECMO circuits. For this study, we used propofol as a model drug to be tested for its adsorption in the oxygenator mimic and compare it to the clinical ex vivo propofol adsorption which we have performed earlier.

Methods: We developed a new oxygenator mimic, namely AER_x. AER_x is built using the same poly(methyl pentene) (PMP) fibers that are used in commercial ECMO oxygenators (Quadrox iD, Getinge, Sweden). The surface area of the PMP fiber in AER_x_G6 (6^th prototype of AER_x) that is exposed to the blood flowing through the circuit is kept similar to Quadrox iD. In order to maintain a consistent temperature, we incorporated a water jacket connected to a heater circuit (Fig 1). AER_x-ECMO circuits (N=3) were assembled similar to a clinical ECMO circuit with a pump, custom perfusion tubing, and replacing the Quadrox iD with AER_x_G6. Circuits were set up in a closed loop with an IV bag representing the “patient”. Circuits were primed with human packed red blood cells (330 mL), fresh frozen plasma (150 mL), plasmalyte (100 mL), heparin (250 units), sodium bicarbonate (3.5 mEq), tromethamine (2.5 g), calcium gluconate (325 mg), and albumin (6.25 g) and maintained at 37^oC with a heating pump, and physiological pH was maintained by adding tromethamine, as needed. Flow was maintained at 1 L/min. Circuits were dosed with propofol (Diprivan®) to achieve a target concentration of 50 µg/mL. Samples were collected at different time points (1, 5, 15, 30, 60, 120, 180, and 240 mins) and were processed to separate the plasma. Plasma samples were then analyzed for propofol concentration using High Performance Liquid Chromatography. The adsorption of propofol in AER_x-ECMO system was then compared to propofol adsorption in the clinical ex vivo ECMO circuit.

Results: AER_x_G6 performed very similar to Quadrox iD in mimicking the adsorption profiles of propofol. At 5 mins, AER_x_G6 demonstrated 66% propofol recovery whereas Quadrox iD demonstrated 58% propofol recovery (Fig 2). At 1 hour, AER_x_G6 demonstrated 30% propofol recovery whereas Quadrox iD demonstrated 17% propofol recovery. At 2 hours, AER_x_G6 demonstrated 17% propofol recovery whereas Quadrox iD demonstrated 15% propofol recovery. None of the time points investigated demonstrated any significant differences between AER_x_G6 and Quadrox iD. AER_x_G6 and Quadrox iD showed a significantly high correlation (P < 0.000014) with an R² value of 0.9648. Finally, the cost of Quadrox iD is approximately $3000, whereas AER_x_G6 was built for $172 which is 17 times cheaper.

Conclusion: We successfully built a significantly cheaper mimic of the commercially available ECMO oxygenator. Our mimic performs similar to the commercial oxygenator in terms of adsorption after administration of propofol. AER_x_G6 can be used for further high throughput screening for drugs. Future tests would include investigation of other commonly used drugs in ECMO patients.

References: [1] T.J. Preston, A.B. Hodge, J.B. Riley, C. Leib-Sargel, K.K. Nicol, In vitro drug adsorption and plasma free hemoglobin levels associated with hollow fiber oxygenators in the extracorporeal life support (ECLS) circuit, J Extra Corpor Technol, 39 (2007) 234-237.
[2] E.D. Wildschut, M.J. Ahsman, K. Allegaert, R.A. Mathot, D. Tibboel, Determinants of drug absorption in different ECMO circuits, Intensive Care Med, 36 (2010) 2109-2116.
[3] C.E. Imburgia, J.E. Rower, D.J. Green, A.M. Mcknite, W.E. Kelley, C.A. Reilly, K.M. Watt, Remdesivir and GS-441524 Extraction by Ex Vivo Extracorporeal Life Support Circuits, ASAIO J, 68 (2022) 1204-1210.
[4] N. Khurana, T. Sünner, O. Hubbard, C. Imburgia, G.J. Stoddard, V. Yellepeddi, H. Ghandehari, K.M. Watt, Micellar Encapsulation of Propofol Reduces its Adsorption on Extracorporeal Membrane Oxygenator (ECMO) Circuit, AAPS J, 25 (2023) 52.

Acknowledgements: This work was funded by the University of Utah Clinical and Translational Science Institute pilot award program and by the National Institute of Health under the award number R01HD097775.


Fig 1. Design (A) and schematic (B) of AERx_G6.


Fig 2. Recovery of propofol from AERx_G6 and Quadrox iD at different time points.