UCSF

Extracorporeal membrane oxygenation (ECMO) is a life support technology capable of providing full replacement of respiratory and, if necessary, cardiac function. ECMO allows a patient to remain hemodynamically and metabolically stable while recovering from acute respiratory distress or surgical intervention, or while awaiting a lung transplant. In essence, ECMO provides time to heal.

In its simplest form, an ECMO circuit withdraws blood with low oxygen content and high carbon dioxide content from the body and passes it over a gas exchange membrane. A continuously flowing gas supply, or sweep gas, flows on the opposite side of the membrane from the blood. The gas concentration gradients existing between the blood and sweep gas enables diffusive transport of oxygen into the blood (oxygenation), while carbon dioxide diffuses out of the blood (ventilation). The oxygenated and ventilated blood is returned to the patient’s systemic circulation, where it can satisfy the metabolic needs of tissues and organs.

While a number of vital components comprise an ECMO circuit, including blood pumps, heat exchangers, and blood-contacting tubing, the most critical part is the oxygenator, which contains the gas exchange membranes and guides blood flow in contact with the membrane surface. Current clinically used oxygenators typically employ hollow fiber membranes; blood flows around bundles of thousands of individual membrane fibers while sweep gas flows through the fiber lumens. This design provides a large cross-sectional area for the blood flow, with millions of potential paths for red blood cells to pass through the interwoven plastic fibers, resulting in a low pressure drop and correspondingly low fluid shear stress through the oxygenator. However, this design consequently allows blood to follow a path of least resistance through the oxygenator, leading to incomplete utilization of the gas exchange capability of all hollow fibers. This inefficiency requires the use of excessive membrane area, on the order of 1 m2, to meet the gas exchange requirements of adult patients.

The purpose of the work outlined in this dissertation is to evaluate an alternative gas exchange membrane, in which a micron-thin polydimethylsiloxane (PDMS) film is mechanically supported by a silicon micropore membrane (SμM) with highly uniform pore geometry and distribution. The PDMS-SμMs revisit some of the earliest work on membrane oxygenators using new technologies derived from the semiconductor industry and the field of soft lithography to create a robust, planar gas exchange membrane. These membranes allow rapid gas exchange per unit area and also open the possibility for new oxygenator geometries in which blood flow is tightly controlled, as in the alveolar capillaries of the native lung, enabling highly efficient gas exchange in parallel plate oxygenators with surface areas less than one tenth of existing designs.

Using microfabrication techniques, I fabricated an oxygenation membrane consisting of a 4.63 μm thick PDMS film supported by a rigid SμM with 500 nm wide rectangular pores, which provided a planar gas exchange surface without the need for structural supports in the blood channel, capable of tolerating transmembrane pressure loads of 76 cmHg. Membrane mass transfer coefficients for oxygen and carbon dioxide were measured to be 5.49 ± 0.89 and 31.76 ± 0.80 mL STP min-1 m-2 cmHg-1, respectively. As a proof of concept, full oxygen saturation of porcine whole blood was achieved in a miniaturized ex vivo circuit for 3 hours without formation of gross clots, as well as demonstration of respiratory assist in a live ovine model, with an observed maximum oxygen exchange with blood of 1.94 ± 0.02 mL min-1 m-2 cmHg-1.