In Vivo Testing of an Ambient Air Based, Portable, and Automated CO 2 Removal Controller for Artificial Lungs.

Portable artificial lung (AL) systems are under development, but there are few technologies available that adjust the carbon dioxide (CO 2 ) removal in response to changes in patient metabolic needs. Our work describes the second generation of a CO 2 -based portable servoregulation system that automatically adjusts CO 2 removal in ALs. Four adult sheep (68 ± 14.3 kg) were used to test the servoregulator. The servoregulator controlled air sweep flow through the lung to meet a target exhaust gas CO 2 (tEGCO 2 ) level in normocapnic and hypercapnic (arterial partial pressure of CO 2 [PaCO 2 ] >60 mm Hg) conditions at varying flow rates (0.5-1.5 L/min) and at tEGCO 2 levels of 10, 20, and 40 mm Hg. In hypercapnic sheep, average post-AL blood partial pressure of CO 2 (pCO 2 ) values were 22.4 ± 3.6 mm Hg for tEGCO 2 of 10 mm Hg, 28.0 ± 4.1 mm Hg for tEGCO 2 of 20 mm Hg and 40.6 ± 4.8 mm Hg for tEGCO 2 of 40 mm Hg. The controller successfully and automatically adjusted the sweep gas flow to rapidly (<10 minutes) meet the tEGCO 2 level when challenged with changes in inlet blood flow or target EGCO 2 levels for all animals. These in vivo data demonstrate an important step toward portable ALs that can automatically modulate CO 2 removal and allow for substantial changes in patient activity or disease status in ambulatory applications.

A Portable Servoregulation Controller to Automate CO2 Removal in Artificial Lungs.

Artificial lung (AL) systems provide respiratory support to patients with severe lung disease, but none can adapt to the changing respiratory needs of the patients. Precisely, none can automatically adjust carbon dioxide (CO2) removal from the blood in response to changes in patient activity or disease status. Because of this, all current systems limit patient comfort, activity level, and rehabilitation. A portable servoregulation controller that automatically modulates CO2 removal in ALs to meet the real-time metabolic demands of the patient is described. The controller is based on a proportional-integral-derivative (PID) based closed-loop feedback control system that modulates sweep gas (air) flow through the AL to maintain a target exhaust gas CO2 partial pressure (target EGCO2 or tEGCO2). The presented work advances previous research by (1) using gas-side sensing that avoids complications and clotting associated with blood-based sensors, (2) incorporating all components into a portable, battery-powered package, and (3) integrating smart moisture removal from the AL to enable long term operation. The performance of the controller was tested in vitro for ∼12 h with anti-coagulated bovine blood and 5 days with distilled water. In tests with blood, the sweep gas flow was automatically adjusted by the controller rapidly (<2 min) meeting the specified tEGCO2 level when confronted with changes in inlet blood partial pressure of CO2 (pCO2) levels at various AL blood flows. Overall, the CO2 removal from the AL showed a strong correlation with blood flow rate and blood pCO2 levels. The controller successfully operated continuously for 5 days when tested with water. This study demonstrates an important step toward ambulatory AL systems that automatically modulate CO2 removal as required by lung disease patients, thereby allowing for physiotherapy, comfort, and activity.

A Parametric Analysis of Capillary Height in Single-Layer, Small-Scale Microfluidic Artificial Lungs.

Microfluidic artificial lungs (µALs) are being investigated for their ability to closely mimic the size scale and cellular environment of natural lungs. Researchers have developed µALs with small artificial capillary diameters (10-50 µm; to increase gas exchange efficiency) and with large capillary diameters (~100 µm; to simplify design and construction). However, no study has directly investigated the impact of capillary height on µAL properties. Here, we use Murray's law and the Hagen-Poiseuille equation to design single-layer, small-scale µALs with capillary heights between 10 and 100 µm. Each µAL contained two blood channel types: capillaries for gas exchange; and distribution channels for delivering blood to/from capillaries. Three designs with capillary heights of 30, 60, and 100 µm were chosen for further modeling, implementation and testing with blood. Flow simulations were used to validate and ensure equal pressures. Designs were fabricated using soft lithography. Gas exchange and pressure drop were tested using whole bovine blood. All three designs exhibited similar pressure drops and gas exchange; however, the µAL with 60 µm tall capillaries had a significantly higher wall shear rate (although physiologic), smaller priming volume and smaller total blood contacting surface area than the 30 and 100 µm designs. Future µAL designs may need to consider the impact of capillary height when optimizing performance.

Toward a Servoregulation Controller to Automate CO2 Removal in Wearable Artificial Lungs.

A laptop-driven, benchtop control system that automatically adjusts carbon dioxide (CO2) removal in artificial lungs (ALs) is described. The proportional-integral-derivative (PID) feedback controller modulates pump-driven air sweep gas flow through an AL to achieve a desired exhaust gas CO2 partial pressure (EGCO2). When EGCO2 increases, the servoregulator automatically and rapidly increases sweep flow to remove more CO2. If EGCO2 decreases, the sweep flow decreases to reduce CO2 removal. System operation was tested for 6 hours in vitro using bovine blood and in vivo in three proof-of-concept sheep experiments. In all studies, the controller automatically adjusted the sweep gas flow to rapidly (<1 minute) meet the specified EGCO2 level when challenged with changes in inlet blood or target EGCO2 levels. CO2 removal increased or decreased as a function of arterial pCO2 (PaCO2). Such a system may serve as a controller in wearable AL systems that allow for large changes in patient activity or disease status.

A micro passive preconcentrator for micro gas chromatography.

We describe a microfabricated passive preconcentrator (µPP) intended for integration into gas chromatographic microsystems (µGC) for analyzing volatile/semi-volatile organic compounds (S/VOC). Devices (8 × 8 mm) were made from a silicon-on-insulator top layer and a glass bottom layer. The top layer has 237 apertures (47 × 47 µm) distributed around the periphery of a circular region (5.2 mm o.d.) through which ambient vapors diffuse at predictable rates. Two internal annular cavities offset from the apertures are packed with ∼800 µg each of commercial carbon adsorbents. Thin-film heaters thermally desorb captured vapors, which are drawn by a pump through a central exit port to a micro injector for analysis with a bench scale GC. The 15 test compounds spanned a vapor pressure range of 0.033 to 1.1 kPa. Effective (diffusional) µPP sampling rates ranged from 0.16 to 0.78 mL min-1 for short-duration exposures to ∼mg m-3 vapor concentrations. Observed and modeled sampling rates generally agreed within 15%. Sampling rates for two representative compounds declined by ≤30% between 0.25 and 24 h of continuous exposure. For one of these, the sampling rate declined by only 8% over a ∼2300-fold concentration range (0.25 h samples). Desorption (transfer) efficiencies were >95% for most compounds (250-275 °C, 60 s, 5 mL min-1). Sampling rates for mixtures matched those for the individual compounds. Dissipating no energy while sampling, additional advantages of this novel device include short- or long-term sampling, high capacity and transfer efficiency for a diverse set of S/VOCs, low transfer flow rate, and a robust fabrication process.

Assessing and improving the biocompatibility of microfluidic artificial lungs.

Microfluidic artificial lungs (µALs) have the potential to improve the treatment and quality of life for patients with acute or chronic lung injury. In order to realize the full potential of this technology (including as a destination therapy), the biocompatibility of these devices needs to be improved to produce long-lasting devices that are safe for patient use with minimal or no systemic anticoagulation. Many studies exist which probe coagulation and thrombosis on polydimethyl siloxane (PDMS) surfaces, and many strategies have been explored to improve surface biocompatibility. As the field of µALs is young, there are few studies which investigate biocompatibility of functioning µALs; and even fewer which were performed in vivo. Here, we use both in vitro and in vivo models to investigate two strategies to improve µAL biocompatibility: 1) a hydrophilic surface coating (polyethylene glycol, PEG) to prevent surface fouling, and 2) the addition of nitric oxide (NO) to the sweep gas to inhibit platelet activation locally within the µAL. In this study, we challenge µALs with clottable blood or platelet-rich plasma (PRP) and monitor the resistance to blood flow over time. Device lifetime (the amount of time the µAL remains patent and unobstructed by clot) is used as the primary indicator of biocompatibility. This study is the first study to: 1) investigate the effect of NO release on biocompatibility in a microfluidic network; 2) combine a hydrophilic PEG coating with NO release to improve blood compatibility; and 3) perform extended in vivo biocompatibility testing of a µAL. We found that µALs challenged in vitro with PRP remained patent significantly longer when the sweep gas contained NO than without NO. In the in vivo rabbit model, neither approach alone (PEG coating nor NO sweep gas) significantly improved biocompatibility compared to controls (though with larger sample size significance may become apparent); while the combination of a PEG coating with NO sweep gas resulted in significant improvement of device lifetime. STATEMENT OF SIGNIFICANCE: The development of microfluidic artificial lungs (µALs) can potentially have a massive impact on the treatment of patients with acute and chronic lung impairments. Before these devices can be deployed clinically, the biocompatibility of µALs must be improved and more comprehensively understood. This work explores two strategies for improving biocompatibility, a hydrophilic surface coating (polyethylene glycol) for general surface passivation and the addition of nitric oxide (NO) to the sweep gas to quell platelet and leukocyte activation. These two strategies are investigated separately and as a combined device treatment. Devices are challenged with clottable blood using in vitro testing and in vivo testing in rabbits. This is the first study to our knowledge that allows statistical comparisons of biocompatible µALs in animals, a key step towards eventual clinical use.

Advancing Front Oxygen Transfer Model for the Design of Microchannel Artificial Lungs.

Microchannel artificial lungs may provide highly efficient, long-term respiratory support, but a robust predictive oxygen transfer (VO2) model is needed to better design them. To meet this need, we first investigated the predictive accuracy of Mikic, Benn, and Drinker's advancing front (AF) oxygen transfer theory by applying it to previous microchannel lung studies. Here, the model that included membrane resistance showed no bias toward overprediction or underprediction of VO2 (median error: -1.13%, interquartile range: [-26.9%, 19.2%]) and matched closely with existing theory. Next, this theory was expanded into a general model for investigating a family of designs. The overall model suggests that, for VO2 = 100 ml/min, fraction of delivered oxygen (FDO2) = 40%, wall shear stress ((Equation is included in full-text article.)) = 30 dyn/cm, and blood channel height = 20-50 µm, a compact design can be achieved with priming volume ((Equation is included in full-text article.)) = 5.8-32 ml; however, manifolding may be challenging to satisfy the rigorous total width ((Equation is included in full-text article.)) requirement ((Equation is included in full-text article.)= 76-475 m). In comparison, 100-200 µm heights would yield larger dimensions ((Equation is included in full-text article.)122-478 ml) but simpler manifolding ((Equation is included in full-text article.)4.75-19.0 m). The device size can be further adjusted by varying FDO2, (Equation is included in full-text article.), or VO2. This model may thus serve as a simple yet useful tool to better design microchannel artificial lungs.

Low-Resistance, Concentric-Gated Pediatric Artificial Lung for End-Stage Lung Failure.

Children with end-stage lung failure awaiting lung transplant would benefit from improvements in artificial lung technology allowing for wearable pulmonary support as a bridge-to-transplant therapy. In this work, we designed, fabricated, and tested the Pediatric MLung-a dual-inlet hollow fiber artificial lung based on concentric gating, which has a rated flow of 1 L/min, and a pressure drop of 25 mm Hg at rated flow. This device and future iterations of the current design are designed to relieve pulmonary arterial hypertension, provide pulmonary support, reduce ventilator-associated injury, and allow for more effective therapy of patients with end-stage lung disease, including bridge-to-transplant treatment.

Design Analysis and Optimization of a Single-Layer PDMS Microfluidic Artificial Lung.

OBJECTIVE: Microfluidic artificial lungs (µALs) are being researched for future clinical use due to the potential for increased gas exchange efficiency, small blood contacting surface area, small priming volume, and biomimetic blood flow paths. However, a current roadblock to clinical use is the need to stack hundreds to thousands of these small-scale µALs in parallel to reach clinically relevant blood flows. The need for so many layers not only increases the complexity and projected cost to manufacture a µAL, but also could result in devices which are cumbersome, and, therefore, not suitable for portable artificial lung systems. METHODS: Here, we describe the design analysis and optimization of a single-layer µAL that simultaneously calculates rated blood flow, blood contacting surface area, blood volume, pressure drop, and shear stress as a function of blood channel height using previously developed closed-form mathematical equations. A µAL designed using this procedure is then implemented and tested. RESULTS: The resulting device exhibits a rated flow of 17 mL/min and reduces the number of layers required for clinically relevant µAL devices by a factor of up to 32X compared to previous work. CONCLUSION: This procedure could significantly reduce manufacturing complexity as well as eliminate a barrier to the clinical application of these promising devices. SIGNIFICANCE: The described method results in the highest rated flow for any single-layer µAL to date.

Stability of Polyethylene Glycol and Zwitterionic Surface Modifications in PDMS Microfluidic Flow Chambers.

Blood-material interactions are crucial to the lifetime, safety, and overall success of blood contacting devices. Hydrophilic polymer coatings have been employed to improve device lifetime by shielding blood contacting materials from the natural foreign body response, primarily the intrinsic pathway of the coagulation cascade. These coatings have the ability to repel proteins, cells, bacteria, and other micro-organisms. Coatings are desired to have long-term stability, so that the nonthrombogenic and nonfouling effects gained are long lasting. Unfortunately, there exist limited studies which investigate their stability under dynamic flow conditions as encountered in a physiological setting. In addition, direct comparisons between multiple coatings are lacking in the literature. In this study, we investigate the stability of polyethylene glycol (PEG), zwitterionic sulfobetaine silane (SBSi), and zwitterionic polyethylene glycol sulfobetaine silane (PEG-SBSi) grafted by a room temperature, sequential flow chemistry process on polydimethylsiloxane (PDMS) over time under ambient, static fluid (no flow), and physiologically relevant flow conditions and compare the results to uncoated PDMS controls. PEG, SBSi, and PEG-SBSi coatings maintained contact angles below 20° for up to 35 days under ambient conditions. SBSi and PEG-SBSi showed increased stability and hydrophilicity after 7 days under static conditions. They also retained contact angles ≤40° for all shear rates after 7 days under flow, demonstrating their potential for long-term stability. The effectiveness of the coatings to resist platelet adhesion was also studied under physiological flow conditions. PEG showed a 69% reduction in adhered platelets, PEG-SBSi a significant 80% reduction, and SBSi a significant 96% reduction compared to uncoated control samples, demonstrating their potential applicability for blood contacting applications. In addition, the presented coatings and their stability under shear may be of interest in other applications including marine coatings, lab on a chip devices, and contact lenses, where it is desirable to reduce surface fouling due to proteins, cells, and other organisms.

Characterization of an S-nitroso-N-acetylpenicillamine-based nitric oxide releasing polymer from a translational perspective.

Due to the role of nitric oxide (NO) in regulating a variety of biological functions in humans, numerous studies on different NO releasing/generating materials have been published over the past two decades. Although NO has been demonstrated to be a strong antimicrobial and potent antithrombotic agent, NO-releasing (NOrel) polymers have not reached the clinical setting. While increasing the concentration of the NO donor in the polymer is a common method to prolong the NO-release, this should not be at the cost of mechanical strength or biocompatibility of the original material. In this work, it was shown that the incorporation of S-nitroso-penicillamine (SNAP), an NO donor molecule, into Elast-eon E2As (a copolymer of mixed soft segments of polydimethylsiloxane and poly(hexamethylene oxide)), does not adversely impact the physical and biological attributes of the base polymer. Incorporating 10 wt % of SNAP into E2As reduces the ultimate tensile strength by only 20%. The inclusion of SNAP did not significantly affect the surface chemistry or roughness of E2As polymer. Ultraviolet radiation, ethylene oxide, and hydrogen peroxide vapor sterilization techniques retained approximately 90% of the active SNAP content, where sterilization of these materials did not affect the NO-release profile over an 18 day period. Furthermore, these NOrel materials were shown to be biocompatible with the host tissues as observed through hemocompatibility and cytotoxicity analysis. In addition, the stability of SNAP in E2As was studied under a variety of storage conditions, as they pertain to translational potential of these materials. SNAP-incorporated E2As stored at room temperature for over 6 months retained 87% of its initial SNAP content. Stored and fresh films exhibited similar NO release kinetics over an 18 day period. Combined, the results from this study suggest that SNAP-doped E2As polymer is suitable for commercial biomedical applications due to the reported physical and biological characteristics that are important for commercial and clinical success.

Achieving 12 Hour Normothermic Ex Situ Heart Perfusion: An Experience of 40 Porcine Hearts.

Although total body perfusion with extracorporeal life support (ECLS) can be maintained for weeks, individual organ perfusion beyond 12 hours has yet to be achieved clinically. Normothermic ex situ heart perfusion (ESHP) offers the potential for prolonged cardiac preservation. We developed an ESHP system to study the effect of perfusate variables on organ preservation, with the ultimate goal of extending organ perfusion for ≥24 hours. Forty porcine hearts were perfused for a target of 12 hours. Hearts that maintained electromechanical activity and had a <3× increase in vascular resistance were considered successful preservations. Perfusion variables, metabolic byproducts, and histopathology were monitored and sampled to identify factors associated with preservation failure. Twenty-two of 40 hearts were successfully preserved at 12 hours. Successful 12 hour experiments demonstrated lower potassium (4.3 ± 0.8 vs. 5.0 ± 1.2 mmol/L; p = 0.018) and lactate (3.5 ± 2.8 vs. 4.5 ± 2.9 mmol/L; p = 0.139) levels, and histopathology revealed less tissue damage (p = 0.003) and less weight gain (p = 0.072). Results of these early experiments suggest prolonged ESHP is feasible, and that elevated lactate and potassium levels are associated with organ failure. Further studies are necessary to identify the ideal perfusate for normothermic ESHP.

Reply to the 'Comment on "The promise of microfluidic artificial lungs"' by G. Wagner, A. Kaesler, U. Steinseifer, T. Schmitz-Rode and J. Arens, Lab Chip, 2016, 16.

This response explores and discusses the critiques of Wagner et al. in their "Comment on 'The promise of microfluidic artificial lungs' by Joseph A. Potkay, Lab Chip, 2014, 14, 4122-4138".

Flexible, Structured MWCNT/PDMS Sensor for Chronic Vascular Access Monitoring.

Graft wall pulsation amplitude sensing can provide a measure of functional status, e.g. in hemodialysis access grafts. Current implantable graft monitoring sensors require graft modification and direct bloodstream contact. We propose a new class of piezoresistive flexible pulsation sensors which can be wrapped around the graft to measure wall movement. These sensors must be highly flexible to prevent graft constriction; typical strain sensors are too rigid and the strain sensing range is too limited for this application. We describe a novel additive manufacturing (AM) method for printing polydimethylsiloxane (PDMS) with an internal porous structure, such that material compliance may be tuned anisotropically for a given sensor geometry. Prototype flexible pulsation sensors (FPS) consisting of structured PDMS with an embedded conductive PDMS sensor layer were fabricated and tested. Initial tests demonstrated reliable sensor response to 1-Hz cyclic elongation of 20%, and a sensor gauge factor of 1.0.

The promise of microfluidic artificial lungs.

Microfluidic or microchannel artificial lungs promise to enable a new class of truly portable, therapeutic artificial lungs through feature sizes and blood channel designs that closely mimic those found in their natural counterpart. These new artificial lungs could potentially: 1) have surface areas and priming volumes that are a fraction of current technologies thereby decreasing device size and reducing the foreign body response; 2) contain blood flow networks in which cells and platelets experience pressures, shear stresses, and branching angles that copy those in the human lung thereby improving biocompatibility; 3) operate efficiently with room air, eliminating the need for gas cylinders and complications associated with hyperoxemia; 4) exhibit biomimetic hydraulic resistances, enabling operation with natural pressures and eliminating the need for blood pumps; and, 5) provide increased gas exchange capacity enabling respiratory support for active patients. This manuscript reviews recent research efforts in microfluidic artificial lungs targeted at achieving the advantages above, investigates the ultimate performance and scaling limits of these devices using a proven mathematical model, and discusses the future challenges that must be overcome in order for microfluidic artificial lungs to be applied in the clinic. If all of these promising advantages are realized and the remaining challenges are met, microfluidic artificial lungs could revolutionize the field of pulmonary rehabilitation.

The effects of PEG-based surface modification of PDMS microchannels on long-term hemocompatibility.

The current study demonstrates the first surface modification for poly(dimethylsiloxane) (PDMS) microfluidic networks that displays a long shelf life as well as extended hemocompatibility. Uncoated PDMS microchannel networks rapidly adsorb high levels of fibrinogen in blood contacting applications. Fibrinogen adsorption initiates platelet activation, and causes a rapid increase in pressure across microchannel networks, rendering them useless for long term applications. Here, we describe the modification of sealed PDMS microchannels using an oxygen plasma pretreatment and poly(ethylene glycol) grafting approach. We present results regarding the testing of the coated microchannels after extended periods of aging and blood exposure. Our PEG-grafted channels showed significantly reduced fibrinogen adsorption and platelet adhesion up to 28 days after application, highlighting the stability and functionality of the coating over time. Our coated microchannel networks also displayed a significant reduction in the coagulation response under whole blood flow. Further, pressure across coated microchannel networks took over 16 times longer to double than the uncoated controls. Collectively, our data implies the potential for a coating platform for microfluidic devices in many blood-contacting applications.

A simple, closed-form, mathematical model for gas exchange in microchannel artificial lungs.

Microfabrication techniques are attractive for constructing artificial lungs due to the ability to create features similar in size to those in the natural lung. However, a simple and intuitive mathematical model capable of accurately predicting the gas exchange performance of microchannel artificial lungs does not currently exist. Such a model is critical to understanding and optimizing these devices. Here, we describe a simple, closed-form mathematical model for gas exchange in microchannel artificial lungs and qualify it through application to experimental data from several research groups. We utilize lumped parameters and several assumptions to obtain a closed-form set of equations that describe gas exchange. This work is intended to augment computational models by providing a more intuitive, albeit potentially less accurate, understanding of the operation and trade-offs inherent in microchannel artificial lung devices.

Bio-inspired, efficient, artificial lung employing air as the ventilating gas.

Artificial lungs have recently been utilized to rehabilitate patients suffering from lung diseases. However, significant advances in gas exchange, biocompatibility, and portability are required to realize their full clinical potential. Here, we have focused on the issues of gas exchange and portability and report a small-scale, microfabricated artificial lung that uses new mathematical modeling and a bio-inspired design to achieve oxygen exchange efficiencies much larger than current devices, thereby enabling air to be utilized as the ventilating gas. This advancement eliminates the need for pure oxygen required by conventional artificial lung systems and is achieved through a device with feature sizes and structure similar to that in the natural lung. This advancement represents a significant step towards creating the first truly portable and implantable artificial lung systems for the ambulatory care of patients suffering from lung diseases.

Long term, implantable blood pressure monitoring systems.

An overview of implantable measurement systems suitable for the long-term, continuous monitoring of blood pressure is presented in this paper. The challenges, design considerations and tradeoffs inherent in these systems are overviewed and implantable sensors from both industrial and research environments are reviewed. The paper is concluded with an outlook of future directions for implantable blood pressure monitoring systems.

First-generation hybrid MEMS gas chromatograph.

The fabrication, assembly, and initial testing of a hybrid microfabricated gas chromatograph (microGC) is described. The microGC incorporates capabilities for on-board calibration, sample preconcentration and focused thermal desorption, temperature-programmed separations, and "spectral" detection with an integrated array of microsensors, and is designed for rapid determinations of complex mixtures of environmental contaminants at trace concentrations. Ambient air is used as the carrier gas to avoid the need for on-board gas supplies. The microsystem is plumbed through an etched-Si/glass microfluidic interconnection substrate with fused silica capillaries and employs a miniature commercial pump and valve subsystem for directing sample flow. The latest performance data on each system component are presented followed by first analytical results from the working microsystem. Tradeoffs in system performance as a function of volumetric flow rate are explored. The determination of an 11-vapor mixture of typical indoor air contaminants in less than 90 s is demonstrated with projected detection limits in the low part-per-billion concentration range for a preconcentrated air-sample volume of 0.25 L.