Background and objectives Carbon monoxide (CO) inhalation is the leading cause of poison‐related deaths in the United States. CO binds to hemoglobin (Hb), displaces oxygen, and reduces oxygen delivery to tissues. The optimal treatment for CO poisoning in patients with normal lung function is the administration of hyperbaric oxygen (HBO). However, hyperbaric chambers are only available in medical centers with specialized equipment, resulting in delayed therapy. Visible light dissociates CO from Hb with minimal effect on oxygen binding. In a previous study, we combined a membrane oxygenator with phototherapy at 623 nm to produce a “mini” photo‐ECMO (extracorporeal membrane oxygenation) device, which improved CO elimination and survival in CO‐poisoned rats. The objective of this study was to develop a larger photo‐ECMO device (“maxi” photo‐ECMO) and to test its ability to remove CO from a porcine model of CO poisoning. Study design/materials and methods The “maxi” photo‐ECMO device and the photo‐ECMO system (six maxi photo‐ECMO devices assembled in parallel), were tested in an in vitro circuit of CO poisoning. To assess the ability of the photo‐ECMO device and the photo‐ECMO system to remove CO from CO‐poisoned blood in vitro, the half‐life of COHb (COHb‐t1/2), as well as the percent COHb reduction in a single blood pass through the device, were assessed. In the in vivo studies, we assessed the COHb‐t1/2 in a CO‐poisoned pig under three conditions: (1) While the pig breathed 100% oxygen through the endotracheal tube; (2) while the pig was connected to the photo‐ECMO system with no light exposure; and (3) while the pig was connected to the photo‐ECMO system, which was exposed to red light. Results The photo‐ECMO device was able to fully oxygenate the blood after a single pass through the device. Compared to ventilation with 100% oxygen alone, illumination with red light together with 100% oxygen was twice as efficient in removing CO from blood. Changes in gas flow rates did not alter CO elimination in one pass through the device. Increases in irradiance up to 214 mW/cm2 were associated with an increased rate of CO elimination. The photo‐ECMO device was effective over a range of blood flow rates and with higher blood flow rates, more CO was eliminated. A photo‐ECMO system composed of six photo‐ECMO devices removed CO faster from CO‐poisoned blood than a single photo‐ECMO device. In a CO‐poisoned pig, the photo‐ECMO system increased the rate of CO elimination without significantly increasing the animal's body temperature or causing hemodynamic instability. Conclusion In this study, we developed a photo‐ECMO system and demonstrated its ability to remove CO from CO‐poisoned 45‐kg pigs. Technical modifications of the photo‐ECMO system, including the development of a compact, portable device, will permit treatment of patients with CO poisoning at the scene of their poisoning, during transit to a local emergency room, and in hospitals that lack HBO facilities.

Background Extracorporeal membrane oxygenators (ECMO) are currently utilized to mechanically ventilate blood when lung or lung and heart function are impaired, like in cases of acute respiratory distress syndrome (ARDS). ARDS can be caused by severe cases of carbon monoxide (CO) inhalation, which is the leading cause of poison‐related deaths in the United States. ECMOs can be further optimized for severe CO inhalation using visible light to photo‐dissociate CO from hemoglobin (Hb). In previous studies, we combined phototherapy with an ECMO to design a photo‐ECMO device, which significantly increased CO elimination and improved survival in CO‐poisoned animal models using light at 460, 523, and 620 nm wavelengths. Light at 620 nm was the most effective in removing CO. Objective The aim of this study is to analyze the light propagation at 460, 523, and 620 nm wavelengths and the 3D blood flow and heating distribution within the photo‐ECMO device that increased CO elimination in CO‐poisoned animal models. Methods Light propagation, blood flow dynamics, and heat diffusion were modeled using the Monte Carlo method and the laminar Navier‐Stokes and heat diffusion equations, respectively. Results Light at 620 nm propagated through the device blood compartment (4 mm), while light at 460 and 523 nm only penetrated 48% to 50% (~2 mm). The blood flow velocity in the blood compartment varied with regions of high (5 mm/s) and low (1 mm/s) velocity, including stagnant flow. The blood temperatures at the device outlet for 460, 523, and 620 nm wavelengths were approximately 26.7°C, 27.4°C, and 20°C, respectively. However, the maximum temperatures within the blood treatment compartment rose to approximately 71°C, 77°C, and 21°C, respectively. Conclusions As the extent of light propagation correlates with efficiency in photodissociation, the light at 620 nm is the optimal wavelength for removing CO from Hb while maintaining blood temperatures below thermal damage. Measuring the inlet and outlet blood temperatures is not enough to avoid unintentional thermal damage by light irradiation. Computational models can help eliminate risks of excessive heating and improve device development by analyzing design modifications that improve blood flow, like suppressing stagnant flow, further increasing the rate of CO elimination.

No abstract