A miniaturized oxygenator device that is perfused like an artificial placenta via the umbilical vessels may have significant potential to save the lives of newborns with respiratory insufficiency. Recently we presented the concept of an integrated modular lung assist device (LAD) that consists of stacked microfluidic single oxygenator units (SOUs) and demonstrated the technical details and operation of SOU prototypes. In this article, we present a LAD prototype that is designed to accommodate the different needs of term and preterm infants by permitting changing of the number of parallel-stacked microfluidic SOUs according to the actual body weight. The SOUs are made of polydimethylsiloxane, arranged in parallel, and connected though 3D-printed polymeric interconnects to form the LAD. The flow characteristics and the gas exchange properties were tested in vitro using human blood. We found that the pressure drop of the LAD increased linearly with flow rate. Gas exchange rates of 2.4-3.8 μL/min/cm(2) (0.3-0.5 mL/kg/min) and 6.4-10.1 μL/min/cm(2) (0.8-1.3 mL/kg/min) for O2 and CO2 , respectively, were achieved. We also investigated protein adsorption to provide preliminary information on the need for application of anticoagulant coating of LAD materials. Albumin adsorption, as measured by gold staining, showed that surface uptake was evenly distributed and occurred at the monolayer level (>0.2 μg/cm(2) ). Finally, we also tested the LAD under in vivo conditions using a newborn piglet model (body weight 1.65-2.0 kg). First, the effect of an arteriovenous bypass via a carotid artery-to-jugular vein shortcut on heart rate and blood pressure was investigated. Heart rate and mean arterial blood pressure remained stable for extracorporeal flow rates of up to 61 mL/kg/min (101 mL/min). Next, the LAD was connected to umbilical vessels (maximum flow rate of 24 mL/min [10.4 mL/kg/min]), and O2 gas exchange was measured under hypoxic conditions (Fi O2 = 0.15) and was found to be 3.0 μL/min/cm(2) . These results are encouraging and support the feasibility of an artificial placental design for an LAD.