Comparison of pulsatile and nonpulsatile perfusion of the lung in an extracorporeal large animal model

Objective: Extracorporeal lung-perfusion models are widely used to evaluate pulmonary preservation techniques and reperfusion injury. However, these models mainly depend on nonpulsatile flow, which is not physiological and can subsequently lead to pulmonary edema. Observation in a standardized setting and reliability of functional and structural data assessment are therefore limited. To overcome these limitations we developed a new extracorporeal large animal lung perfusion model utilizing pulsatile flow to perfuse the pulmonary vasculature. Methods: Lungs of juvenile domestic pigs were in situ preserved with 2 liters Perfadex(R) and stored for 3 h at 10degreesC. Thereafter, reperfusion of the lung was performed in an extracorporeal blood perfusion circuit employing either a modified roller pump with pulsatile module (300 ml/min; pulsation rate 90/min) or a standardized roller pump with continuous flow (30 ml/min). Ventilation was performed with physiologic room air (350 ml; 16/min) for 1 h. Pulsatile and nonpulsatile perfusion was performed in 2 groups (group NP: nonpulsatile; group P: pulsatile flow, n = 7) during reperfusion. Peak inspiratory pressure (PIP), mean pulmonary artery pressure (PAP), and oxygenation capacity (DeltaPO(2)) were continuously measured. For control of the effectiveness of the pulsatile perfusion pressure waveforms were obtained directly from the native pulmonary artery of both groups. Malondialdehyde (MDA) as a parameter for lipid peroxidation and endothelial cell damage was assessed at 10, 30 and 50 min reperfusion. At the end of the study, pulmonary water content was assessed by means of wet-to-dry ratio (W/D ratio). The tissue was further processed for microscopic analysis. Results: PIP increased significantly in both groups during reperfusion. Mean PAP in both groups increased to 60 mm Hg after 20 min followed by a decrease after 60 min to 40 mm Hg. Pressure waveforms of the pulmonary artery showed sufficient pulsatility in the pulmonary vasculature with a systolic/cliastolic pressure difference of 15 mm Hg whereas the pressure difference was 3-5 mm Hg in the nonpulsatile group. DeltaPO(2) was stable in groups NP and P during reperfusion (30 min: NP: 66.4 (62.2-88) mm Hg; P: 74.8 (65-81.7) mm Hg) without any statistically significant differences between the groups. MDA in group NP decreased over the reperfusion period from 6.2 (3.3-6.3) muM at 10 min to 5.2 (3.2-6.1) muM at 50 min, whereas in group P the level increased and was significantly higher after 50 min reperfusion compared to group NP [6.6 (6.19.2) muM at 50 min; p = 0.016]. W/D ratio was 6.7 (6.3-7.0) in group NP and 6.8 (6.3-7.6) in group P. Light microscopy evaluation showed no differences between both groups regarding severity of intra-alveolar and interstitial edema and numbers of intra-alveolar, intracapillary and interstitial granulocytes. Conclusion: Although effective pulsatile perfusion of the pulmonary vasculature was achieved by means of a modified roller pump, this measure obviously did not improve functional parameters nor did it significantly reduce the edema formation after 3 h ischemia in this extracorporeal lung perfusion model. The use of pulsatile perfusion is therefore not mandatory in the extracorporeal setting of a large animal lung perfusion model. Copyright (C) 2002 S. Karger AG, Basel.