BACKGROUND: Continuous venous air emboli have been detected in the inferior vena cava and smaller veins using transesophageal echocardiography in patients with positive pressure ventilation and associated pulmonary barotrauma. The authors hypothesized that gas entered the venous circulation, following dissection of small vessels at several sites in the subcutaneous or retro-peritoneal soft tissues. OBJECTIVE: The present study was designed to determine if a comparable venous gas embolism occurred in anesthetized dogs, after creation of a pneumomediastinum. DESIGN: Using transesophageal echocardiography, we observed 11 anesthetized dogs mechanically ventilated with positive end-expiratory pressure, while mediastinal air was introduced through a catheter at a rate of 0.5 ml/kg/min. RESULTS: A continuous stream of bubbles appeared in the inferior vena cava in 8/11 dogs (73%) after an infusion period of 280 +/- 81 min. A surge of bubbles was commonly observed following abdominal massage and was often associated with a transient decrease of end-tidal carbon dioxide tensions. In two dogs the air infusion rate was reduced to 0.25 mg/kg/min, and bubbles were detected in the inferior vena cava for as long as 16 consecutive hours. CONCLUSION: We conclude that in anesthetized dogs mechanically ventilated with positive end-expiratory pressure, unremitting pneumomediastinum is usually followed by continuous venous air embolism. A mechanism hypothesized for venous gas entry in the clinical condition of positive end-expiratory pressure ventilation with subcutaneous gas is suggested by this model.
BACKGROUND: Continuous venous air emboli have been detected in the inferior vena cava and smaller veins using transesophageal echocardiography in patients with positive pressure ventilation and associated pulmonary barotrauma. The authors hypothesized that gas entered the venous circulation, following dissection of small vessels at several sites in the subcutaneous or retro-peritoneal soft tissues. OBJECTIVE: The present study was designed to determine if a comparable venous gas embolism occurred in anesthetized dogs, after creation of a pneumomediastinum. DESIGN: Using transesophageal echocardiography, we observed 11 anesthetized dogs mechanically ventilated with positive end-expiratory pressure, while mediastinal air was introduced through a catheter at a rate of 0.5 ml/kg/min. RESULTS: A continuous stream of bubbles appeared in the inferior vena cava in 8/11 dogs (73%) after an infusion period of 280 +/- 81 min. A surge of bubbles was commonly observed following abdominal massage and was often associated with a transient decrease of end-tidal carbon dioxide tensions. In two dogs the air infusion rate was reduced to 0.25 mg/kg/min, and bubbles were detected in the inferior vena cava for as long as 16 consecutive hours. CONCLUSION: We conclude that in anesthetized dogs mechanically ventilated with positive end-expiratory pressure, unremitting pneumomediastinum is usually followed by continuous venous air embolism. A mechanism hypothesized for venous gas entry in the clinical condition of positive end-expiratory pressure ventilation with subcutaneous gas is suggested by this model.