Evaluation and development of a capnodynamic method for estimation of pulmonary blood flow in a porcine model

Abstract: Effective pulmonary blood flow (EPBF) i.e. cardiac output (CO) minus shunt flow could be estimated by a capnodynamic equation. Via introduction of short pauses in the ventilatory pattern the required alterations of carbon dioxide are induced. By integrating the alterations into the equation EPBF could be calculated. The mathematical formula also includes an equation term, effective lung volume (ELV) that has previously been shown to correlate to FRC. The capnodynamic method (COEPBF) was evaluated during haemodynamic and ventilatory alterations in a porcine model before and after lung lavage. An ultrasonic flow probe positioned around the pulmonary trunc (COTS) was used as a reference method for CO. ELV was compared to the sulphur hexafluoride (SF6) method and stability during CO alterations was assessed. Two different ventilation patterns were studied, in papers I-III a pattern based on inspiratory holds and in paper IV a pattern based on expiratory holds. Bland Altman statistics were used for evaluation of the agreement for absolute values and the four-quadrant and polar plot methodologies were used to assess trending ability. COEPBF based on inspiratory holds showed good agreement and trending abilities at PEEP 5 cmH2O. However, a paradoxical increase was seen when PEEP was increased to 12 cmH2O. Lung lavage resulted in a significant decrease in lungfunction with a two-fold increase in shunt fraction and the performance of COEPBF was significantly impaired. However, the trending ability was largely preserved when assessed by the four-quadrant methodology. The shunt fraction was dependent of CO regardless of the degree of lung injury and PEEP level. When shunt levels were > 20%, COEPBF underestimated COTS. ELV was significantly affected by CO alterations at an unchanged PEEP level. A small difference between ELV and FRCPEEP was seen at PEEP 5 cmH2O before lavage and at PEEP 12 cmH2O after lavage. Since a ventilatory pattern based on inspiratory holds is likely to affect the pulmonary capillary blood flow per se a modification of the pattern might improve the performance. A ventilatory pattern based on expiratory holds could be assumed to reduce this variation. In paper IV, when COEPBF was based on expiratory holds, this paradoxical increase in COEPBF was not displayed and the performance was improved. This capnodynamic method could be considered a non-invasive alternative for estimation of EPBF and ELV in mechanically ventilated subjects without significant lung pathology. Considering that the ability to detect trends is good, this device might have the prerequisite to be used to guide goal-directed treatment protocols for CO optimisation.