1343267 - In vivo hepatic flow distribution by computational fluid dynamics can predict pulmonary flow distribution in Fontan patients

Background: In Fontan patients, hepatic blood reaches the pulmonary arteries via a surgically implanted conduit. A lung receiving too little hepatic blood due to the initial or post-surgical pulmonary arterial configuration, may develop pulmonary arterial vascular malformations (PAVMs) in the form of intrapulmonary shunts. These shunts may lead to reduced oxygenation, exercise tolerance, and quality of life.

While PAVMs may heal following invasive interventions on the pulmonary arteries aimed at improving the pulmonary hepatic flow distribution, the association between the fraction of hepatic blood to a lung and the amount of shunting and resulting oxygenation is not known.

Therefore, we aimed to establish these relationships to help improve initial surgical designs and provide predictable outcomes of interventions.

Methods:

CMR data from eighteen Fontan patients with total cavo-pulmonary connection (age 3–14 years, 8 females) was used to measure the ratio of right-to-left pulmonary vein flow, and to create computational fluid dynamics (CFD) models used to calculate the fraction of hepatic blood to the left lung. CFD was also used to calculate the patient-specific left-to-right ratio of pulmonary vascular resistance (PVR) required to match the observed pulmonary artery flow distribution.

In a simplified circuit with pulmonary flow and PVR, the fraction of right-to-left pulmonary vein flow was expected to be proportional to the left-to-right fraction of PVR (Figure 1).

Peripheral oxygen saturation at rest at the time of CMR examination, and diagnosis of PAVMs were noted. The fraction of hepatic blood to the left lung was correlated to the right-to-left fraction of pulmonary vein flow and to the left-to-right fraction of PVR using log/linear regression.  The fraction of hepatic blood to the left lung was also correlated to peripheral oxygen saturation at rest using quadratic regression to capture low saturation at the extreme ends of the hepatic flow distribution.

Results: In the studied cohort, a strong relationship was found between the hepatic flow distribution and the right-to-left pulmonary vein flow ratio (r = 0.74; p< 0.001). Also, a strong relationship was found between the hepatic flow distribution and the left-to-right ratio of PVR (r = 0.73; p< 0.001). This also demonstrated the association between pulmonary vein flow and PVR as shown in Figure 1. A strong correlation was also found between the hepatic flow distribution and peripheral oxygen saturation at rest (r = 0.75; p< 0.001). Only two of the patients were diagnosed with PAVMs.

Conclusion:

The distribution of hepatic blood to the lungs has a continuous effect on the pulmonary blood flow, PVR and oxygen saturation, even without clinical manifestations of PAVM’s. These findings may help predict changes in these parameters during planning of surgical or endovascular interventions affecting the hepatic flow distribution in Fontan patients. CMR combined with CFD comprises a powerful tool to discover pathophysiological relationships.