Pulmonary artery banding in sheep: a novel large animal model for congestive hepatopathy

Abstract

Congestive hepatopathy is becoming increasingly recognized among Fontan-palliated patients. Elevated central venous pressure is thought to drive the pathologic progression, characterized by sinusoidal dilatation, congestion, and fibrosis. A clinically relevant large animal model for congestive hepatopathy would provide a valuable platform for researching novel biomarkers, treatment, and prevention. Here, we report on a titratable, sheep pulmonary artery banding model for this disease application. Pulmonary artery banding was achieved by progressively inflating the implanted pulmonary artery cuff. Right ventricular catheter was implanted to draw venous blood samples and measure pressure. The pulmonary artery cuff pressure served as a surrogate for the intensity of pulmonary artery banding and was measured weekly. After about 9 wk, animals were euthanized, and the liver was harvested for histopathological assessment. Nine animal subjects received pulmonary artery banding for 64 ± 8 days. Four of the nine subjects exhibited moderate to severe liver injury, and three of those four exhibited bridging fibrosis. Increasing pulmonary artery cuff pressure significantly correlated with declining mixed venous oxygen saturation (P = 3.29 × 10^-5), and higher congestive hepatic fibrosis score (P = 0.0238), suggesting that pulmonary artery banding strategy can be titrated to achieve right-sided congestion and liver fibrosis. Blood analyses demonstrated an increase in plasma bile acids, aspartate aminotransferase, and γ-glutamyltransferase among subjects with moderate to severe injury, further corroborating liver tissue findings. Our large animal pulmonary artery banding model recapitulates congestive hepatopathy and provides a basis to bridge the current gaps in scientific and clinical understanding about the disease.NEW & NOTEWORTHY We present here a large animal platform for congestive hepatopathy, a disease growing in clinical prevalence due to the increasing number of Fontan-palliated patients. Further data are needed to develop a better clinical management strategy for this poorly characterized patient population. Previous reports of animal models to study this disease have mostly been in small animals with limited fidelity. We show that congestive hepatopathy can be replicated in a chronic, progressive pulmonary artery banding model in sheep. We also show that the banding strategy can be controlled to titrate the level of liver injury. To date, we do not know of any other large animal model that can achieve this level of control over disease phenotype and clinical relevance.

Keywords: Fontan-associated liver disease; congestive hepatopathy; large animal model; liver fibrosis; pulmonary artery banding; sheep.

Graphical abstract

Figure 1.

Overview of sheep pulmonary artery cuff inflation model for inducing congestive hepatopathy, and pulmonary artery cuff pressure area under the curve (AUC) quantification. RV, right ventricle.

Figure 2.

Pulmonary artery banding can be titrated to control the level of liver injury and fibrosis. A: each row of panel shows the animal subject number and its corresponding pulmonary artery (PA) cuff pressure trajectory, area under the curve (AUC), and liver histological profiles stained with H&E and Masson’s Trichrome (Group A: minimal-to-mild, n = 5 subjects; Group B: moderate-to-severe injuries, n = 4 subjects). Yellow star indicates subjects that displayed congestive hepatic fibrosis score of 3 characterized by bridging fibrosis. Scale bar = 100 µm provided on Trial A5 panels. B: AUC between subjects with normal (n = 5) and injured (n = 4) liver profiles. C: AUC between subjects with congestive hepatic fibrosis score of below 3 (n = 6) vs. score with 3 or above (n = 3). Data in B and C are presented as median and interquartile range. Asterisk indicates significant difference between groups (P < 0.05 from Mann–Whitney U test). H&E, hematoxylin and eosin.

Figure 3.

Right-sided congestion and ventricular function during the sheep congestive hepatopathy model in subjects with moderate to severe liver injuries (black) and mild to minimal injuries (white); A: mixed venous oxygen saturation (SvO₂). B: right ventricular (RV) systolic pressure. Data are presented as means ± SD per week per group. n = 1–4 subjects for moderate-to-severe, n = 1–5 subjects for minimal-to-mild. Sample size per week is provided as a table under each figure.

Figure 4.

Longitudinal blood plasma profiles of liver enzymes and bile acids in animal subjects with moderate to severe liver injuries (black) and mild to minimal injuries (white): alanine aminotransferase (ALT, A); aspartate aminotransferase (AST, B); γ-glutamyltransferase (GGT, C). Plasma concentrations of cholic acid (D), taurocholic acid (E), glycocholic acid (F), taurodeoxycholic acid (G), and glycodeoxycholic acid (H). Data are presented as means ± SD per week per group. n = 2–4 subjects for moderate-to-severe, n = 2–5 subjects for minimal-to-mild. Sample size for each week is provided as tables below or next to the corresponding figures.