Late Breaking Clinical Trials
Virtual Recording
Sezen Ugan Atik, MD
cardiac Mri fellow
The Hospital for Sick Children
toronto, Ontario, Canada
Sezen Ugan Atik, MD
cardiac Mri fellow
The Hospital for Sick Children
toronto, Ontario, Canada
Osami Honjo,, MD
Head, Div of Cardiovas Surgery
The Hospital for Sick Children, Canada
Mike Seed, MD
staff
The Hospital for Sick Children
toronto, Ontario, Canada
Shi Joon Yoo, MD, PhD
staff
The Hospital for Sick Children
toronto, Ontario, Canada
Israel Valverde, MD, PhD
Pediatric Cardiologist
The Hospital for Sick Children
SickKids
Toronto, Ontario, Canada
Matteo Ponzoni, MD
Clinical fellow
The Hospital for Sick Children, Canada
John Coles
md
the hospital for
toronto, Canada
Rajiv R. Chaturvedi, MD, PhD
md
The Hospital for Sick Children
toronto, Ontario, Canada
anne Dipchand, MD
staff
The Hospital for Sick Children
toronto, Canada
Christopher Z. Lam, MD
Staff Radiologist
The Hospital for Sick Children
Toronto, Ontario, Canada
In adolescent Fontan patients, extracardiac conduits (ECC) may become relatively undersized with growth, contributing to poor outcomes and end-organ sequelae. We evaluated longitudinal ECC geometry by MRI during adolescence and explored associations with clinical and hepatic outcomes.
Methods:
We retrospectively identified 63 pediatric (≤18 y) ECC Fontan patients who underwent fasting, non-sedated, same-day heart–liver MRI at a single tertiary center. Patients with ≥2 MRIs were included; conduit measurements were obtained on every scan. We excluded interrupted IVC–Kawashima physiology and incomplete ECC phase-contrast/geometry. CSA was measured on SSFP images using double-oblique planes orthogonal to the conduit centerline; inner-to-inner diameters were taken at two level: suprahepatic native IVC and visually narrowest segment. Elliptical lumens used π·(AP/2)·(TR/2), circular lumens π·(D/2)² ; CSA was reported in mm². The primary exposure was level-specific change in CSA from first→last MRI. Outcomes included hepatic/portal features (PLE, portal venous/lymphatic congestion scores, splenomegaly/ascites/liver fibrosis) and functional status (VO₂peak %pred, VE/VCO₂ slope). Within-patient change used paired non-parametric tests; between-group comparisons used Mann–Whitney and χ²/Fisher with ORs (95% CI). Associations used Spearman and robust OLS (HC3). Two-sided p< 0.05 was considered significant.
Results:
All baseline demographic and clinical characteristics are summarized in Table 1. At the first MRI, the mean in-vivo ECC lumen CSA at the narrowest-conduit level was smaller than the nominal implanted graft CSA (279 vs 363 mm²; mean Δ −84 mm², −23%), consistent with paired diameter testing (Δ −2.66 mm; p< 0.0001). Longitudinally, CSA increased across conduit levels, with a modest but significant rise at the narrowest-conduit level (median %Δ +17.9%; p< 0.001). When patients were stratified by CSA increase at the narrowest-conduit level (< mean − 2 SD vs others), those below −2 SD showed more adverse hepatic/clinical phenotypes; specifically, PLE patients had systematically lower CSA increase at this level (p=0.004). A higher native-IVC:narrowest-conduit CSA ratio >1.0 (i.e., relatively smaller conduit vs IVC) was associated with liver fibrosis (OR 64.66; 95% CI 3.59–1164.53; p< 0.001) and higher portal lymphatic scores (OR 15.00; 95% CI 0.77–291.85; p=0.022). Functionally, VE/VCO₂ slope worsened as CSA increase became smaller at the narrowest-conduit level (ρ −0.31; p≈0.03), whereas VO₂peak % predicted correlated positively with the native-IVC:narrowest-conduit CSA ratio (ρ +0.29; p=0.041).
Conclusion: Decreases in CSA of Fontan tube were associated with adverse hepatic and clinical phenotypes. These findings support MRI-based surveillance of ECC geometry during adolescence and raise the hypothesis that timely upsizing or stenting of restrictive segments could modify downstream outcomes.
Table 1:Cohort demographic and diagnostic characters Parameter Value Sample size (n) 63 Sex: male (%) 39 (62) Sex: female (%) 24 (38) Age at first MRI (y) — median (IQR) 8.93 (5.19–13.11) Age at last MRI (y) — median (IQR) 16.51 (13.77–17.33) Years between first–last MRI — median (IQR) 5.12 (3.57–8.80) — Diagnoses (n, %) — HLHS 19 (30.2%) Tricuspid atresia 17 (27.0%) DORV 9 (14.3%) DILV 9 (14.3%) Others 9 (14.3%) — Fenestration — Yes 60 (95.2%) No 3 (4.8%) Table 2. Implant vs first in-vivo diameter Level N=63 Implant diameter (mm) — mean First in-vivo diameter (mm) — mean Δ (mm) [95% CI] Paired p</b> ECC–IVC anastomosis 21.15 19.19 -1.96 (-3.05 to -0.87) 0.000824 Mid-segment 21.15 18.20 -2.95 (-3.71 to -2.18) 2.26e-10 PA anastomosis 21.15 19.88 -1.27 (-1.98 to -0.55) 0.0009349