Respiration-Driven Modulations in Fontan Flow Energetics: Impact of Age and General Anesthesia

Background: Flow imaging is essential in the clinical management of patients with Fontan anatomies. 4D flow MRI can detect 3D metrics of flow energetics that are associated with Fontan clinical outcomes, such as kinetic energy (KE) and viscous energy loss (EL)^1–5. Changes in intrathoracic pressure over the respiratory cycle modulate Fontan flow^6,7. General anesthesia (GA) and positive pressure ventilation in young pediatric patients during CMR may further impact these modulations. 4D flow imaging captures only a single respiratory state—end-expiration. The impacts of respiration-driven changes in chest pressure on Fontan flow efficiency, including those modulated by GA/ventilation, are not well understood.

5D flow MRI is a new technique that can capture 3D hemodynamics over both the cardiac and respiratory cycles^8,9. Here, we use 5D flow MRI to measure KE and EL over respiration in adult Fontan patients compared to age-appropriate controls and in pediatric patients with and without GA.

Methods: N=19 Fontan patients and N=9 controls (age 26±6 yrs) were prospectively recruited for 5D flow MRI and were grouped by: Fontan Adults ( > 18 yrs, N=9), Fontan Peds (< 18 yrs, no GA, N=5), and Fontan Peds-GA (< 18 yrs, GA, N=5). 5D flow analysis is detailed in Figure 1. KE and EL were computed in the aorta, pulmonary arteries (PA), and caval veins (SVC, IVC), integrated over the cardiac cycle, and indexed to vessel size.

We evaluated differences in respiratory variability (max % change) of each metric between Fontan and controls using Tukey tests. Alterations in respiration-driven dynamics (% change in each respiratory phase) were identified using repeated measures ANOVA and Tukey tests.

Results: Compared to controls, Fontan patients showed increased respiratory variability in KE in the SVC, IVC, and PAs (50-92% vs. 15-29%, all p< 0.05, Fig. 2a). Fontan Peds-GA showed increased respiratory variability in PA EL (63% vs. 29%, p< 0.05, Fig. 2b). In Fontan Adults, respiratory variability in EL/KE was increased in PAs (61% vs. 27%, p< 0.05, Fig. 2c) and decreased in the aorta (15% vs. 29%, p< 0.05) compared to controls. No significant differences were found between Fontan groups

Compared to controls, Fontan Adults had increased inspiratory KE (+20-38% vs. -6-+2%, p< 0.01, Fig. 3a) and decreased expiratory KE (-36-38% vs. -3-0%, p< 0.01) in the Fontan vessels. EL was also modulated (Fig. 3b), resulting in increased expiratory EL/KE (+26-30% vs. +2-10%, p< 0.01, Fig. 3c). The same effect was seen in Fontan Peds IVC (inspiratory KE +46% p< 0.01, expiratory KE -35% p< 0.05, expiratory EL/KE +52% p< 0.001). Fontan Peds-GA did not show significant differences from controls in particular respiratory phases.

Conclusion: Our results indicate that respiration modulates both adult and pediatric Fontan flow efficiency. Ventilation during GA may disrupt this effect. Capturing flow over the entire respiratory cycle is needed to understand the full range of Fontan function, especially in patients that undergo CMR with GA.

Figure 1: Methods. a) Cohort and Scan Parameters. N=19 Fontan patients were subdivided into 3 groups: Fontan Adult: older than 18 years (N=9). Fontan Ped: Non-anesthetized pediatric patients (N=5). Fontan Ped-GA: General anesthesia pediatric patients (N=5). N=9 adult volunteers were included as controls. b) Data analysis. 5D flow data was reconstructed with four respiratory states from inspiration to expiration. Analysis included 3D segmentation of the aorta, pulmonary arteries (PAs), and caval veins (IVC, SVC) and calculation of total kinetic energy (KE) and total energy loss (EL) over the cardiac cycle indexed to vessel size. Energy loss fraction (EL/KE) was computed as a metric of flow inefficiency.
Figure 2: Respiratory variability in KE, EL, and EL/KE, computed as the 100%*(max-min)/mean. a) KE: Fontan Adults (red) had increased respiratory variability in SVC (50%, p<0.05), IVC (74%, p<0.05), and PA KE (74%, p<0.01) compared to controls (gray, 15-29%). Fontan Peds (black) had increased respiratory variability in IVC KE (92%, p<0.01) compared to controls. Fontan Peds-GA (blue) had increased respiratory variability in SVC (55%, p<0.05) and PA KE (74%, p<0.05) compared to controls. b) EL: Fontan Peds-GA had increased respiratory variability in PA EL (63%, p<0.05) compared to controls (29%). c) EL/KE: Fontan Adults had increased respiratory variability in PA EL/KE (61% p<0.05) and decreased in aorta EL/KE (15% p<0.05) compared to controls (27-29%).
Figure 3: Alterations in respiratory-driven dynamics in KE, EL, and EL/KE. For each respiratory phase, the % change in each metric relative to the mean over all respiratory phases was computed. Tukey tests were performed following . Repeated measures ANOVAs. a) KE: Fontan Adults (red) had increased inspiratory KE in SVC (+20% vs. -6%, p<0.01), IVC (+36% vs. +1%, p<0.01), and PAs (+38% vs. -2%, p<0.001), and decreased expiratory KE in IVC (-38% vs. -2%, p<0.01) and PAs (-36% vs. -0.3%, p<0.01) compared to controls (gray). Fontan Peds (black) had increased inspiratory IVC KE (+46%, p<0.01) and decreased expiratory IVC KE (-35%, p<0.05) compared to controls. b) EL: Fontan Adults had increased inspiratory IVC EL (+5% vs. -21%, p<0.01) and decreased expiratory PA EL (-14% vs. +8%, p<0.05) compared to controls. c) EL/KE: Fontan Adults and Peds had increased expiratory IVC EL/KE (+26%, +52% vs. +2%, p<0.05, p<0.001). Fontan adults had increased expiratory PA EL/KE (+31% vs. +10%, p<0.01).