Liver fibrosis marker is a potential predictor of the development of Fontan-associated liver diseases^†

Abstract

OBJECTIVES

To evaluate how well liver fibrosis markers (fibrosis-4 index, aspartate aminotransferase to platelet ratio index, and model for end-stage liver disease excluding international normalized ratio score) can predict early detection of Fontan-associated liver disease and to identify risk factors for Fontan-associated liver disease development.

METHODS

This retrospective multicentre study included patients who underwent the Fontan procedure between 2004 and 2020 with at least 3 years of follow-up. Blood tests and imaging were conducted to diagnose Fontan-associated liver disease. The predictive value of these markers was assessed using receiver operating characteristic curve analysis. Risk factors for Fontan-associated liver disease development were identified using Fine–Gray subdistribution hazard analysis.

RESULTS

This study included 137 patients. The fibrosis-4 index, measured at 2 years post-Fontan, was a strong predictor for Fontan-associated liver disease development 10 years later (area under the curve: 0.81, optimal cutoff value: 0.17, 83.1% sensitivity, and 73.0% specificity). Fine–Gray subdistribution hazard analysis shows that a fibrosis-4 index level was a key risk factor for Fontan-associated liver disease. Patients with a fibrosis-4 index >0.17 after 2 years had a higher incidence of Fontan-associated liver disease after 10 years (45.6%) than patients with fibrosis-4 index ≤0.17 (3.9%, P = 0.002). These patients also had higher pulmonary artery pressure 5 years later.

CONCLUSIONS

The fibrosis-4 may be a useful marker for early detection of Fontan-associated liver disease, which, in this study, was identified as a risk factor for the disease’s development.

CLINICAL REGISTRATION NUMBER

Kitasato University, No. B23-130; 7 February 2024.

Open in new tabDownload slide

INTRODUCTION

Since its introduction in 1971, the Fontan procedure has significantly improved, along with better postoperative care, reducing surgical mortality from 15–30% to <5%, with a 20-year survival rate nearing 85% [1, 2]. Despite improved early and mid-term survival over the past 4 decades, organ complications remain concerning, including plastic bronchitis and protein-losing enteropathy (PLE) [3]. Fontan-associated liver disease (FALD) commonly occurs in patients with over 10 years of follow-up, with incidence rates ranging from 20% to 60% [4, 5]. FALD can cause serious complications like hepatopulmonary syndrome, hepatorenal syndrome, and hepatocellular carcinoma [4], significantly reducing quality of life and becoming life-threatening. Treatment options are limited, with liver transplantation as a possibility, but due to the lack of large-scale studies, the effectiveness and timing of interventions remain unclear [6].

Early detection and intervention are crucial to preventing FALD progression. Managing factors like elevated central venous pressure (CVP), conduit kinking, and potential graft narrowing due to growth through surgical or catheter procedures may slow liver disease progression [1]. Liver fibrosis markers are useful for early detection [7–9], such as the FIB-4 (fibrosis-4) index (age, aspartate aminotransferase (AST), alanine aminotransferase (ALT), and platelet count), APRI (aspartate aminotransferase to platelet ratio index) score (AST and platelet count), and MELD-XI score (bilirubin and creatinine) [10–12]. Though validated in chronic liver disease, their use in Fontan patients remains understudied. Further research is needed to confirm their effectiveness in detecting FALD [10, 13].

This study aimed to evaluate how well liver fibrosis markers predict FALD onset after the Fontan procedure and identify FALD risk factors.

MATERIALS AND METHODS

Ethical statement

This study was approved by the Institutional Review Board of Kitasato University Hospital (approval no. B23-130; 7 February 2024). Informed consent was obtained via an opt-out method on the IRB-approved website.

Study design and objectives

Selection criteria

We reviewed patients who underwent Fontan procedures at 3 institutions (Kitasato University, Jichi Medical University Tochigi Children’s Medical Center, Gunma Children’s Medical Center) between 2004 and 2020. Only patients followed up for at least 3 years after the procedure were included (Fig. 1) [14]. We excluded patients who died from non-liver-related causes within 3 years, patients without laboratory testing within 3 years, and patients who developed liver diseases within 3 years, as the onset was earlier compared to previous studies, suggesting different underlying causes [14–18]. The necessary data were retrieved in an anonymized form from the electronic medical records stored at each hospital. Follow-up data were recorded using the most recent clinical data obtained at the latest visit.

Hepatic assessment

Haemodynamic assessment

The haemodynamic assessment included echocardiography and cardiac catheterization. Echocardiography focused on systemic ventricular contractility and valve regurgitation, with atrioventricular valve regurgitation (AVVR) defined as moderate or greater in severity. In this study, cardiac catheterization was performed as part of routine follow-up care for patients at 1, 5, and 10 years post-Fontan. We believe that this assessment is essential for evaluating the haemodynamic status of Fontan circulation and potential complications after the Fontan procedures. Cardiac catheterization measured key parameters like pulmonary artery pressure (PAP), systemic ventricular end-diastolic pressure, single ventricle ejection fraction (SVEF), pulmonary artery vascular resistance (PVR), pulmonary artery index (PAI, Nakata index), CVP, and cardiac index (CI). Cardiac catheterization was routinely performed pre-Fontan and at 1, 5, and 10 years post-Fontan to track haemodynamic changes over time.

Surgical details of Fontan procedure

Fontan procedures used extracardiac total cavopulmonary connection (TCPC) with expanded polytetrafluoroethylene (ePTFE) grafts except for 1 patient who had a lateral tunnel TCPC. Fifty-eight percent (80/137) received a 4- to 6-mm fenestration between the conduit and atrium. A 16-mm ePTFE graft is preferred for extracardiac TCPC due to its haemodynamic advantages, including lower energy loss and reduced thrombus risk compared to larger conduits [22].

Statistical analysis

Baseline and postoperative variables are presented as median with interquartile range (IQR) for continuous variables and frequency (percentage) for categorical variables. We assessed 2 groups based on the FIB-4 index at 2 years after the Fontan operation: Comparison 1 included group F (FIB-4 index >0.17) and group N (FIB-4 index ≤0.17); Comparison 2 consisted of group H (FIB-4 index >0.20) and group L (FIB-4 index ≤0.20) for Supplementary Data. Fisher’s exact test evaluated categorical differences, while the Mann–Whitney U test was used for continuous variables between the 2 groups. The Fine–Gray method, using R software version 4.3.2 for Windows (Development Core Team), estimated the cumulative incidence of FALD and Fontan-associated events, with death as a competing factor [15]. Additionally, the Fine–Gray model was used to identify risk factors for FALD in univariable and multivariable analyses. Variables with P-values <0.10 in univariable analysis were included in multivariable Fine–Gray models. Follow-up time was estimated using the inverse Kaplan–Meier method, which determines median follow-up as the time point at which 50% of participants remain under observation. The person-time follow-up rate (PTFR) was calculated as the ratio of observed person-time to total person-time assuming no dropouts. Observed person-time was determined by summing individual follow-up times until an end-point was reached (event, dropout or study end), while total person-time assumed all participants were followed for the maximum duration of 10 years [23]. Time-dependent receiver operating characteristic curve (ROC curve) analysis assessed the predictive value of liver fibrosis markers for FALD, with ROC curves and AUC (area under the curve) values calculated via marginal weighting in R. Pairwise comparisons of AUCs were conducted using DeLong’s test with Holm–Bonferroni correction for multiple comparisons. The predictive performance was evaluated using decision curve analysis (DCA) and calibration plots. DCA was conducted using the dcurves package in R, with net benefit calculated across a range of threshold probabilities (0–1). Calibration analysis was performed using the rms package in R to assess the agreement between predicted probabilities and observed outcomes. The calibration curve was generated using bootstrap resampling (1000 repetitions) to minimize overfitting. All other statistical analyses were performed with JMP version 17.0 (SAS Institute, NC, USA), with P < 0.05 considered significant.

RESULTS

We reviewed 151 patients who underwent Fontan procedures at 3 institutions (Kitasato University, 47 patients; Jichi Medical University Tochigi Children’s Medical Center, 53 patients; and Gunma Children’s Medical Center, 51 patients) between 2004 and 2020. Only patients followed up for at least 3 years after the procedure were included (Fig. 1) [14]. We excluded 2 patients who died from non-liver-related causes within 3 years, 10 patients without laboratory testing within 3 years, and 2 patients who developed liver diseases within 3 years, as the onset was earlier compared to previous studies, suggesting different underlying causes [14–18]. Finally, 137 patients were included in this study.

Table 1 summarizes clinical and demographic data. The median (IQR) age at the first Fontan procedure was 2.3 years (1.4–3.6). Of the patients, 60 (43.8%) had right ventricular dominance, 68 (49.6%) had left ventricular (LV) dominance, and 9 (6.6%) had biventricular dominance (Table 1). Most patients (99.3%) underwent extracardiac TCPC; 81 (59.1%) had fenestrated TCPC, with 1 patient (0.7%) undergoing lateral tunnel TCPC. Conduit sizes were 12 mm in 1 patient (0.7%), 14 mm in 5 (3.6%), 16 mm in 117 (85.4%), and 18 mm in 13 (9.5%) (Table 1). Before Fontan completion, 52 patients (38.0%) had pulmonary artery banding, 64 (46.7%) received a Blalock–Thomas–Taussig shunt, 21 (15.3%) underwent the Norwood procedure, and 15 (10.9%) had pulmonary artery plasty (Table 1). Most patients with 12-mm or 14-mm conduits underwent total cavopulmonary shunt (Kawashima procedure) following polysplenia syndrome (4/6).

Cumulative incidence of FALD

The median (IQR) follow-up was 10.3 years (7.3–16.4). The PTFR was calculated as 76.7%. One patient with asplenia died 7.1 years after TCPC due to pulmonary haemorrhage following re-conduit replacement for an infection. FALD incidence post-Fontan was 2.5% at 5 years, 16.3% at 10 years, and 40.4% at 15 years (Fig. 2A), highlighting the rising risk of liver complications over time in Fontan patients.

Hepatic and haemodynamic assessment following the Fontan procedure

Supplementary Material, Table S1 shows the temporal changes in liver function on blood exams. At our institution, routine blood tests are performed annually during the first 3 years post-Fontan and subsequently at 5 and 10 years. Cardiac catheterization is routinely conducted at 1, 5, and 10 years post-Fontan. There were no missing data for blood tests or angiography at 1 and 3 years. However, in 5 years, data were missing for 9 patients (8% of 106), and in 10 years, data were missing for 5 patients (11% of 45). These results confirm that follow-up evaluations were conducted with high accuracy, particularly in the early post-Fontan period. Median (IQR) platelet count decreased from 300 × 10⁹/l (251–358) pre-Fontan to 197 × 10⁹/l (147–239) after 10 years, while γ–GTP and creatinine levels significantly increased post-Fontan (Supplementary Material, Table S1), indicating declining liver and kidney function. The APRI score reached 0.5 at 1 year post-Fontan, suggesting mild liver fibrosis [8], with no further changes. The MELD-XI score remained stable, indicating it was not suitable for follow-up. The FIB-4 score increased from 0.05 (0.03–0.10) preoperatively to 0.25 (0.19–0.34) 5 years post-Fontan, with a continued increase (Supplementary Material, Table S2). Haemodynamic parameters, summarized in Table 2, remained stable throughout.

During the follow-up period, imaging studies were performed for patients who exhibited abnormal blood test results, clinical findings such as hepatomegaly on physical examination, or abnormalities detected on X-rays. Ultrasound was performed in 26 patients (19.0%), enhanced CT scans in 16 patients (11.7%), and MRI in 8 patients (5.8%).

Predictive utility of liver fibrosis markers for FALD development

We conducted a time-dependent ROC analysis to predict FALD 10 years post-Fontan. The FIB-4 index at 2 years had an AUC of 0.81 (optimal cutoff value: 0.17; 83.1% sensitivity and 73.0% specificity) (Fig. 2B). In comparison, the APRI had an AUC of 0.65, and the MELD-XI score had an AUC of 0.49 (Fig. 2B). Pairwise comparisons of AUCs showed statistically significant differences between all markers after Holm–Bonferroni correction (FIB-4 index versus APRI: P = 0.037, FIB-4 index versus MELD-XI score: P < 0.001, APRI versus MELD-XI score: P = 0.037). The Fib-4 index demonstrated the highest predictive performance for FALD in the ROC analysis.

The calibration curves for FIB-4 index, APRI and MELD-XI scores at 2 years were assessed (Supplementary Material, Fig. S1). The FIB-4 index showed moderate calibration with a tendency to underestimate risk at higher predicted probabilities, while the APRI score demonstrated poor calibration, with significant deviations between predicted and observed risks. The MELD-XI score exhibited the poorest calibration among the 3 markers, with substantial discrepancies between predicted and observed risks across the entire range of probabilities. While there is some deviation between predicted and observed probabilities, the errors are relatively small, indicating that the FIB-4 index may be a useful tool for predicting FALD.

The DCA curves demonstrated that across clinically relevant threshold probabilities (0.1–0.5), the FIB-4 index at 2 years consistently showed the highest net benefit, followed by APRI. This suggests that the FIB-4 index could provide the most clinical utility in predicting FALD development (Supplementary Material, Fig. S2).

Analysis of risk factors for FALD development

Fine–Gray subdistribution hazard analysis identified a significant risk factor for FALD development: FIB-4 index (hazard ratio [HR]: 5.19 [95% CI 1.59–16.93, P = 0.006]) (Table 3).

Patients were divided into 2 groups according to the FIB-4 index at 2 years post-Fontan: group F (FIB-4 index >0.17) and group N (FIB-4 index ≤0.17). A comparative analysis showed that the cumulative incidence of FALD was significantly higher in group F than in group N (P = 0.002) (Fig. 3). At 5 years, the cumulative incidence was 2.4% for both groups. At 10 years, it was 45.6% in group F and 3.9% in group N. Additionally, group F showed lower CI and SVEF and higher PAP at 5 years than group N (Table 4). The CI was 3.3 (2.9–3.6) L/min/m² in group F and 3.8 (3.2–4.5) L/min/m² in group N (P = 0.03); SVEF was 62% (58%–64%) in group F and 63% (58%–69%) in group N (P = 0.02); and PAP was 12 (9–14) mmHg in group F and 9 (8–11) mmHg in group N (P = 0.01). Group F tended to have a higher PVR and a lower PAI than group N.

To explore the clinical implications of an alternative cutoff, patients were stratified into 2 groups: those with a marginally elevated FIB-4 index (Group H: FIB-4 index >0.20) and those with a normal or low FIB-4 index (Group L: FIB-4 index ≤0.20) at 2 years post-Fontan. The cumulative incidence of FALD was significantly higher in Group H compared to Group L (P < 0.001) (Supplementary Material, Fig. S3). This finding indicated that patients with a marginally elevated FIB-4 index were at a substantially increased risk for developing FALD. Supplementary Material, Table S3 details the comparison of cardiac catheterization parameters between the 2 groups at 5 years post-Fontan. Group H exhibited higher median values for pulmonary artery (PA) pressure and PA vascular resistance as well as lower CI and EF of single ventricle, suggesting that those with a marginally elevated FIB-4 index may experience worse cardiac function over time.

DISCUSSION

FALD is a common complication in Fontan patients, but its mechanisms and risk factors are not well understood. Establishing definitive diagnostic criteria for FALD has been difficult, requiring various diagnostic modalities [24, 25] without consensus on optimal testing timing. FALD typically appears late, often >10 years post-Fontan, by which time it may have advanced significantly. This study aimed to explore early prediction of FALD using accessible methods like blood tests. We found that the FIB-4 index at 2 years post-Fontan was a highly sensitive and specific predictor of FALD 10 years later. A high FIB-4 index post-Fontan was a significant risk factor for FALD development.

The FIB-4 index, which includes age, AST, ALT, and platelet count, reflects early liver function changes [10, 13]. Although it naturally increases with age, its value at 2 years post-Fontan remains a strong predictor of FALD. AST and ALT indicate liver damage, while lower platelet count suggests portal hypertension [26]. These parameters allow the FIB-4 index to detect early liver issues, enabling timely interventions to improve long-term outcomes. The FIB-4 index is also effective in detecting liver fibrosis in other conditions, like chronic hepatitis [13]. Its non-invasive nature and simple calculation make it practical for regular monitoring, supporting its role in predicting FALD progression in post-Fontan management.

Patients with a FIB-4 index >0.17 (group F) had a significantly higher cumulative incidence of FALD at 5- and 10-year post-Fontan than those with a FIB-4 index ≤0.17 (group N). Group F also exhibited compromised cardiac function, with lower CI, SVEF, and elevated PAP. This likely results from a cycle where impaired cardiac function increases CVP, causing hepatic congestion, which worsens liver and cardiac function. Liver dysfunction can lead to arterio-venous malformations, increasing shunt volume and potentially overloading the heart [27]. As liver fibrosis progresses, it restricts hepatic blood flow, leading to systemic complications, including reduced cardiac function and elevated PAP [28, 29]. Increased portal pressure from fibrosis raises hepatic venous pressure and right heart load [30], contributing to declines in CI, EF, and increased PAP [31]. Chronic hepatic congestion from fibrosis affects cardiac function, showing the connection between liver and heart conditions in Fontan patients [31, 32]. Additionally, systemic inflammation and immune responses from liver damage can worsen these conditions, leading to further declines in CI, SVEF, and increases in PAP [32].

The FIB-4 index, incorporating age, AST, ALT, and platelet count, offers a comprehensive evaluation of liver function and hematological state [10]. Compared to APRI [11], which uses only AST and platelet count, and the MELD-XI score [12], which relies on bilirubin and creatinine [13], the FIB-4 index provides a more accurate reflection of liver fibrosis, making it a superior predictor of FALD in our study. While APRI was designed for chronic hepatitis patients, it has lower sensitivity and specificity than the FIB-4 index [9]. Similarly, MELD-XI, primarily used for assessing liver transplantation needs, is less effective in detecting early fibrosis [33]. In this study, the FIB-4 index showed higher AUC and predictive value than these other markers. Additionally, its non-invasive nature and simple calculation make it practical for early FALD detection and management in Fontan patients.

Despite favourable pre- and post-Fontan haemodynamic (e.g. low pulmonary vascular resistance and minimal AVVR), our study observed the progression of FALD. This finding highlights the complex, multifactorial nature of FALD, suggesting that factors beyond haemodynamic, such as chronic inflammation and microvascular changes, may drive its evolution [7, 16]. Although FALD may be an almost inevitable long-term complication of Fontan procedure, several strategies could help delay its onset or slow progression. These include optimizing haemodynamic by reducing CVP and improving cardiac output, detecting early liver changes through regular imaging and biochemical surveillance, and potentially applying new interventions such as mechanical circulatory support or alternative surgical techniques [16]. Nevertheless, the timing and severity of FALD differ substantially among individuals, emphasizing the need for personalized management and ongoing research into innovative therapeutic approaches.

Recent evidence has highlighted the potential risks associated with smaller conduits in extracardiac TCPC. A study by Kisamori et al. (2024) demonstrated a significantly higher incidence of liver cirrhosis in patients with extracardiac TCPC compared to those with lateral tunnel Fontan procedures, particularly when smaller conduits (e.g. ≤16 mm) were used [34]. These findings suggested that prolonged exposure to Fontan circulation with smaller conduits might exacerbate hepatic venous hypertension and increase the risk of FALD. In our practice, we previously performed extracardiac TCPC at around 1 year of age using 16 mm grafts, based on computational fluid dynamics (CFD) studies indicating that this size could achieve favourable haemodynamics in smaller patients [22]. However, recognizing the long-term risks associated with smaller conduits and prolonged Fontan circulation, we have recently shifted our treatment strategy to delay Fontan completion until patients reach a body size that can accommodate an 18 mm graft. This change aims to optimize Fontan haemodynamic and minimize liver damage during the period before conduit upsizing becomes necessary [34]. This approach aims to optimize early haemodynamics while reducing the risk of FALD during the interval before conduit upsizing becomes necessary. Further studies are needed to evaluate whether delaying Fontan completion or upsizing conduits in adulthood can mitigate the progression of liver disease and improve long-term outcomes for Fontan patients.

A study by Miranda et al. investigated the relationship between liver fibrosis markers and exercise testing results in relation to liver disease progression in patients who underwent the Fontan procedure [35]. Their findings suggested that exercise-induced increases in venous pressure may exacerbate liver disease, establishing a mechanistic link between exercise capacity and liver health. Specifically, their study demonstrated a correlation between worsening exercise test results and the progression of FALD. Although their study focused on patients with a mean age of approximately 30 years, the results imply that incorporating cardiopulmonary exercise testing alongside liver fibrosis markers into routine follow-up could provide clinicians with valuable insights into changes in liver function [36]. For instance, patients with a marginally elevated FIB-4 index at 2 years post-Fontan could benefit from regular exercise testing to monitor changes in haemodynamics and liver function over time in patients undergoing the Fontan procedure. This approach might enable earlier detection of high-risk patients and facilitate timely interventions to prevent FALD progression [17, 35].

Incorporating the FIB-4 index into routine follow-up can facilitate early FALD detection and proactive management. The findings also suggest that other long-term Fontan complications, such as PLE, arrhythmias, and thromboembolic events, could be predicted using similar biomarkers. Further research should validate these markers in larger Fontan populations and explore interventions for identified risks.

Limitations

This study has limitations. First, as a retrospective multicentre study, data were collected from 3 institutions with varying clinical practices and patient management, potentially affecting generalizability. The limited follow-up may also not fully capture the long-term risk of FALD. Second, while the FIB-4 index and other liver fibrosis markers are useful, they do not fully predict FALD, underscoring the need for additional diagnostic tools. Factors such as infections, nutrition, and comorbidities, which influence liver disease progression, were not fully accounted for. Third, FALD severity varies, and accurate grading often requires invasive procedures like liver biopsy, which was not performed. Thus, early detection of FALD was not guaranteed. In addition, imaging modalities such as ultrasound, CT, and MRI were performed selectively based on clinical indications such as abnormal blood tests or hepatomegaly on physical examination. This approach may have led to underdiagnosis of FALD, as some patients with subclinical liver disease might not have undergone imaging. Routine imaging for all Fontan patients could improve the detection of early-stage FALD, but logistical challenges such as sedation requirements for paediatric patients and resource constraints limit its feasibility. Fourth, the calibration analysis revealed a limited range of predicted probabilities (0–0.3) for all liver fibrosis markers, suggesting these models may be more suitable for identifying low- to moderate-risk patients rather than high-risk individuals. This restricted range could be attributed to the relatively low incidence of FALD in the study population, the limited follow-up period, which may not fully capture long-term FALD risk, and potential limitations in the predictive capacity of these markers for FALD in Fontan patients. These findings underscore the need for further refinement of predictive models, particularly for identifying high-risk individuals. Future research should focus on extending the follow-up period, incorporating additional risk factors, and validating these markers in larger, more diverse Fontan populations. Finally, hepatic imaging (ultrasound, CT, MRI) was not routinely scheduled, only being done when liver damage was suspected, possibly underestimating FALD incidence.

CONCLUSIONS

The FIB-4 index at 2 years post-Fontan effectively predicts FALD development. Patients with FIB-4 index >0.17 showed a higher incidence of FALD and signs of declining CI, SVEF, and PAP at 5 years post-procedure. This study suggests that incorporating the FIB-4 index in follow-up protocols for Fontan patients could enable earlier detection and management of FALD.

SUPPLEMENTARY MATERIAL

Supplementary material is available at EJCTS online.

FUNDING

None declared.

Conflict of interest: None declared.

ACKNOWLEDGMENTS

We would like to express our sincere gratitude to Dr Takuya Kawahara at the Clinical Research Promotion Center, The University of Tokyo Hospital, for his invaluable assistance in analysing our research.

DATA AVAILABILITY

The data underlying this article are available in the article. The data underlying this article will be shared on reasonable request to the corresponding author.

Author contributions

Sakura Horie: Data curation; Project administration; Writing—original draft. Fumiaki Shikata: Formal analysis; Project administration; Writing—original draft. Norihiko Oka: Conceptualization; Writing—original draft; Writing—review & editing. Toru Okamura: Conceptualization; Supervision; Writing—review & editing. Yoshikiyo Matsunaga: Data curation; Writing—review & editing. Kenta Matsui: Data curation; Visualization. Tsutomu Hataoka: Data curation; Validation; Visualization. Tadashi Kitamura: Data curation; Writing—review & editing. Masaomi Fukuzumi: Formal analysis; Visualization. Yoichiro Hirata: Data curation; Writing—review & editing. Ryoichi Kondo: Data curation; Writing—review & editing. Kagami Miyaji: Conceptualization; Supervision; Writing—review & editing

Reviewer information

European Journal of Cardio-Thoracic Surgery thanks Julie Cleuziou, Andrew C. Chatzis and the other anonymous reviewers for their contribution to the peer review process of this article.

REFERENCES

ABBREVIATIONS

ABBREVIATIONS
 
• APRI

  Aspartate aminotransferase to platelet ratio index

 
• AUC

  Area under the curve

 
• CI

  Cardiac index

 
• EF

  Ejection fraction

 
• FALD

  Fontan-associated liver disease

 
• FIB-4

  Fibrosis-4

 
• IQR

  Interquartile range

 
• MELD-XI

  Model for end-stage liver disease excluding international normalized ratio

 
• PAP

  Pulmonary artery pressure

 
• PLE

  Protein-losing enteropathy

 
• TCPC

  Total cavopulmonary connection

 
• T-Bil

  Total bilirubin

Author notes