Unplanned Hospitalization Due to All‐Cause Morbidity and Its Real‐World Management Practices in Patients With Fontan Circulation Authors: Hideo Ohuchi, Kohta Takei, Jun Muneuchi, Shingo Kasahara, Yuichi Ishikawa, Shota Muraji, Masanori Tsukada, Masaki Nii, Shin Ono, Motoki Takamuro, Hirofumi Saiki, Mitsuhiro Fujino, Kenji Kuraishi, Lisheng Lin, Aya Miyazaki, Ichiro Sakamoto, Satoshi Masutani, Yasunobu Hayabuchi, Hiroyuki Ohashi, Kenji Yasuda, Mitsuru Seki, Aki Mori, Michikazu Nakai Published: 2025 ABSTRACT AbstractBackgroundManagement strategies of unplanned hospitalizations (UPH) vary and have not been well characterized in patients with Fontan circulation.MethodsWe prospectively followed 3226 Fontan patients for 3 years to characterize the current UPH status and elucidate the associations between in‐hospital management strategies and postdischarge outcomes.ResultsA total of 243 UPHs occurred (2.5 cases/100 person‐years). Fontan‐related UPHs accounted for 154 cases (63%), most commonly due to heart failure (19%), protein‐losing enteropathy (16%), hemorrhage (14%), arrhythmia (9%), thromboembolism (3%), and plastic bronchitis (2%). Fontan‐unrelated UPHs (89 cases; 37%) were mainly due to infections (19%). Adult patients were more frequently hospitalized for hemorrhage and thromboembolism (P<0.05). The longest length of hospital stay was observed in protein‐losing enteropathy, followed by thromboembolism and heart failure. Among nonpharmacological treatments, oxygen therapy was commonly used, followed by surgical procedures and catheter‐based interventions. Among pharmacological treatments, diuretics were commonly administered, followed by heparin and catecholamines. Of 240 patients discharged alive, 76 (15.4 cases/100 person‐years) were readmitted. Protein‐losing enteropathy was the most frequent cause of readmission, followed by heart failure and hemorrhage. In addition to conventional risk factors, in‐hospital management practices, such as oxygen therapy, surgical intervention, and diuretic adjustment were associated with readmission risk (P<0.01–0.001).ConclusionsWe clarified real‐world practices related to UPHs in Fontan patients. Specific in‐hospital treatment strategies were associated with the risk of readmission. Further studies are warranted to determine the causal relationships between treatment approaches and outcomes for standardized management strategies.Keywords: Fontan, heart failure, hospitalization, management, protein‐losing enteropathySubject Categories: Quality and Outcomes, Complications Nonstandard Abbreviations and Acronyms hypoplastic left heart syndrome protein‐losing enteropathy unplanned hospitalization Clinical Perspective What Is New? We elucidated the full scope and management strategies of all unplanned hospitalizations (UPHs) in Fontan patients, finding that 63% were Fontan related and 37% were not, with thromboembolic and hemorrhagic UPHs significantly more frequent in adults than in pediatric patients and infections commonly observed in both groups. We elucidated the full scope and management strategies of all unplanned hospitalizations (UPHs) in Fontan patients, finding that 63% were Fontan related and 37% were not, with thromboembolic and hemorrhagic UPHs significantly more frequent in adults than in pediatric patients and infections commonly observed in both groups. In addition to patient‐specific risk factors, management strategies during UPH—such as oxygen therapy, nonoral treatments, and diuretic administration—were associated with readmissions after discharge, which often occurred for reasons similar to those of the previous UPH. In addition to patient‐specific risk factors, management strategies during UPH—such as oxygen therapy, nonoral treatments, and diuretic administration—were associated with readmissions after discharge, which often occurred for reasons similar to those of the previous UPH. What Are the Clinical Implications? In addition to traditional risk factors, our findings suggest that management strategies for UPH may be associated with prognosis, highlighting the importance of appropriate UPH management as a significant prognostic factor and the need, from an infection prevention perspective, to raise awareness about vaccination and medication adjustments during UPH while remaining vigilant against clinical inertia. In addition to traditional risk factors, our findings suggest that management strategies for UPH may be associated with prognosis, highlighting the importance of appropriate UPH management as a significant prognostic factor and the need, from an infection prevention perspective, to raise awareness about vaccination and medication adjustments during UPH while remaining vigilant against clinical inertia. The introduction of the Fontan procedure has dramatically improved the prognosis of patients with a single functional ventricle. 1 However, because the Fontan circulation is a unique circulatory system, the long‐term frequency of postoperative unplanned hospitalizations (UPH) and mortality is higher in Fontan patients compared with those with other congenital heart diseases. 2 The primary causes of these UPHs include heart failure (HF), arrhythmias, protein‐losing enteropathy (PLE), thromboembolism, and hemorrhage. Numerous reports have been published regarding the frequency of these complications and associated risk factors. 3 , 4 However, there is little information on the actual clinical management of these UPHs and the resulting outcomes. Therefore, we aimed to clarify the following in this prospective multicenter study: (1) to reassess the full scope of all long‐term UPHs in Fontan survivors, (2) to identify the clinical characteristics and management practices (hospitalization duration, treatment methods) for each category of Fontan‐related UPHs, and (3) to investigate the relationship between the management of UPHs and postdischarge outcomes. The goal of these investigations was to clarify the clinical practices for Fontan patients who require UPH and to establish a foundation for developing future treatment strategies for these patients. METHODS Data Availability The data that support the findings of this study are available from the corresponding author upon reasonable request, subject to institutional and ethical committee approvals. Study Population and Data Collection This was a prospective cohort observational study conducted across 20 institutions in Japan. Participating centers were classified by the number of patients under follow‐up: 6 large centers (≥200 patients), 4 medium‐sized centers (100–199 patients), and 10 small centers (<100 patients). The study population included all patients who required UPH during a 3‐year period from a total cohort of 3226 post‐Fontan patients. The enrollment period spanned from April 1, 2018, to September 29, 2023. Follow‐up after discharge continued until March 31, 2024, with a minimum follow‐up duration of 6 months after discharge. This study was conducted with the approval of the ethics committee at all participating institutions. Written informed consent was obtained from all patients enrolled in the hospitalization registry or from their legal guardians. Clinical data were entered by research collaborators at each institution into a standardized software template (FileMaker), specifically developed for this study. The FileMaker database was distributed to all participating centers. After the follow‐up period (March 31, 2024), it was returned by mail to the principal investigator (H.O.). Collected data were centrally merged and analyzed. Any uncertainties in the data were clarified and corrected through direct communication (eg, email) with each site. The collected data consisted of 3 major components: (1) patient baseline characteristics—including diagnosis, surgical history, hemodynamic data immediately after surgery and at recent follow‐up, presence of major complications (e.g., PLE, pulmonary arteriovenous fistulae, history of arrhythmia), and current medications; (2) clinical data during UPH—including reason for admission, length of hospital stay (LOS, in days), physical examination findings at admission and discharge, ECG results, echocardiographic findings including atrioventricular valve regurgitation, complete blood count, serum biochemistry, and biomarkers such as BNP (brain natriuretic peptide); and (3) postdischarge outcomes – specifically, data on any subsequent unplanned rehospitalizations (re‐UPH), including the reason and date of readmission. Classification of UPH Causes UPH causes were classified into 2 main categories: Fontan‐related and Fontan‐unrelated UPHs (F‐UPH and FU‐UPH, respectively). F‐UPHs were further divided into 6 subcategories based on the primary cause: (1) HF, (2) arrhythmias, (3) PLE, (4) thromboembolism, (5) hemorrhage, and (6) plastic bronchitis. HF was defined as the presence of at least 1 of the following clinical signs: orthopnea, paroxysmal nocturnal dyspnea, pulmonary edema, increasing peripheral edema with or without inappropriate weight gain, or radiographic evidence of pulmonary congestion. In this study, UPHs due to atrioventricular valve regurgitation and hypoxia due to Fontan circulatory failure were also included in the HF category. Arrhythmia‐related UPHs included all rhythm disturbances requiring inpatient treatment, such as atrial tachycardia, atrial fibrillation, sinus node dysfunction, atrioventricular block, and ventricular tachycardia. 5 Device‐related events involving pacemakers were also included. PLE‐related UPHs were defined by the presence of clinical symptoms or hypoalbuminemia in patients with an established diagnosis of PLE. 6 In cases of newly diagnosed PLE, confirmation of protein leakage from the gastrointestinal tract by diagnostic testing was required. Thromboembolism‐related UPHs included symptomatic events confirmed by imaging studies, such as pulmonary embolism and stroke. Hemorrhage‐related UPHs included bleeding events from any organ system requiring admission, including hemoptysis, gastrointestinal bleeding, and gynecological hemorrhages. Hemorrhage was considered an F‐UPH, as it was frequently observed in patients receiving anticoagulation therapy due to problematic Fontan circulation or arrhythmias. Diagnostic confirmation for both thromboembolism and hemorrhage was made using ultrasonography or contrast‐enhanced computed tomography. Plastic bronchitis was defined by the expectoration of bronchial casts. The cause of UPH was determined by the contributing researchers at each participating facility. In cases of multiple concurrent causes, the primary cause was prioritized and the contributing researcher and central adjudication committee (H.O.) made the final decision. FU‐UPHs were divided into 2 subcategories: (1) infectious causes, including infections such as endocarditis and (2) other causes, which included UPHs unrelated to the 6 Fontan‐related subcategories and not caused by infection. These included gastrointestinal conditions (eg, acute appendicitis), central nervous system disorders (eg, seizures), and trauma (eg, fractures). The classification of UPH according to cause was done by the responsible investigator at each participating site. In cases involving multiple coexisting conditions, selecting the principal cause of UPH was at the discretion of the site investigator. In‐Hospital Management and Clinical Course of UPHs In‐Hospital Management To assess the clinical management of UPHs, we analyzed LOS and treatment modalities. Interventions were categorized as nonpharmacological (eg, mechanical ventilation, oxygen therapy, surgery, catheter interventions, and device therapy) or pharmacological. Pharmacological treatments were divided into intravenous (catecholamines, vasoactive agents, diuretics, heparin) and oral drugs (diuretics, angiotensin‐converting enzyme inhibitors [ACEIs] or angiotensin II receptor blockers [ARBs], beta blockers, pulmonary vasodilators). Supplementation therapies included immunoglobulins (IgG, including subcutaneous ones, ie, Hizentra), albumin, red blood cell transfusions, and other blood products. Changes in Clinical Status During UPH To assess the impact of treatment, we evaluated pre‐ and posttreatment changes in clinical parameters, including physical findings (body weight, temperature, heart rate, and blood pressure), laboratory data (biochemical markers), and BNP. Postdischarge Follow‐Up Postdischarge follow‐up focused on the occurrence of re‐UPH. We evaluated the association between re‐UPH and various clinical parameters, including in‐hospital management during the prior UPH. Statistical Analysis Data are expressed as mean±SD or median with interquartile range, as appropriate. For comparisons between 2 independent groups, the unpaired t test or Mann–Whitney U test was used, depending on the distribution of the data. For paired or matched data, the paired t test or Wilcoxon signed‐rank test was applied. Categorical variables were compared using the chi‐square test or Fisher’s exact test, as appropriate. For comparisons among ≥3 groups, the Kruskal–Wallis test followed by Dunn’s post hoc test or 1‐way ANOVA followed by Scheffé’s post hoc test was performed, depending on the data distribution. To evaluate the association between clinical variables and all‐cause UPH, Cox proportional hazards models were used. In this model, hazard ratios (HRs) for systemic ventricular ejection fraction (EF) and atrioventricularvalve regurgitation were calculated using ordinal grading: preserved EF=3, reduced EF=2, poor EF=1; and severe atrioventricular regurgitation=3, moderate=2, mild or less=1. Systemic ventricular morphology was classified as left ventricular type or nonleft ventricular type. Event free from all‐cause UPHs was estimated using the Kaplan–Meier method, and comparisons between groups were performed with the log‐rank test. For the analysis of HRs for clinical variables associated with re‐UPH after discharge, we calculated both univariate HRs and multivariate HRs adjusted for hospital volume (categorized as large, medium sized, or small), sex, and age group (pediatric versus adult). All statistical analyses were conducted using JMP Pro 12 software (SAS Institute Inc., Cary, NC, USA). A P value of <0.05 was considered statistically significant. RESULTS Frequency of UPH by Cause and Clinical Characteristics in the Entire UPH Cohort Among 3226 patients, 243 experienced UPH, yielding an overall incidence of 2.5 events per 100 person‐years. F‐UPHs were significantly more common, with 154 cases (63%, 1.6 per 100 person‐years), compared with 89 FU‐UPH cases (37%, 0.9 per 100 person‐years; P<0.001) (Figure 1A). Comparison of the frequency of Fontan‐related and Fontan‐unrelated unplanned hospitalizations (A); comparison of UPH frequencies between pediatric and adult patients (B); comparison of UPHs due to thromboembolism (C) and hemorrhage (D) between pediatric and adult patients. UPH indicates unplanned hospitalization. Clinical Background Table 1 summarizes the clinical background of the entire UPH cohort, including comparisons between pediatric (< 18 years) and adult patients. In children, the majority had undergone extracardiac conduit total cavopulmonary connection with a higher prevalence of hypoplastic left heart syndrome (HLHS). They underwent the Fontan procedure at a younger age and had higher rates of anticoagulant and ACEI/ARB use. Table 1Clinical Characteristics of Fontan Patients at the Time of First Unplanned Hospitalization During Study PeriodPatientsAllChildrenAdults P valueCases, n24315291‐Age at hospitalization, y18±1011±528±7<0.001Male sex, %5655590.471Body mass index18±517±321±5<0.001Age at first Fontan, y3 (2.1–4.6)2.7 (2.0–3.8)4.4 (2.7–9.3)<0.001Y after first Fontan14±58±522±6<0.001Type of FontanAtriopulmonary connection101‐TCPC (intracardiac rerouting)23617‐TCPC (extracardiac rerouting)21414569‐Bjork202‐Unknown325‐Diagnosis, nHeterotaxy6744230.534Univentricular heart694821‐Hypoplastic left heart syndrome33258‐Tricuspid atresia29920‐Double outlet right ventricle261511‐Mitral atresia17107‐Pulmonary atresia1486‐Atrioventricular discordance1394‐Atrioventricular septal defect1293‐Others301911‐Complications (n, %)Protein‐losing enteropathy54 (22%)31 (20%)23 (25%)0.379Pulmonary arteriovenous fistulae18 (7%)13 (9%)5 (5%)0.361Pacemaker implantation39 (16%)21 (14%)18 (20%)0.225Medications, %Diuretics7174670.270Anticoagulation3743290.026Angiotensin‐converting enzyme inhibitor/angiotensin receptor antagonist5158410.009Beta blocker3534380.440Antiarrhythmia149230.002Pulmonary artery dilator2325210.461Latest hemodynamicsCentral venous pressure, mm Hg12±313±312±30.194Cardiac index, L/min/m2 3.3±1.43.4±1.53.2±1.20.117End‐diastolic volume index of the systemic ventricle, ml/m2 81±2980±2581±320.820Ejection fraction, %55±1055±1154±90.946Atrioventricular valve regurgitation ≥ moderate, %2223200.158Artery oxygen saturation, %91±690±792±40.046Values are mean±SD or median with interquartile range.TCPC indicates total cavopulmonary connection. Clinical Characteristics of Fontan Patients at the Time of First Unplanned Hospitalization During Study Period Values are mean±SD or median with interquartile range. TCPC indicates total cavopulmonary connection. Association Between Clinical Background and Overall UPH We analyzed the relationship between clinical variables (age, sex, primary cardiac anatomy, HLHS, type of Fontan repair, and hospital volume) and F‐UPH versus FU‐UPH. There was no overall difference in UPH rates by age group (P=0.136); however, F‐UPHs were more frequent in adults (HR, 1.47 [95% CI, 1.02–1.94], P=0.036; Figure 1B). Among these, hemorrhage (P=0.023) and thromboembolism (P=0.003) were particularly more common in adults (Figures 1C and 1D). No significant differences were observed between sexes for F‐UPHs or FU‐UPHs. However, PLE was significantly more frequent in male patients (P=0.027). Although there was a slight difference in overall UPH by anatomical diagnosis, no significant difference was observed in F‐UPH. However, FU‐UPHs were more frequent in patients with mitral atresia or HLHS (Figure S1A through S1C), with the latter group showing a significantly higher rate of infection‐related UPHs (P=0.038). There was no difference in the frequency of UPH among different Fontan procedures (Figure S1D through S1F). Regarding hospital size, large‐volume centers had a lower overall UPH rate compared with medium‐ and small‐volume institutions (Figure S2A). The proportion of F‐UPHs was higher at large and medium size‐institutions (P<0.05–0.001) (Figures S2B and S2C), whereas in small centers, no significant difference was observed between F‐UPHs and FU‐UPHs (Figure S2D). Frequency of UPH Categories The frequency of each UPH category is shown in Figures 2A and 2B. Among F‐UPHs, the most frequent cause was HF (n=47, 19%), followed by PLE (n=40, 16%), hemorrhage (n=33, 14%), arrhythmia (n=23, 9%), thromboembolism (n=7, 3%), and plastic bronchitis (n=4, 2%). In contrast, the most frequent causes of FU‐UPHs were infections (n=46, 19%) and other conditions (n=43, 18%). Contrarily, infection was the second most common cause of UPH overall, with the highest frequency in pediatric patients (20%) and the third most common in adults (16%), with no significant difference between the 2 groups (Figure S3). Causes of the first UPH during the study period (A), and cumulative incidence of UPHs shown using Kaplan–Meier curves (B). HF indicates heart failure; PLE, protein‐losing enteropathy; and UPH, unplanned hospitalization. Table 2 compares the frequency (per 100 person‐years), LOS, and clinical background for each pathogenetic category. Among F‐UPH categories, hospitalizations for hemorrhage and thromboembolism were associated with older age. The age at Fontan operation was higher among patients admitted for HF and thromboembolism. Diuretic use was more common in UPHs due to HF and PLE. There were no significant differences in hemodynamics across the pathogenetic categories. Table 2Comparison of Clinical Characteristics Among Fontan Patients Based on the Cause of Unplanned HospitalizationUnplanned hospitalizationFontan‐related P value* Fontan‐unrelatedCausesHeart failurePLEHemorrhageArrhythmiaThromboembolismPBInfectionOthersCases, n47403323744643Incidence, n/100 person‐y0.490.410.340.240.070.04<0.0010.480.44Length of hospital stay, d13 (5–27)18 (9–35)8 (3–28)8 (5–19)17 (4–42)7 (5–58)0.0357 (3–20)5 (3–14)Age at admission, y18 (11–33)14 (8–21)20 (13–25)15 (11–23)24 (20–33)12 (9–17)0.04414 (7–24)14 (9–22)Adult, %433052358600.0423330Male sex, %51734561861000.0825744Body mass index, kg/m2 19±517±419±419±519±318±70.25118±519±4Age at first Fontan, y4 (3–5)3 (2–4)3 (2–5)3 (2–5)5 (3–9)2 (2–2)0.0083 (2–5)3 (2–5)Y after first Fontan14 (9–23)11 (4–19)16 (10–20)13 (8–21)19 (6–26)10 (7–15)0.12211 (5–21)11 (5–21)Diagnosis, %Heterotaxy2815393529250.1792628Left ventricle type systemic ventricle383036305700.6752837Hypoplastic left heart syndrome9203914500.188247Complications, %PLE988991475<0.00177Pulmonary arteriovenous fistulae481800250.05077Pacemaker implantation1113182614250.4241519Medications, %Diuretics79954870671000.0007258Anticoagulation383342262900.0073528Angiotensin‐converting enzyme inhibitor/angiotensin receptor antagonistACEI555848704300.0343953Beta blocker3430394329250.4364035Antiarrhythmia17812351400.436157Pulmonary artery dilator233312001000.5393719Last FontanCentral venous pressure, mm Hg13±312±312±312±310±215±30.47513±312±4Cardiac index, L/min per m2 3.0±1.13.3±1.13.5±1.32.6±0.53.0±0.73.8±0.80.0834.2±2.03.4±1.3End‐diastolic volume index of the systemic ventricleI, mL/m2 89±4073±2380±3095±2459±9570.09277±3080±22Ejection fraction, %54±1154±1055±1052±846580.72558±1055±9Atrioventricular valve regurgitation ≥moderate, %191312131400.451132Artery oxygen saturation, %91 (86–94)94 (88–96)93 (88–94)94 (90–95)92 (87–95)88 (85–92)0.54293 (89–94)93 (89–94)PB, plastic bronchitis; PLE, protein‐losing enteropathy. Values are mean±SD or median with interquartile range.* P values are calculated among all groups without those with plastic bronchitis due to the small sample size (n=4). Comparison of Clinical Characteristics Among Fontan Patients Based on the Cause of Unplanned Hospitalization PB, plastic bronchitis; PLE, protein‐losing enteropathy. Values are mean±SD or median with interquartile range. P values are calculated among all groups without those with plastic bronchitis due to the small sample size (n=4). Cause by UPH Category Detailed causes for each UPH category are illustrated in Figure 3. Detailed breakdown of the causes of the first unplanned hospitalization by category: heart failure (A), hemorrhage (B), arrhythmia (C), thromboembolism (D), infection (E), and other causes (F). AVB indicates atrioventricular block; AVVR, atrioventricular valve regurgitation; CRAO, central retinal artery occlusion; GI, gastrointestinal; IE, infective endocarditis; PA, pulmonary artery; PM, pacemaker; RV, right ventricle; TIA, transient ischemic attack; URI, upper respiratory infection; and VA, ventricular arrhythmia. Fontan‐Related UPH HF: The most common underlying issues were weight gain and peripheral edema (64%), followed by hypoxemia (21%) and worsening atrioventricular valve regurgitation (9%). Hemorrhage: Hemoptysis was the most frequent bleeding source (36%), followed by gastrointestinal (30%) and ovarian bleeding (12%). Arrhythmia: Atrial tachycardia or atrial flutter was most common (61%), followed by sinus node dysfunction (13%). Thromboembolism: There were 7 cases of thromboembolism, including 2 cases of pulmonary embolism; the remaining 5 cases involved other thromboembolism events. Non‐Fontan‐Related UPH Infection: Infection was the second most common cause of UPH overall. It was the most frequent cause in pediatric patients (20%) and the third most common in adults (16%), with no significant age‐related differences. The most common type was pneumonia (30%). There were 4 cases (9%, including 1 pediatric case) of COVID‐19–related hospitalizations: 2 due to infection 2 two due to vaccine‐related adverse events, including 1 case of myocarditis. Other causes: Gastrointestinal conditions, including acute appendicitis and abdominal pain, were the most common (19%), followed by seizures (14%) and trauma (12%). Length of Hospital Stay and Treatment Modalities Length of Stay LOS is summarized in Table 2. The median LOS for all UPH cases was 9 days (4–26 days). F‐UPHs had a significantly longer LOS (median 12 [5–28] days) compared with FU‐UPHs (7 3 , 4 , 5 , 6 , 7 , 8 , 9 , 10 , 11 , 12 , 13 , 14 , 15 , 16 days, P=0.002). Three in‐hospital deaths were recorded: 2 due to HF and 1 due to pacemaker‐related infection. Figure 4 shows LOS by hospital and UPH category, with PLE having the longest and other causes the shortest LOSs. Length of hospital stay for the first unplanned hospitalization by underlying cause. The vertical axis is shown on a logarithmic scale. PLE indicates protein‐losing enteropathy. No significant differences in LOS were observed by age or sex. However, patients with HLHS had significantly longer LOS compared with those without HLHS (19 [7–42] versus 9 [4–23] days, P<0.01), and admissions at large‐volume centers had significantly longer LOS compared with medium or small centers (17 [7–35] versus 6 [3–15] days, P<0.001). Among F‐UPH, PLE was associated with the longest median LOS, at 18 days (9–35), followed by UPHs due to thromboembolism, with a median LOS of 17 days, and HF, with a median LOS of 13 days. In contrast, for FU‐UPH, the median LOS was shorter, with infections accounting for a median of 7 days and other causes resulting in a median LOS of 5 days. Treatment Modalities Treatment strategies, including nonpharmacological interventions, intravenous, and subcutaneous therapies, are presented in Tables 3, 4, and 5. Table 3Nonpharmacological Managements of First Fontan‐Related and ‐Unrelated Unplanned HospitalizationCausesSurgery or endoscopyPMIIVNMechanical ventilationO2 NONo.(n, %)(n, %)(n, %)(n, %)(n, %)(n, %)All UPHs24333 (14%)10 (4%)24 (10%)10 (4%)93 (38%)5 (2%)Fontan related15417 (11%)10 (6%)23 (15%)7 (5%)63 (41%)3 (2%)Heart failure4710 (21%)3 (6%)13 (28%)2 (4%)19 (40%)2 (4%)Protein‐losing enteropathy402 (5%)05 (13%)1 (3%)21 (53%)0Hemorrhage333 (9%)04 (12%)3 (9%)12 (36%)0Arrhythmia234 (9%)6 (26%)005 (22%)1 (4%)Thromboembolism75 (9%)01 (14%)03 (43%)0Plastic bronchitis46 (9%)1 (25%)01 (25%)3 (75%)0Fontan unrelated8916 (18%)01 (1%)3 (4%)30 (34%)2 (2%)Infection464 (9%)002 (4%)23 (50%)2 (4%)Others4312 (28%)01 (2%)1 (2%)7 (16%)0IVN indicates catheter intervention; NO, nitric oxide; O2, oxygen inhalation; PMI, pacemaker implantation; and UPH, unplanned hospitalization. Values are median with interquartile range. Nonpharmacological Managements of First Fontan‐Related and ‐Unrelated Unplanned Hospitalization IVN indicates catheter intervention; NO, nitric oxide; O2, oxygen inhalation; PMI, pacemaker implantation; and UPH, unplanned hospitalization. Values are median with interquartile range. Table 4Pharmacological Managements of First Fontan‐Related and ‐Unrelated Unplanned HospitalizationCausesNo.Intravenous administrationCatecholamineInodilatorDiureticsHeparinSteroidNo.DurationNo.DurationNo.DurationNo.DurationNo.(rate, %)(d, IQR)(rate, %)(d, IQR)(rate, %)(d, IQR)(rate, %)(d, IQR)(rate, %)All unplanned hospitalizations24322 (9%)10 (5–24)20 (8%)8 (5–15)68 (28%)8 (4–15)26 (11%)18 (8–33)5 (2%)Fontan related15416 (10%)11 (6–29)18 (12%)8 (5–14)55 (36%)8 (4–15)26 (17%)18 (8–33)4 (3%)Heart failure478 (17%)10 (5–17)12 (26%)6 (3–9)25 (53%)6 (2–11)3 (6%)7 (5–77)1 (2%)Protein‐losing enteropathy404 (10%)16 (7–47)5 (13%)9 (6–20)22 (55%)8 (6–14)19 (48%)23 (14–34)2 (5%)Hemorrhage331 (3%)400…5 (15%)11 (7–39)0…0Arrhythmia232 (9%)91 (4%)122 (9%)15 (7–39)0…0Thromboembolism71 (14%)…0…0 (0%)…3 (43%)10 (4–16)0Plastic bronchitis40…0…1 (25%)…1 (25%)51 (25%)Fontan unrelated896 (7%)7 (3–20)2 (2%)1513 (15%)13 (4–30)0…1 (1%)Infection462 (4%)5 (5–5)1 (2%)215 (11%)15 (5–44)0…0Others434 (9%)9 (2–23)1 (2%)58 (19%)12 (2–15)0…1 (2%)IQR indicates interquartile range. Pharmacological Managements of First Fontan‐Related and ‐Unrelated Unplanned Hospitalization IQR indicates interquartile range. Table 5Supplementation for First Fontan‐Related and Unrelated Unplanned HospitalizationCausesNo.Intravenous or subcutaneous supplementationAlbumin (intravenous)IgG (intravenous or subcutaneous)RBCOthersAll unplanned hospitalizations24344 (18%)46 (19%)37 (15%)15 (6%)Fontan related15438 (25%)43 (28%)31 (20%)13 (8%)Heart failure4711 (23%)9 (19%)9 (19%)5 (11%)Protein‐losing enteropathy4023 (58%)30 (75%)8 (20%)5 (13%)Hemorrhage331 (3%)1 (3%)11 (33%)1 (3%)Arrhythmia232 (9%)2 (9%)2 (9%)2 (9%)Thromboembolism7001 (14%)0Plastic bronchitis41 (25%)1 (25%)00Fontan unrelated896 (7%)3 (3%)6 (7%)2 (2%)Infection464 (9%)3 (7%)2 (4%)1 (2%)Others432 (5%)04 (9%)1 (2%)Others include fresh frozen plasma and platelet transfusion. RBC indicates red blood cell transfusion. Supplementation for First Fontan‐Related and Unrelated Unplanned Hospitalization Others include fresh frozen plasma and platelet transfusion. RBC indicates red blood cell transfusion. All UPHs Oxygen therapy was the most frequently employed nonpharmacological intervention, used in 38% of cases, followed by surgical procedures in 14% and catheter‐based interventions in 10%. Regarding pharmacological treatment, diuretics were the most administered in 28% of cases, with heparin and catecholamines used in 11% and 9%, respectively. As for supplemental therapies, intravenous IgG was the most frequently administered (19%), followed by albumin (18%) and red blood cell transfusions (15%). The duration of these therapies is depicted in Figure 5, which shows that heparin had the longest median treatment duration, followed by oxygen therapy and catecholamines. Duration of treatment by therapeutic modality during the first unplanned hospitalization. The vertical axis is shown on a logarithmic scale. ANP indicates atrial natriuretic peptide, PDEi, phosphodiesterase inhibitor. Fontan‐Related UPHs For UPH patients with HF, nonpharmacological treatments frequently included oxygen therapy used in 40% of cases, catheter‐based interventions in 28%, and surgical procedures in 21%. The surgical treatments primarily consisted of atrioventricular valve repair and thoracoabdominal drainage, each accounting for 4 cases and comprising 80% of the surgical group. Catheter interventions typically involved collateral coil closure and pulmonary artery dilation. Pharmacological therapies were dominated by diuretics, used in 53% of cases, followed by vasopressors in 26% and catecholamines in 17%. Among supplemental therapies, albumin was administered in 23% of cases, red blood cell transfusions in 19%, and intravenous IgG in 19%. In PLE, the most common nonpharmacological intervention was oxygen therapy, used in 53% of cases, followed by catheter interventions in 13%. The catheter‐based procedures were predominantly fenestration creation and pulmonary artery dilation, together accounting for 80% of cases. Pharmacological treatment primarily involved diuretics (55%) and heparin (48%). Supplemental therapies included IgG in 75% of cases and albumin in 58%, whereas steroid use was observed in only 5%. For UPHs due to hemorrhage, oxygen therapy was administered in 36% of cases, with catheter interventions applied in 12%. Diuretics were used in 15%, and red blood cell transfusions were the most common supplemental treatment, given in 33%. Among patients with UPH for arrhythmia, pacemaker therapy was implemented in 26% of cases, and electrical cardioversion in 13%. Oxygen therapy was used in 22%. Pharmacologically, both diuretics and catecholamines were used in 9%. Supplemental treatments—including albumin, IgG, and red blood cell transfusions—were each used in 9%. For those UPHs with thromboembolism, oxygen therapy was used in 43%. Heparin was the primary pharmacological treatment, also used in 43%. One case required red blood cell transfusion. In patients with plastic bronchitis, oxygen therapy was the most common nonpharmacological treatment, administered in 75%. Mechanical ventilation, surgery, pacemaker intervention, diuretics, heparin, and steroid therapy were used in 25% each. Fontan Unrelated‐UPSs Infections were most frequently managed with oxygen therapy (34%). A subset of these patients also required intravenous medications or supplemental therapies. For other FU‐UPHs, surgical or endoscopic interventions were common, occurring in 28% of patients. Oral Medication Adjustments During UPH Adjustments in the dosage of oral medications during UPH—including diuretics, beta blockers, ACEIs or ARBs, and pulmonary vasodilators—are summarized in Table 6. Across all cases, although some patients experienced changes in medication dosing, the majority (60%–80%) had no modifications to their prescribed oral medications during their first UPH. Table 6Managements With Oral Medications During First Fontan‐Related and Unrelated Unplanned HospitalizationCases*Diureticsß blockersACEI/ARBPulmonary artery dilatorsNumber of patients prescribed, n (%)194 (80%)93 (38%)127 (52%)115 (47%)Cases at admission, n17791114115New prescription, n172130Oral dosage adjustmentIncreaseNo changeDecreaseIncreaseNo changeDecreaseIncreaseNo changeDecreaseIncreaseNo changeDecrease(n, %)(n, %)(n, %)(n, %)(n, %)(n, %)(n, %)(n, %)(n, %)(n, %)(n, %)(n, %)All unplanned hospitalizations24358 (30%)120 (62%)16 (8%)14 (15%)68 (73%)11 (12%)15 (12%)94 (74%)18 (14%)7 (6%)97 (84%)11 (10%)Fontan related15444 (35%)73 (57%)10 (8%)14 (23%)41 (67%)6 (10%)14 (16%)59 (69%)13 (15%)6 (8%)58 (79%)9 (12%)Heart failure4718 (42%)20 (47%)5 (11%)7 (37%)10 (53%)2 (10%)4 (15%)15 (58%)7 (27%)4 (15%)15 (70%)4 (15%)Protein‐losing enteropathy4015 (38%)22 (55%)3 (7%)2 (17%)10 (83%)02 (9%)19 (82%)2 (9%)020 (91%)2 (9%)Hemorrhage337 (37%)11 (58%)1 (5%)011 (85%)2 (15%)1 (6%)14 (88%)1 (6%)1 (8%)9 (75%)2 (17%)Arrhythmia231 (6%)15 (88%)1 (6%)5 (36%)7 (50%)2 (14%)6 (35%)8 (47%)3 (18%)1 (13%)6 (74%)1 (13%)Thromboembolism72 (50%)2 (50%)002 (100%)01 (25%)3 (75%)001 (100%)0Plastic bronchitis41 (25%)3 (75%)001 (100%)000004 (100%)0Fontan unrelated8914 (21%)47 (70%)6 (9%)027 (84%)5 (16%)1 (2%)35 (85%)5 (12%)1 (2%)39 (93%)2 (5%)Infection468 (22%)26 (70%)3 (8%)015 (88%)2 (12%)1 (6%)16 (89%)1 (6%)1 (4%)27 (96%)0Others436 (20%)21 (70%)3 (10%)012 (80%)3 (20%)019 (83%)4 (17%)012 (86%)2 (14%)PLE=, Cases*=number of patients who required UPH for each cause. ACEI/ARB indicates angiotensin‐converting enzyme inhibitor/angiotensin receptor blocker. Managements With Oral Medications During First Fontan‐Related and Unrelated Unplanned Hospitalization PLE=, Cases*=number of patients who required UPH for each cause. ACEI/ARB indicates angiotensin‐converting enzyme inhibitor/angiotensin receptor blocker. In F‐UPHs, among patients admitted for HF, both diuretics and beta blockers were frequently uptitrated, whereas ACEIs/ARBs were commonly downtitrated. In PLE, diuretics were often increased, whereas other medications remained unchanged. Similarly, in hemorrhage‐related UPHs, diuretics were commonly uptitrated, but changes in other medications were infrequent. Patients admitted for arrhythmia frequently had their beta blockers and ACEIs/ARBs increased. In thromboembolism‐related UPHs, diuretics were uptitrated in ≈50%, with no observed changes in other medications. Among patients with plastic bronchitis, 25% experienced an increase in diuretic dosage. In contrast, medication adjustments were generally less frequent in FU‐UPHs, reflecting a lower clinical need for therapeutic escalation or modification in this population. Association Between In‐Hospital Management and Postdischarge Outcomes Among the 240 patients who survived to be discharged, the median follow‐up period was 2.0 years (range 0.8–3.1 years), during which 76 patients were readmitted, and 9 of these readmissions resulted in death. The overall rate of recurrent UPHs (re‐UPH) in patients with a history of UPH was 15.4 per 100 patient‐years, which is 6.1 times higher than that of prior UPH. The frequency of Fontan‐related readmissions (F‐re‐UPHs) was 56 cases (74%), showing a trend toward an increase compared with prior F‐UPH (63%) (P=0.093). Among F‐re‐UPHs, the most frequent cause was PLE, accounting for 26 cases, followed by HF in 12 cases, hemorrhage in 10 cases, arrhythmia in 6 cases, and both thromboembolism and plastic bronchitis in 1 case each, as illustrated in Figure 6. For FU‐re‐UPHs, readmissions were equally attributed to infections and other causes, with 10 cases each, also shown in Figure 6. Causes of re‐UPH after discharge (A), and cumulative incidence of re‐UPHs shown using Kaplan–Meier curves (B). HF indicates heart failure; PLE, protein‐losing enteropathy; re‐UPH, unplanned rehospitalization; and UPH, unplanned hospitalization. Characteristics of Re‐UPH Causes Many re‐UPH cases were attributable to the same causes as the prior UPH. Among F‐re‐UPHs, 73% of PLE readmissions (19 cases), 50% of HF readmissions (6 cases), 50% of hemorrhagic readmissions (5 cases), 33% of arrhythmia readmissions (2 cases), and 100% of plastic bronchitis readmissions (1 case) occurred for the same reasons as the prior UPH. In FU‐re‐UPHs, 33% of infection readmissions (4 cases) and 75% of readmissions for other causes (6 cases) were also due to the same reasons as the prior UPH. Association Between Clinical Characteristics and Re‐UPH There were no significant differences in the frequency of re‐UPH based on the presence of heterotaxy, ventricular morphology (left ventricular type versus nonleft ventricular type), surgical technique (extracardiac versus intracardiac), age, sex, presence of HLHS, or among the 3 categories of institutional size. However, in contrast to the prior UPH, re‐UPH occurred significantly more frequently in patients followed at large institutions compared with those at medium or small institutions (HR, 1.57 [95% CI, 1.00–2.47]; P=0.049) (Figure S4A). Associations between conventional and prior UPH‐related risk factors and the occurrence of re‐UPH, both before and after adjustment for institutional size, sex, and age group (pediatric versus adult), are presented in Table 7 (Fontan‐related) and in Table S1 (Fontan‐unrelated). Table 7Associations of Conventional and First Unplanned Hospitalization‐Related Risk Factors With the Second Unplanned HospitalizationVariablesCasesFontan‐related re‐UPHUnivariate (unadjusted)Multivariate (adjusted)* HR95% CI P valueHR95% CI P valueRisk factors associated with first UPHManagement during first UPHO2 administration2402.141.27–3.620.0052.001.17–3.450.012Surgery/endoscopy/pacemaker implantation2400.330.10–1.040.0590.310.09–0.990.048Diuretics (intravenous)2402.571.52–4.360.0012.611.52–4.490.001Albumin (intravenous)2402.351.31–4.190.0072.411.33–4.390.004IgG (intravenous or intracoronary)2403.562.08–6.10<0.0012.891.58–5.310.001Changes in diuretic dose during first UPHIncrease vs patients without diuretics2404.001.14–14.30.0334.511.27–16.10.020No change vs patients without diuretics2406.301.95–20.40.0026.962.14–22.60.001Changes in biomarkers during first UPHPlatelet, per μL1700.950.91–0.990.0210.950.91–0.990.026Albumin, per 0.1 g/dL1651.071.03–1.11<0.0011.061.02–1.100.003Conventional risk factors at dischargeComplications at dischargeProtein‐losing enteropathy2403.532.09–5.97<0.0013.612.09–6.25<0.001Pulmonary arteriovenous fistulae2393.571.74–7.320.0033.031.44–6.400.004Recent hemodynamicsCentral venous pressure, mm Hg2011.101.00–1.210.0431.131.03–1.240.011Cardiac index, L/min per m2 1901.221.01–1.440.0441.211.01–1.420.037Diastolic blood pressure, mm Hg2070.970.94–0.990.0170.970.95–1.000.063Oral medications at dischargeDiuretics2405.311.66–17.0<0.0016.171.92–19.90.002Angiotensin‐converting enzyme inhibitor/angiotensin receptor blocker2400.560.32–0.960.0320.570.33–0.980.043Peripheral blood profileWhite blood cell count, ×103/μL1700.840.73–0.950.0050.850.74–0.960.006Lymphocyte, %1680.960.93–0.990.0110.970.93–0.990.016Liver variableAlbumin, per 0.1 g/dL1660.950.92–0.990.0250.930.89–0.980.003HR indicates hazard ratio; re‐UPH, unplanned rehospitalization; and UPH, unplanned hospitalization.*Adjusted with sex, age (child, adult) and hospital scale (large, middle sized, small). Associations of Conventional and First Unplanned Hospitalization‐Related Risk Factors With the Second Unplanned Hospitalization HR indicates hazard ratio; re‐UPH, unplanned rehospitalization; and UPH, unplanned hospitalization. Adjusted with sex, age (child, adult) and hospital scale (large, middle sized, small). Association Between Prior UPH Management and Re‐UPH In F‐re‐UPH, oxygen administration was associated with a higher risk of re‐UPH, whereas surgical or endoscopic interventions, including pacemaker implantation, were associated with a lower risk (Figures 7A and 7B). Patients receiving oxygen therapy exhibited more severe hypoxemia at the time of the prior UPH (89±6% versus 92±6%, P<0.01), higher logarithmically expressed BNP (Ln BNP) levels (4.1±1.5 versus 3.2±1.3, P<0.001), and lower EF (2.6±0.6 versus 2.8±0.5, P<0.05). The use of intravenous and oral diuretics during prior UPH was strongly associated with subsequent re‐UPH (Figures 7C and 7D, respectively), and intravenous diuretic therapy was associated with higher Ln BNP levels (4.2±1.7 versus 3.3±1.2, P<0.001). Patients whose oral diuretic dosage remained unchanged or was increased during prior UPH had a 6.3‐fold and 3.9‐fold higher risk of re‐UPH, respectively, compared with those not receiving diuretics (P<0.05–0.01, Figure 7D). Adjustments of other medications during prior UPH were not associated with re‐UPH. In an analysis excluding patients with PLE, oxygen therapy and unchanged diuretic dosage during prior UPH remained significantly associated with an increased risk of F‐re‐UPH (P<0.05 for both). These associations remained consistent even after adjustment for institutional size, sex, and age (pediatric versus adult) (Table 7). In FU‐re‐UPH, an increased dosage of oral diuretics showed a trend toward association with re‐UPH (P=0.06). This remained consistent even after adjustment for the preceding 3 factors (Table S1). Comparisons of the frequency of unplanned rehospitalization after discharge in patients with and without the use of oxygen therapy (A), surgical/endoscopic/pacemaker treatments (B), intravenous diuretic therapy (C), changes in diuretic doses during prior unplanned hospitalization (D), and the use of diuretics (E) and ACEI/ARB (F) at discharge, are shown using Kaplan–Meier curves. ACEI/ARB indicates angiotensin‐converting enzyme inhibitor/angiotensin receptor blocker; IV, intravenous; PMI, pacemaker implantation; and re‐UPH, unplanned rehospitalization. Association Between Conventional Risk Factors and Re‐UPH In F‐re‐UPH, the presence of PLE, pulmonary arteriovenous fistulae, elevated central venous pressure, high cardiac output, and low arterial diastolic pressure were significantly associated with an increased risk of re‐UPH. At discharge, the use of oral diuretics was associated with a 5.3‐fold higher risk of re‐UPH (P<0.001, Figure 7E), whereas patients receiving ACEI or ARB had a 43% lower risk compared with those not receiving these medications (P<0.05, Figure 7F). These statistically significant associations remained unchanged after adjustment for institutional size, sex, and age group (pediatric versus adult) (Table 7). When patients with PLE were excluded, the presence of pulmonary arteriovenous fistulae, elevated central venous pressure and cardiac output, and the use of diuretics remained significantly associated with a higher risk of re‐UPH (P<0.05–0.01). Association Between Changes in Clinical Findings During Prior UPH and Re‐UPH Significant changes in clinical parameters during prior UPH are summarized in Table 8. Body weight, heart rate, and blood pressures decreased, and arterial oxygen saturation increased (P<0.05–0.001). Regarding laboratory findings, BNP levels and white blood cell counts decreased, and serum potassium, calcium, and lymphocyte counts increased. Some liver‐related markers showed significant changes (P<0.05–0.001). Among these parameters, a decrease in platelet counts and an increase in serum albumin during prior UPH were significantly associated with F‐re‐UPH. This association remained consistent even after adjustment for the three factors. In contrast, a decrease in serum albumin was significantly associated with FU‐re‐UPH after adjustment (P=0.049) (Table S1). Table 8Changes in Clinical Variables During First Unplanned HospitalizationVariablesAdmissionDischargeNo. P valuePhysicalRespiratory rate23±920±4150<0.001Body temperature36.9±0.836.5±0.5216<0.001Body weight, kg38.0±19.036.2±18.0188<0.001Change in body weight, %…−4.2±5.8188Heart rate, bpm96±2689±171050.016Systolic blood pressure, mm Hg102±1599±122130.006Diastolic blood pressure, mm Hg61±1258±112040.005SpO2, %91±692±62120.005Brain natriuretic peptide, pg/mL62 (23–181)31 (17–84)71<0.001ElectrolyteK, meq/L3.9±0.64.0±0.61700.018Ca, mg/dL8.8±0.89.1±0.6130<0.001Peripheral bloodWhite blood cell count, ×103/μL7.9±5.25.7±3.1172<0.001Lymphocyte, %19±1222±121690.002KidneyCreatinine, mg/dL0.58 (0.43–0.81)0.55 (0.40–0.71)1750.004LiverTotal protein, g/dL6.1±1.36.4±1.1161<0.001Albumin, g/dL3.6±0.93.8±0.71670.016Total bilirubin, mg/dL1.0 (0.5–1.7)0.7 (0.5–1.3)1580.007γ‐glutamyl transpeptidase, IU/L62 (38–99)78 (45–133)136<0.001Values are mean±SD or median with interquartile range. Changes in Clinical Variables During First Unplanned Hospitalization Values are mean±SD or median with interquartile range. DISCUSSION In this large‐scale study of Fontan patients, we reconfirmed the major causes of UPH and identified risk factors for re‐UPH, yielding several novel insights. The overall incidence of UPH was 2.5 per 100 person‐years, with 63% classified as F‐UPH and 37% as FU‐UPH. Infections were a frequent cause, including among adults, whereas thromboembolism and hemorrhagic events were significantly more common in adults than in pediatric patients. Although large institutions had lower initial UPH rates, they exhibited higher rates of re‐UPH. UPH due to PLE, thromboembolism, and HF was associated with longer LOS. Regardless of cause, oxygen therapy and diuretics were the most frequently used treatments; patients with PLE often received albumin and IgG supplementation. Re‐UPH occurred at a rate of 15.4 per 100 person‐years among those with prior UPH, often due to the same cause. In‐hospital treatments, inadequate adjustment of oral diuretic dosage during the prior UPH, continued use of diuretics, and absence of ACEI/ARB therapy at discharge were all associated with an increased risk of re‐UPH. Causes and Frequency of Hospitalization Previous large‐scale studies have reported the frequency of hospitalizations based on underlying pathophysiology, primarily focusing on F‐UPHs. According to a report from the Mayo Clinic, mortality increased from 26% to 57% between 10 and 30 years post Fontan operation, with the main causes being PLE, ventricular dysfunction, and arrhythmias. 7 In contrast, a report from the Australian and New Zealand registry showed a marked decrease in mortality from 17% to 7% between 15 and 25 years postoperatively; however, the incidence of complications such as HF, arrhythmias, and thromboembolism increased. 8 These epidemiological findings are consistent with our own observations. Although the overall incidence of UPH was lower in large‐volume centers, the re‐UPH rate was significantly higher. This higher rate of re‐UPH may be partly attributed to the greater proportion of PLE cases in the large‐institution group (29% versus 17%, P=0.02). In terms of FU‐UPHs, infection‐related UPHs in pediatric Fontan patients are particularly concerning, given previous findings suggesting a link between early postoperative infections and increased mortality. 9 Notably, the frequency of infection‐related UPH in adults was comparable to that seen in children and was as high as that of HF‐related UPH. In patients with HF, infection, including influenza, is known to be associated with poor outcomes. 10 , 11 Therefore, infection prevention strategies, including vaccination, should be emphasized in the management of Fontan patients regardless of age. Inpatient Management This study provided a comprehensive overview of current inpatient management practices in Fontan patients. Although previous reports have described LOS in this population, 9 , 12 , 13 few have examined the relationship between treatment strategies and postdischarge outcomes. Length of Hospital Stay Although LOS varied widely across cases, patients admitted for the management of PLE, thromboembolism, and HF had prolonged LOS of ≈2 weeks, suggesting the complexity and challenges involved in treating these conditions. In contrast, patients admitted for arrhythmias or hemorrhage were discharged within 1 week. Notably, LOS in our cohort was longer compared with reports from Western countries, 9 , 12 , 13 which may be attributable to differences in national health care systems, particularly Japan’s universal health insurance model. 14 However, considering the rising health care costs associated with the growing adult Fontan population, 13 , 15 efforts to shorten LOS are increasingly important. Optimizing inpatient management of PLE and HF and preventing re‐UPH should be prioritized. Interestingly, in large‐volume centers, patients with HLHS and those treated at larger institutions had longer LOS compared with their counterparts. Given the improving long‐term outcomes yet high morbidity of patients with HLHS, 16 long‐term management in this subgroup warrants special attention. Treatment Strategies Nonpharmacologic Interventions Oxygen therapy was commonly used across all categories of UPH. Although primarily administered to more severely ill patients, oxygen supplementation in Fontan patients has been shown to improve hypoxemia and lower central venous pressure in the acute setting. 17 Prospective studies are needed to assess its long‐term effects. Notably, percutaneous or surgical interventions such as pacemaker implantation and drainage procedures were associated with a significantly reduced risk of re‐UPH, suggesting their substantial therapeutic impact and supporting their proactive use. Pharmacologic Therapy Diuretic use was widespread and, as in other cardiovascular conditions, appeared closely tied to prognosis in Fontan patients. 3 , 18 Titrating diuretic dosages during UPH based on the patient’s condition may be crucial for improving outcomes. Although no standardized guidelines exist for pharmacologic management in Fontan physiology, the high frequency of diuretic use and apparent treatment inertia may contribute to disease progression. Thus, UPH may serve as a valuable opportunity for medication optimization in these patients. 19 Interestingly, patients treated with ACEIs or ARBs had a lower incidence of re‐UPH. Although prior randomized trials did not demonstrate a benefit of ACEI/ARB in terms of exercise capacity, 20 our findings showed a reduced re‐UPH rate in patients receiving ACEI/ARB at discharge, even though they had a higher prevalence of ≥ moderate atrioventricular valve regurgitation (32% versus 14%, P=0.028). This underscores the need for future studies to evaluate their impact on long‐term outcomes. Supplemental Therapy Interestingly, both albumin and IgG supplementation were used not only in patients with PLE but also in some patients with HF. Given that hypogammaglobulinemia, anemia, and infections are associated with HF exacerbations, 10 , 11 , 21 , 22 it is possible that these supplements were used prophylactically in clinical practice. However, there is currently no established evidence supporting their use in HF management among Fontan patients, indicating a need for further investigation. The notable use of surgical interventions, intravenous therapies, and supplementation even in FU‐UPHs suggests that standard diseases may require more cautious management in Fontan patients due to their unique physiology. Treatment‐Related Changes in Clinical Status In patients with HF, disease progression often occurs with each UPH, eventually leading to death. 23 In this study, many respiratory, circulatory, hepatic, and renal parameters improved during UPH, and most patients were discharged in a clinically improved state. However, the study did not provide clear benchmarks regarding the degree of therapeutic intervention, such as changes in body weight, required to improve long‐term outcomes. Postdischarge Outcomes Several factors related to both patient condition and in‐hospital management were found to be associated with the risk of re‐UPH. Among clinical factors, established comorbidities such as PLE and pulmonary arteriovenous fistulae, along with hemodynamic parameters such as elevated central venous pressure and high cardiac output, were reconfirmed as risk factors. 3 , 4 , 18 Regarding inpatient management, oxygen therapy, intravenous diuretics, supplemental therapies, and failure to adjust oral diuretic dosages, possibly indicative of clinical inertia, were associated with a higher risk of re‐UPH. In contrast, nonpharmacologic interventions (excluding oxygen) and ACEI/ARB use at discharge were associated with a lower risk of re‐UPH. Among these, addressing diuretic treatment inertia may be an especially modifiable factor. Postoperative complications following the Fontan procedure frequently include arrhythmias such as bradycardia, refractory PLE, pleural or peritoneal effusions, hemoptysis, and gastrointestinal bleeding, many of which are difficult to manage pharmacologically. These findings underscore the importance of proactively considering surgical or endoscopic interventions that are tailored to the specific underlying pathophysiology. Furthermore, the recurrent nature of UPH due to persistent underlying conditions highlights the need to establish category‐specific inpatient management protocols aimed at improving long‐term outcomes and minimizing clinical inertia. 19 Study Limitations This study has some limitations. First, the criteria for UPH, treatment, and discharge may have varied across institutions and attending physicians. As a result, cases managed in the outpatient setting were not included. Conversely, some patients with relatively mild clinical conditions may have been hospitalized. Second, the primary cause of UPH may have been difficult to determine in some cases. For instance, when HF, PLE, and arrhythmias coexisted, distinguishing the primary precipitating factor from a secondary consequence may have been challenging. Although the UPH categorization was delegated to the local investigators at each institution, this remains a limitation of the study. Third, UPHs at institutions other than the primary facility where the patient was being followed may have been missed, leading to potential underestimation of UPHs. Fourth, the health care payment system, particularly in relation to the LOS, may have influenced LOS. In Japan, where a universal health insurance system guarantees coverage for inpatient costs, LOS tends to be longer than in Western countries. This factor limits the generalizability of the findings to other health care settings. Additionally, rising health care costs among adult Fontan patients are a growing concern in Japan and warrant further investigation 15 . Fifth, although the study was prospective in design, missing data in clinical test results limited the ability to comprehensively identify prognostic factors. Sixth, this study focused on UPH, and detailed analyses regarding mortality or by UPH category were not performed. Future research should include more in‐depth investigations into mortality and category‐specific UPHs. Finally, the establishment of a causal relationship between in‐hospital management strategies during UPHs and the risk of readmission, as suggested in this study, requires further prospective investigation. CONCLUSIONS This study clarified not only the frequency of UPHs among Fontan patients in real‐world clinical practice in Japan but also the actual management strategies employed during UPH. Regardless of age, the frequency of admissions due to infections and other noncardiac causes was comparable to those due to HF and PLE, whereas UPHs related to thromboembolism and hemorrhage were more common in adult patients. LOS was notably longer in cases of PLE, thromboembolism, and HF. Across all causes of UPH, oxygen therapy and the use of diuretics were commonly employed treatments, and adjustments to oral medications during UPH were infrequently made. After discharge, re‐UPHs often occurred due to the same underlying cause. In‐hospital treatment strategies and discharge medication status, especially clinical inertia regarding diuretic therapy, were associated with the risk of re‐UPH. These findings highlight the need for prospective clinical trials to determine whether improving the management of UPHs can lead to better long‐term outcomes in this patient population. Sources of Funding This study was supported by research grants awarded to Hideo Ohuchi from the Japanese Society of Pediatric Cardiology and Cardiac Surgery (Project A), the Ministry of Health, Labour and Welfare of Japan (22K07905, 24FC1014), and the Miyata Cardiac Research Promotion Foundation. Disclosures None. Supporting information Table S1 Figures S1–S4 This article was sent to Daniel E. Clark, MD, MPH, Associate Editor, for review by expert referees, editorial decision, and final disposition. Supplemental Material is available at https://www.ahajournals.org/doi/suppl/10.1161/JAHA.125.044374 For Sources of Funding and Disclosures, see page 18. References Associated Data Supplementary Materials Table S1 Figures S1–S4 Data Availability Statement The data that support the findings of this study are available from the corresponding author upon reasonable request, subject to institutional and ethical committee approvals.