Incidence, pathophysiology, and treatment of failing Fontan after the total cavopulmonary connection.

BACKGROUND: Failing Fontan poses a significant clinical challenge. This study aims to improve patients' outcomes by comprehensively understanding the incidence, pathophysiology, risk factors, and treatment of failing Fontan after total cavopulmonary connection. METHODS: We performed a retrospective analysis of patients who underwent total cavopulmonary connection at the German Heart Center Munich between 1994 and 2022. The onset of failing Fontan was defined as: protein-losing enteropathy, plastic bronchitis, NYHA class IV, NYHA class III for > one year, unscheduled hospital admissions for heart failure symptoms, and evaluation for heart transplantation. RESULTS: Among 634 patients, 76 patients presented with failing Fontan, and the incidence was 1.48 per 100 patient-years. Manifestations included protein-losing enteropathy (n = 34), hospital readmission (n = 28), NYHA III (n = 18), plastic bronchitis (n = 16), evaluation for heart transplantation (n = 14), and NYHA IV (n = 4). Risk factors for the onset of failing Fontan were dominant right ventricle (p = 0.010) and higher pulmonary artery pressure before total cavopulmonary connection (p = 0.004). A total of 72 interventions were performed in 59 patients, including balloon dilatation/stent implantation in the total cavopulmonary connection pathway (n = 49) and embolization of collaterals (n = 24). Heart transplantation was performed in four patients. The survival after the onset of Fontan failure was 77% at 10 years. Patients with failing Fontan revealed significantly higher zlog-NT-proBNP levels after onset compared to those without (p = 0.021). CONCLUSIONS: The incidence of Fontan failure was 1.5 per 100 patient years. Dominant right ventricle and higher pulmonary artery pressure before total cavopulmonary connection were significant risks for the onset of failing Fontan. Zlog-NT-proBNP is only a late marker of Fontan failure.

Impact of low birth weight on staged single-ventricle palliation.

BACKGROUND: To assess the impact of low birth weight on early and late outcomes after staged palliation for single ventricle. METHODS: Patients after stage 1 palliation for single ventricle in our institution were retrospectively included and divided into two weight groups: 2.5 kg or less (low birth weight) and more than 2.5 kg. The impact of low birth weight on mortality and on the progression to further palliation stages (bidirectional Glenn, stage 2, and total cavopulmonary connection, stage 3) was assessed. RESULTS: A total of 452 patients were included. Patients with low birth weight (n = 37, 8 %) had more frequently associated prematurity and extracardiac anomalies. Early and inter-stage mortality after stage 1 was higher in patients with low birth weight, so that less of these patients reached the next palliation stage (57 % vs. 77 %, p = 0.01, and 38 % vs. 56 %, p = 0.05, for stage 2 and stage 3, respectively). After 5 years, overall survival was inferior in patients with low birth weight (48 % vs. 73 %, p < 0.001). Survival conditioned by stage 2 palliation was inferior in patients with low birth weight compared to the reference group (76 % vs. 89 % after 5 years, p = 0.04). Low birth weight was a risk factor for death in most patients' subgroups, inclusive those with restricted pulmonary blood flow after a systemic-to-pulmonary shunt procedure. CONCLUSIONS: During staged palliation of single-ventricle physiology, low birth weight has a detrimental impact on survival extending to beyond stage 2. This study calls for increased vigilance of these patients beyond the first interstage.

Outcomes after bidirectional cavopulmonary shunt with antegrade pulmonary blood flow in high-risk patients.

OBJECTIVE: We have left antegrade pulmonary blood flow (APBF) at bidirectional cavopulmonary shunt (BCPS) only for high-risk patients. This study evaluates the indication and the outcomes of patients with APBF, compared to those without APBF. METHODS: Patients with APBF after BCPS were identified among patients who underwent BCPS between 1997 and 2022. Outcomes of patients with and without APBF after BCPS were compared. RESULTS: APBF was open in 38 (8.2%) of 461 patients. Median age (7.7 versus 6.3 months, p = 0.55) and weight (5.6 versus 6.1 kg, p = 0.75) at BCPS were similar in both groups. The most frequent indication for APBF was high pulmonary artery pressure (PAP) in 14 patients, followed by hypoxaemia in 10, and hypoplastic left pulmonary artery in 8. The source of APBF was the pulmonary trunk in 10 patients and the aortopulmonary shunt in 28. Median hospital stay after BCPS was longer (22 versus 14 days, p = 0.018) and hospital mortality was higher (10.5 versus 2.1%, p = 0.003) in patients with APBF compared to those without APBF. However, 448 hospital survivors showed similar survival after discharge following BCPS (p = 0.224). Survival after total cavopulmonary connection (TCPC) was similar between the groups (p = 0.753), although patients with APBF were older at TCPC compared to those without (3.9 versus 2.2 years, p = 0.010). CONCLUSION: APBF was left in 8% following BCPS in high-risk patients, mainly due to preoperative high PAP. Hospital survivors after BCPS demonstrated comparable survival in patients with and without APBF. Adding APBF at BCPS might be a useful option for high-risk patients.

Impact of veno-venous collaterals on outcome after the total cavopulmonary connection.

OBJECTIVE: To evaluate the prevalence of veno-venous collaterals (VVCs) after total cavopulmonary connection (TCPC) and analyze their impact on outcomes. METHODS: Patients undergoing TCPC between 1994 and 2022 were evaluated. VVCs were identified using angiograms of cardiac catheterizations and their impact on outcomes was analyzed. RESULTS: A total of 635 patients were included. Median age at TCPC was 2.3 (interquartile ranges (IQR): 1.8-3.3) years. The most frequent diagnosis was hypoplastic left heart syndrome in 173 (27.2%) patients. Prior bidirectional cavopulmonary shunt was performed in 586 (92.3%) patients at a median age of 5.3 (3.6-9.9) months. VVCs were found in 94 (14.8%) patients at a median of 2.8 (0.1-11.8) years postoperatively. The prevalence of VVCs was similar between the dominant right and left ventricle (14.7 vs. 14.9%, p = 0.967). Mean pulmonary artery pressure (16.2 vs. 16.0 mmHg, p = 0.902), left atrial pressure (5.5 vs. 5.7 mmHg, p = 0.480), transpulmonary gradient (4.0 vs. 3.8 mmHg, p = 0.554) and oxygen saturation (81.4 vs. 82.6%, p = 0.103) before TCPC were similar between patients with and without VVCs. The development of VVCs did not affect survival after TCPC (p = 0.161). Nevertheless, VVCs were a risk for the development of plastic bronchitis (PB, p < 0.001). Interventional closure of VVCs was performed in 60 (9.4%) patients at a median of 8.9 (0.6-15.1) years after TCPC, and improvement of oxygen saturation was observed in 66% of the patients. CONCLUSIONS: The prevalence of VVCs after TCPC was 15%. VVCs had no impact on survival following TCPC but were associated with a high prevalence of PB.

Quantification of ventricular stress in univentricular hearts during early childhood using age-independent zlog-NT-proBNP.

BACKGROUND: Children with univentricular hearts (UVH) undergo up to three palliative surgical procedures to achieve complete circulatory separation (Fontan circulation). As a marker of cardiac wall stress, NT-proBNP is a promising tool to assess systemic ventricular load in these patients. However, different reference intervals (RI) apply to each stage, as NT-proBNP is highly age-dependent. METHODS: Children undergoing systemic-to-pulmonary (SP) shunt placement (stage 1), bidirectional cavopulmonary shunt (BCPS, stage 2) or total cavopulmonary connection (TCPC, stage 3) between 2011 and 2021 with NT-proBNP measurement within 7 days before surgery were included. Furthermore, outpatients after TCPC with NT-proBNP measurement were enrolled. Biomarker levels were evaluated using its age-adjusted z-score ("zlog-NT-proBNP"; age-independent RI, -1.96 to +1.96), allowing comparison between different stages and revealing changes in systemic ventricular load independent of the marked physiological decline in RI with age. RESULTS: Overall, 289 children (227 before, 62 after TCPC) met the eligibility criteria. Median time between blood sampling and surgery (SP shunt/BCPS/TCPC) was 2 [1-3] days and 3.2 [2.0-4.5] years after TCPC. Age-adjusted zlog-NT-proBNP levels were 3.47 [2.79-3.93] in children with native UVH (before SP shunt), 3.10 [1.89-3.58] at stage 1 (before BCPS), 1.08 [0.51-1.88] at stage 2 (before TCPC), and 1.09 [0.72-1.75] at stage 3 (after TCPC/Fontan completion). Consequently, BCPS revealed the strongest decrease (median - 2.02 logarithmized standard deviations, p < 0.001). CONCLUSIONS: In children with UVH undergoing staged Fontan palliation, zlog-NT-proBNP is a highly promising tool for course assessment of systemic ventricular load, independent of the age-related decline in physiological NT-proBNP concentration.

Bioactive adrenomedullin (bio-ADM) is associated with endothelial dysfunction in infants and children with complex congenital heart disease undergoing open-heart surgery on cardiopulmonary bypass.

OBJECTIVES: Children with congenital heart disease (CHD) undergoing cardiac surgery on cardiopulmonary bypass (CPB) are at risk for systemic inflammation leading to endothelial dysfunction associated with increased morbidity. Bioactive adrenomedullin (bio-ADM) is a peptide regulating vascular tone and endothelial permeability. The aim of this study was to evaluate the dynamics of plasma bio-ADM in this patient cohort and its role in capillary leak. METHODS: Plasma samples from 73 pediatric CHD patients were collected for bio-ADM measurement at five different timepoints (TP) in the pre-, intra-, and post-operative period. The primary endpoint was a net increase in bio-ADM levels after surgery on CPB. Secondary endpoints included association of bio-ADM levels with clinical signs for endothelial dysfunction. RESULTS: Bio-ADM levels increased after surgery on CPB from pre-operative median of 12 pg/mL (IQR [interquartile range] 12.0-14.8 pg/mL) to a maximum post-operative median of 48.8 pg/mL (IQR 34.5-69.6 pg/mL, p<0.001). Bio-ADM concentrations correlated positively with post-operative volume balance, (r=0.341; p=0.005), increased demand for vasoactive medication (duration: r=0.415; p<0.001; quantity: TP3: r=0.415, p<0.001; TP4: r=0.414, p<0.001), and hydrocortisone treatment for vasoplegia (bio-ADM median [IQR]:129.1 [55.4-139.2] pg/mL vs. 37.9 [25.2-64.6] pg/mL; p=0.034). Patients who required pleural effusion drainage revealed higher bio-ADM levels compared to those who did not (median [IQR]: 66.4 [55.4-90.9] pg/mL vs. 40.2 [28.2-57.0] pg/mL; p<0.001). CONCLUSIONS: Bio-ADM is elevated in children after cardiac surgery and higher levels correlate with clinical signs of capillary leakage. The peptide should be considered as biomarker for endothelial dysfunction and as potential therapeutic target in this indication.

Influence of Shunt Type on Survival and Right Heart Function after the Norwood Procedure for Aortic Atresia.

The study objective was to compare the results after Norwood procedure between modified Blalock-Taussig shunt (MBTS) and right ventricle-to-pulmonary artery conduit (RVPAC) according to Sano in patients with hypoplastic left heart syndrome (HLHS) and aortic atresia (AA). A total of 146 neonates with HLHS and AA who underwent the Norwood procedure at our institution between 2001 and 2020 were divided into 2 groups according to shunt type (MBTS or RVPAC). Survival after the Norwood procedure was compared between the groups. Longitudinal right ventricular and tricuspid valve function in each group were evaluated using cubic splines method. RVPAC was performed in 103 patients and MBTS in 43 according to surgeon preference. There were no differences in the 30-day mortality rates (16.5% vs 16.3%, P  = 0.973). Survival at 0.5, 1 and 3 years was 79.6%, 74.6%, and 68.9% in RVPAC and 66.8%, 64.3%, and 58.5% in MBTS (P  =  0.293). Among 23 patients undergoing tricuspid valve procedure, different mechanisms of tricuspid regurgitation were observed between the groups. Longitudinal analysis revealed greater prevalence of late right ventricular dysfunction in RVPAC patients. In 77 patients who completed Fontan procedure, the postoperative N-terminal pro B-type natriuretic peptide value was significantly higher in RVPAC vs MBTS (554 vs 276 ng/L, P  =  0.007). No survival advantage of RVPAC over MBTS was observed in neonates with HLHS and AA undergoing the Norwood procedure. Longitudinal analysis demonstrated a greater prevalence of right ventricular dysfunction and higher N-terminal pro B-type natriuretic peptide values during late follow-up in patients with RVPAC.

Predicting Major Adverse Cardiovascular Events in Children With Age-Adjusted NT-proBNP.

BACKGROUND: N-terminal pro-B-type natriuretic peptide (NT-proBNP) is frequently used as a valuable prognostic biomarker in cardiac diseases. In children, however, it has not been established because of its strong age dependency. To overcome this obstacle, we recently introduced the zlog value of N-terminal pro-B-type natriuretic peptide (zlog-proBNP) as an age-adjusted reference. OBJECTIVES: This study evaluates the prognostic power of zlog-proBNP for the occurrence of major adverse cardiovascular events (MACE) throughout childhood in patients with congenital heart diseases (CHD). METHODS: A total of 910 children with CHD (median age 5 months; range 0.0-18.0 years) were included. MACE was defined as death, resuscitation, mechanical circulatory support, or hospitalization caused by cardiac decompensation. Because the physiological NT-proBNP concentration decreases significantly during childhood, zlog values were applied for an age-independent evaluation. RESULTS: MACE occurred in 138 children during a median follow-up of 6 months (range 1 day to 7.6 years). High zlog-proBNP values (>+3.0) were most strongly associated with adverse events (n = 93; adjusted HR: 21.1; 95% CI: 2.9-154.2; P < 0.001). Among all evaluated indicators, zlog-proBNP was the best predictor for MACE (adjusted HR: 1.52; 95% CI: 1.31-1.76; P < 0.001) along with age and predictively superior to absolute NT-proBNP values. A cutoff value of +1.96 (age-independent upper limit of the physiological NT-proBNP concentration) achieved a negative predictive value of >96%. CONCLUSIONS: Zlog-proBNP overcomes the strong age dependency of NT-proBNP and is a powerful prognostic marker for age-independent exclusion and prediction of MACE in children with CHD. We therefore expect zlog-proBNP to play a pivotal role in the future management of children with heart diseases.

Continuous, complete and comparable NT-proBNP reference ranges in healthy children.

Background: NT-proBNP is one of the most important biomarkers for the diagnosis and risk assessment of heart failure in adults. Age- and gender-independent reference intervals (RIs) have been reported. In contrast, RIs in children are strongly age-dependent, do not exist for all ages and reveal a right-skewed distribution. Accordingly, no common Z-score can be formed and a cross-age interpretive method, so far, is missing. Methods: Within the paper on hand, new evaluation techniques are applied to already published NT-proBNP study results and additionally to newly gained data. Upper limits (ULs), lower limits (LLs) and 50th percentiles are tested for power-like behavior as a function of age using linear regression analysis. Functions for continuous RIs are derived and reference limits are calculated on a per day basis. A corresponding Zlog formula is deduced and its usefulness is stated in two clinical examples. Results: The power-like behavior of NT-proBNP concentration from birth to 18 years is demonstrated. With age in days t and measured NT-proBNP value x in pg/mL, an age-specific Zlog value may directly be calculated using the equation:ZlogNT-proBNP=log x+0.512⋅log t-3.4171.489+0.014⋅log t⋅3.92${\rm{Zlo}}{{\rm{g}}_{{\rm{NT - proBNP}}}} = {{\log \;x + 0.512 \cdot \log \;t - 3.417} \over {1.489 + 0.014 \cdot \log \;t}} \cdot 3.92$. Conclusions: Using formulas for UL and LL, continuous RIs from 0 to 18 years may be obtained. Continuity corresponds to physiological changes in the body much better than discrete RIs. With the advent of an NT-proBNP-specific Zlog value, a cross-age Z-score equivalent is providing an easy interpretation aid in everyday pediatric practice. This new approach allows to identify clinical worsening much better, sooner and more clearly than previous absolute values.