Long-term outcomes of common atrioventricular valve replacement for patients with functional single ventricle Authors: Keisuke Shibagaki, Yuji Tominaga, Kenta Imai, Motoki Komori, Rieko Kutsuzawa, Hiroyuki Kamiya, Shigemitsu Iwai Published: 2026 ABSTRACT ObjectiveThis study compared the long-term outcomes of common atrioventricular valve replacement after common atrioventricular valve plasty and common atrioventricular valve plasty alone in patients with a functional single ventricle.MethodsWe retrospectively evaluated 30 patients with functional single ventricle who experienced controlled common atrioventricular valve regurgitation with repeated common atrioventricular valve plasty alone (group R(−)) and 19 patients with functional single ventricle who experienced controlled common atrioventricular valve regurgitation with common atrioventricular valve replacement (group R(+)) at our institution. All patients underwent common atrioventricular valve plasty. We evaluated the survival and cumulative incidence of Fontan completion after the initial common atrioventricular valve plasty and hemodynamic parameters after Fontan completion in both groups.ResultsThe cumulative incidence of common atrioventricular valve replacement was 35% at 10 years after the initial common atrioventricular valve plasty. Multivariable competing risk regression analysis identified 2 factors associated with common atrioventricular valve replacement: The initial common atrioventricular valve plasty was performed before the bidirectional Glenn procedure (hazard ratio, 3.13; 95% CI, 1.19-8.19; P = .020) and severe common atrioventricular valve regurgitation at the initial common atrioventricular valve plasty (hazard ratio, 2.84; 95% CI, 1.07-7.59; P = .037). The 10-year survivals of the R(−) (73%) and R(+) (78%) groups were not significantly different. Groups R(−) and R(+) had comparable Fontan completion rates (73% vs 66%) after 3 years. The Fontan pressure, ventricular end-diastolic pressure, and cardiac index were comparable between groups R(−) and R(+) (10 mm Hg vs 10 mm Hg [P = .27], 8 mm Hg vs 7 mm Hg [P = .97], and 2.9 L/min/m2 vs 3.1 L/min/m2 [P = .60], respectively).ConclusionsCommon atrioventricular valve replacement may be a reasonable option for treating uncontrolled common atrioventricular valve regurgitation in patients with a functional single ventricle. Survival and cumulative incidence of Fontan completion. Central MessageCAVV replacement may be one of the therapeutic options for uncontrollable CAVV regurgitation in patients with FSV. PerspectiveIdentifying factors associated with CAVV replacement in patients with FSV can guide optimal timing of intervention. Our findings indicate that CAVV replacement, when CAVV regurgitation is uncontrollable, yields survival and Fontan outcomes comparable to repeated plasty, supporting its consideration as a reasonable option for selected high-risk patients. Survival and cumulative incidence of Fontan completion. Central Message CAVV replacement may be one of the therapeutic options for uncontrollable CAVV regurgitation in patients with FSV. Perspective Identifying factors associated with CAVV replacement in patients with FSV can guide optimal timing of intervention. Our findings indicate that CAVV replacement, when CAVV regurgitation is uncontrollable, yields survival and Fontan outcomes comparable to repeated plasty, supporting its consideration as a reasonable option for selected high-risk patients. The function of the common atrioventricular valve (CAVV) is crucial for the prognosis of patients with a functional single ventricle (FSV).1, 2, 3, 4 CAVV regurgitation is caused by anatomic abnormalities of the valve leaflets and subvalvular structures in the FSV, and it is associated with mortality, morbidity, and Fontan circulation failure in patients with FSV.3, 4, 5, 6, 7 Therefore, CAVV plasty is required to control regurgitation. However, despite improvements in CAVV plasty techniques and the ability to perform repeated CAVV plasty, sometimes CAVV regurgitation cannot be controlled.5,8, 9, 10 Additionally, repeated CAVV plasty can cause ventricular dysfunction.11 However, in patients with FSV, CAVV replacement is a reasonable treatment option for unrepairable CAVV regurgitation because of its acceptably low mortality rate.12, 13, 14, 15, 16 Few studies have reported the long-term outcomes of CAVV replacement in patients with FSV because of the limited number of patients who have undergone this procedure.12 Therefore, CAVV replacement for patients with FSV remains controversial. Additionally, comparisons of the long-term outcomes of CAVV replacement and those of CAVV plasty alone have rarely been reported. Therefore, this study aimed to evaluate and compare the long-term outcomes of CAVV replacement and those of CAVV plasty alone in patients with FSV. Materials and Methods Ethical Approvals This study was approved by the ethics committee of our hospital on December 20, 2019 (Approval Number: 19092). Opt-out consent was obtained instead of individual written informed consent. The ethics committee approved the use and publication of anonymized study data. Patients Forty-nine patients with FSV underwent CAVV plasty as their initial surgery for CAVV regurgitation between 1995 and 2022, of whom 19 eventually required a CAVV replacement. We separated the patients into 2 groups: Group R(−) underwent 1 more CAVV plasty but never underwent replacement, and group R(+) underwent CAVV replacement at any point after their initial CAVV plasty (Figure 1). For outcome analyses, patients were categorized based on their final treatment status during follow-up (CAVV plasty alone vs eventual CAVV replacement), rather than considering CAVV replacement as a time-dependent exposure. The median age of patients at the time of the initial CAVV plasty was 1.1 years, and 26 patients (53%) were female (Table 1). Furthermore, 26 patients (55%) and 20 patients (41%) exhibited asplenia and total anomalous pulmonary venous connections, respectively.Figure 1Flow chart of patient enrollment. One patient underwent CAVV replacement because of CAVV stenosis and regurgitation. FSV, Functional single ventricle.Table 1Preoperative and perioperative patient characteristicsVariableAll (n = 49)Group R(−) (n = 30)Group R(+) (n = 19)P valueAge at initial CAVV plasty, y1.1 (0.5-1.8)1.4 (0.7-2.3)0.8 (0.2-1.2).047Body weight at initial CAVV plasty, kg6.3 (5.3-8.7)7.5 (5.9-11.0)5.4 (3.5-6.4).002Female, n26 (53)16 (53)10 (53).96Primary diagnosis, n Unbalanced atrioventricular defect34 (69)21 (70)13 (68).91 Single ventricle (right)13 (27)9 (30)4 (21).49 Single ventricle (left)2 (4)0 (0)2 (11).07Associated cardiac defect, n Heterotaxy Asplenia26 (55)18 (60)8 (42).22 Polysplenia8 (16)4 (13)4 (21).48 DORV25 (51)12 (40)13 (68).052 PS/pulmonary atresia23 (47)14 (47)9 (47).96 TAPVC20 (41)12 (40)8 (42).88 PVO12 (25)8 (27)4 (21).66Dominant ventricle morphology, n Right42 (86)28 (93)14 (74).055Timing of the initial CAVV plasty, n Before Glenn12 (24)4 (13)8 (42).039 At Glenn24 (49)15 (50)9 (47).86 Between Glenn and Fontan4 (8)4 (13)0 (0).097 At Fontan9 (18)7 (23)2 (11).26Echocardiography before initial CAVV plasty, n CAVV regurgitation grade Mild to moderate14 (29)11 (37)3 (16).12 Moderate18 (37)13 (43)5 (26).23 Moderate to severe8 (16)4 (13)4 (21).48 Severe9 (18)2 (7)7 (37).008 Ejection fraction, % ≦401 (2)0 (0)1 (5).20 41-6023 (47)12 (40)11 (58).22 60<17 (35)12 (40)5 (26).37 Unknown8 (16)6 (20)2 (11).38 Small mural leaflet8 (16)3 (10)5 (26).132 Abnormalities of subvalvular structures8 (16)3 (10)5 (26).132CPB time, min170 (122-312)197 (162-218)273 (204-309).007Aortic crossclamp time, min76 (55-86)54 (45-93)95 (65-117).077Concomitant procedure, n PA banding9 (18)5 (17)4 (21).70 Systemic pulmonary shunt3 (6)2 (7)1 (5).84 Ventricle to pulmonary artery conduit8 (16)2 (7)6 (32).022 Norwood procedure4 (8)1 (3)3 (16).12 Damus-Kaye-Stansel anastomosis7 (14)5 (17)2 (11).55 Glenn procedure24 (49)17 (57)7 (37).18 Fontan procedure8 (16)6 (20)2 (11).38 PA plasty13 (27)7 (23)6 (32).52 TAPVC repair6 (12)2 (7)4 (21).13 PVO release4 (8)2 (7)2 (11).63 Aortic valve plasty1 (2)0 (0)1 (5).20 Cryoablation2 (4)2 (7)0 (0).25Initial CAVV plasty procedures Interannular bridging32 (65)21 (70)11 (58).52 Edge to edge16 (33)7 (23)9 (47).058 Commissure closure11 (22)6 (20)5 (26).60 Annuloplasty14 (29)8 (27)6 (32).62 Cleft suture7 (14)5 (17)2 (11).60 Leaflet suture9 (18)7 (23)2 (11).26Values are expressed as n (%) or median (IQR). CAVV, Common atrioventricular valve; CPB, cardiopulmonary bypass; DORV, double outlet right ventricle; PA, pulmonary artery; PM, pacemaker; PS, pulmonary stenosis; PVO, pulmonary venous obstruction; TAPVC, total anomalous pulmonary venous connection. Flow chart of patient enrollment. One patient underwent CAVV replacement because of CAVV stenosis and regurgitation. FSV, Functional single ventricle. Table 1Preoperative and perioperative patient characteristicsVariableAll (n = 49)Group R(−) (n = 30)Group R(+) (n = 19)P valueAge at initial CAVV plasty, y1.1 (0.5-1.8)1.4 (0.7-2.3)0.8 (0.2-1.2).047Body weight at initial CAVV plasty, kg6.3 (5.3-8.7)7.5 (5.9-11.0)5.4 (3.5-6.4).002Female, n26 (53)16 (53)10 (53).96Primary diagnosis, n Unbalanced atrioventricular defect34 (69)21 (70)13 (68).91 Single ventricle (right)13 (27)9 (30)4 (21).49 Single ventricle (left)2 (4)0 (0)2 (11).07Associated cardiac defect, n Heterotaxy Asplenia26 (55)18 (60)8 (42).22 Polysplenia8 (16)4 (13)4 (21).48 DORV25 (51)12 (40)13 (68).052 PS/pulmonary atresia23 (47)14 (47)9 (47).96 TAPVC20 (41)12 (40)8 (42).88 PVO12 (25)8 (27)4 (21).66Dominant ventricle morphology, n Right42 (86)28 (93)14 (74).055Timing of the initial CAVV plasty, n Before Glenn12 (24)4 (13)8 (42).039 At Glenn24 (49)15 (50)9 (47).86 Between Glenn and Fontan4 (8)4 (13)0 (0).097 At Fontan9 (18)7 (23)2 (11).26Echocardiography before initial CAVV plasty, n CAVV regurgitation grade Mild to moderate14 (29)11 (37)3 (16).12 Moderate18 (37)13 (43)5 (26).23 Moderate to severe8 (16)4 (13)4 (21).48 Severe9 (18)2 (7)7 (37).008 Ejection fraction, % ≦401 (2)0 (0)1 (5).20 41-6023 (47)12 (40)11 (58).22 60<17 (35)12 (40)5 (26).37 Unknown8 (16)6 (20)2 (11).38 Small mural leaflet8 (16)3 (10)5 (26).132 Abnormalities of subvalvular structures8 (16)3 (10)5 (26).132CPB time, min170 (122-312)197 (162-218)273 (204-309).007Aortic crossclamp time, min76 (55-86)54 (45-93)95 (65-117).077Concomitant procedure, n PA banding9 (18)5 (17)4 (21).70 Systemic pulmonary shunt3 (6)2 (7)1 (5).84 Ventricle to pulmonary artery conduit8 (16)2 (7)6 (32).022 Norwood procedure4 (8)1 (3)3 (16).12 Damus-Kaye-Stansel anastomosis7 (14)5 (17)2 (11).55 Glenn procedure24 (49)17 (57)7 (37).18 Fontan procedure8 (16)6 (20)2 (11).38 PA plasty13 (27)7 (23)6 (32).52 TAPVC repair6 (12)2 (7)4 (21).13 PVO release4 (8)2 (7)2 (11).63 Aortic valve plasty1 (2)0 (0)1 (5).20 Cryoablation2 (4)2 (7)0 (0).25Initial CAVV plasty procedures Interannular bridging32 (65)21 (70)11 (58).52 Edge to edge16 (33)7 (23)9 (47).058 Commissure closure11 (22)6 (20)5 (26).60 Annuloplasty14 (29)8 (27)6 (32).62 Cleft suture7 (14)5 (17)2 (11).60 Leaflet suture9 (18)7 (23)2 (11).26Values are expressed as n (%) or median (IQR). CAVV, Common atrioventricular valve; CPB, cardiopulmonary bypass; DORV, double outlet right ventricle; PA, pulmonary artery; PM, pacemaker; PS, pulmonary stenosis; PVO, pulmonary venous obstruction; TAPVC, total anomalous pulmonary venous connection. Preoperative and perioperative patient characteristics Values are expressed as n (%) or median (IQR). CAVV, Common atrioventricular valve; CPB, cardiopulmonary bypass; DORV, double outlet right ventricle; PA, pulmonary artery; PM, pacemaker; PS, pulmonary stenosis; PVO, pulmonary venous obstruction; TAPVC, total anomalous pulmonary venous connection. Initial CAVV plasty was usually performed for patients with worse than moderate CAVV regurgitation during the Glenn procedure. Figure 2 shows the timing of the initial CAVV plasty. When the ventricular function decreased, CAVV plasty was performed before the Glenn procedure. Among the 49 patients, 12 underwent CAVV plasty before the Glenn procedure and 24 underwent CAVV plasty during the Glenn procedure. At our institution, valve plasty is the first-line treatment. Surgical procedures comprising CAVV plasty at our institution include interannular bridging, edge-to-edge repair, cleft closure, commissure closure, and annuloplasty; various combinations of these procedures are performed depending on the cause of CAVV regurgitation or structures (Table 1). Additionally, CAVV plasty was performed in patients with worse than mild CAVV regurgitation when regurgitation worsened due to hemodynamic changes. After initial CAVV plasty, a residual CAVV regurgitation grade of less than moderate was defined as “controlled CAVV regurgitation.” Therefore, subsequent CAVV plasty was usually performed when CAVV regurgitation progressed to worse than moderate after the previous CAVV plasty. Furthermore, CAVV replacement was performed before the Glenn procedure when circulatory dynamics collapse because of uncontrollable CAVV regurgitation after CAVV plasty.Figure 2Patients with each CAVV regurgitation grade at the time of initial CAVV plasty. Patients with each CAVV regurgitation grade at the time of initial CAVV plasty. The baseline demographic characteristics of patients who underwent CAVV replacement are shown in Table E1. The median interval between the initial CAVV plasty and subsequent CAVV replacement was 10.7 months (interquartile range [IQR], 3.4-28.8). Six patients underwent CAVV replacement before the Glenn procedure. Surgical Procedure CAVV surgery was performed via full median sternotomy under cardiopulmonary bypass with cardioplegic arrest. Our first-line CAVV plasty techniques are annuloplasty, primarily the Kay–Reed technique, and the edge-to-edge techniques. Regarding CAVV replacement, a prosthetic valve was implanted in the intra-annular position using pledgeted interrupted sutures. To preserve ventricular function and avoid complete atrioventricular block, we attempted to preserve part of the chordal attachments of the posterior bridging leaflet whenever possible. In principle, we selected the appropriate prosthetic valve based on the annular diameter at the time of surgery. When the annulus was too small to accommodate the prosthetic valve, the appropriate prosthetic valve was implanted using a short segment of vascular graft. All patients underwent CAVV replacement using mechanical valves. Ten of these patients received ATS valves (ATS Medical Inc), and 9 patients received St Jude Medica valves (St Jude Medical Inc) (Figure E1). Anticoagulation Management After Common Atrioventricular Valve Replacement After postoperative hemodynamics stabilized, all patients were administered intravenous heparin and the activated partial thromboplastin time was maintained within the range of 45 to 55 seconds. After extubation and the resumption of oral intake, all patients received warfarin and the prothrombin time international normalized ratio was maintained within the range of 2.0 to 3.0. Data Collection and Assessment of Cardiac Function All variables were obtained from the patients’ medical records, including echocardiography and catheter examination reports. Time-related outcomes were assessed based on data from the most recent outpatient visits recorded in the electronic medical records. Follow-up was completed in 92% of patients. CAVV regurgitation was measured preoperatively and postoperatively using transthoracic echocardiography. The severity of CAVV regurgitation was assessed and classified as none, trivial, mild, mild-to-moderate, moderate, moderate-to-severe, or severe regurgitation. Outcomes The overall cumulative incidence and factors associated with CAVV replacement were examined. These factors included age, body weight, severe CAVV regurgitation at the time of the initial CAVV plasty, and undergoing the initial CAVV plasty before the Glenn procedure. Next, we evaluated the survival of groups R(−) and R(+) after the initial CAVV plasty. The cumulative incidence of Fontan completion after the initial CAVV plasty and hemodynamic parameters after Fontan completion in groups R(−) and R(+) were also assessed. Post-Fontan cardiac catheterization was routinely performed at 5-year intervals, beginning at 1 year after surgery (ie, at 1, 5, and 10 years, and thereafter). Hemodynamic parameters, including Fontan pressure, ventricular end-diastolic pressure, and cardiac index, of patients who underwent Fontan completion were measured using cardiac catheterization at the final follow-up. Statistical Analysis All analyses were performed using SPSS statistics software (version 30.0; IBM Corp) and EZR (Saitama Medical Center, Jichi Medical University).17 Continuous variables are presented as medians with IQRs, and categorical variables are presented as numbers or percentages. Group comparisons (group R(−) vs group R(+)) were performed using the Mann–Whitney U test for continuous variables and Fisher exact test for categorical variables, as appropriate. Survival and the cumulative incidence of Fontan completion in groups R(−) and R(+) were estimated using Kaplan–Meier curves and compared using a time-varying covariate model. However, the cumulative incidence of CAVV replacement was estimated using a competing risk analysis, and competing risk regression was performed to determine the risk factors for CAVV replacement. The competing risk analysis included the following factors: age, body weight, undergoing the initial CAVV plasty before the Glenn procedure, and severe CAVV regurgitation at the initial CAVV plasty. Hazard ratios (HRs) and 95% CIs were calculated for all variables. Results Patients’ Characteristics Compared with patients in group R(−), patients in group R(+) were significantly younger (1.4 years vs 0.8 years; P = .047) and had lower weights (7.5 kg vs 5.4 kg; P = .002). No significant difference was observed in the sex of the patients between the 2 groups. Additionally, compared with patients in group R(−), more patients in group R(+) underwent the initial CAVV plasty before the Glenn procedure (13% vs 42%; P = .039). Compared with patients in group R(−), more patients in group R(+) had severe CAVV regurgitation according to the echocardiography results (7% vs 37%; P = .008). However, no significant differences were observed in the ventricular ejection fraction or surgical repair techniques between the 2 groups. Table 2 presents the postoperative outcomes after the initial CAVV plasty. Twenty patients had worse than moderate CAVV regurgitation, and 14 of those patients had severe CAVV regurgitation. Compared with patients in group R(−), more patients in group R(+) had severe CAVV regurgitation (n = 14 [74%] vs 0 [0%]; P = .009). No significant differences in extracorporeal membrane oxygenation cannulation after CAVV plasty were observed between the 2 groups.Table 2Postoperative patient characteristics at the initial common atrioventricular valve plastyVariableAll (n = 49)Group R(−) (n = 30)Group R(+) (n = 19)P valueEchocardiography after the initial CAVV plasty CAVV regurgitation grade Better than moderate29 (59)26 (53)3 (16).001 Moderate5 (10)2 (7)3 (16).30 Moderate to severe11 (22)2 (7)9 (47).001 Severe4 (8)0 (0)14 (74).009 Ejection fraction, % <402 (4)0 (0)2 (11).07 41-505 (10)2 (7)3 (16).95 51-6017 (35)6 (20)11 (58).72 >6015 (31)7 (23)8 (42).45 Unknown10 (20)8 (27)2 (11).17ECMO cannulation5 (10)3 (10)2 (11).30Echocardiography at latest follow-up CAVV regurgitation grade Less than moderate21 (70) Moderate8 (27) Moderate to severe1 (3) Severe0 (0)Values are expressed as n (%) or median (IQR). CAVV, Common atrioventricular valve; ECMO, extracorporeal membrane oxygenation. Table 2Postoperative patient characteristics at the initial common atrioventricular valve plastyVariableAll (n = 49)Group R(−) (n = 30)Group R(+) (n = 19)P valueEchocardiography after the initial CAVV plasty CAVV regurgitation grade Better than moderate29 (59)26 (53)3 (16).001 Moderate5 (10)2 (7)3 (16).30 Moderate to severe11 (22)2 (7)9 (47).001 Severe4 (8)0 (0)14 (74).009 Ejection fraction, % <402 (4)0 (0)2 (11).07 41-505 (10)2 (7)3 (16).95 51-6017 (35)6 (20)11 (58).72 >6015 (31)7 (23)8 (42).45 Unknown10 (20)8 (27)2 (11).17ECMO cannulation5 (10)3 (10)2 (11).30Echocardiography at latest follow-up CAVV regurgitation grade Less than moderate21 (70) Moderate8 (27) Moderate to severe1 (3) Severe0 (0)Values are expressed as n (%) or median (IQR). CAVV, Common atrioventricular valve; ECMO, extracorporeal membrane oxygenation. Postoperative patient characteristics at the initial common atrioventricular valve plasty Values are expressed as n (%) or median (IQR). CAVV, Common atrioventricular valve; ECMO, extracorporeal membrane oxygenation. Characteristics of Patients Who Underwent Common Atrioventricular Valve Replacement Nineteen patients underwent CAVV replacement after CAVV plasty. The cumulative incidence rates of CAVV replacement were 27%, 33%, and 35% after 1, 5, and 10 years, respectively (Figure 3, A). Table 3 shows the factors associated with CAVV replacement according to the univariable and multivariable competing risk regression analyses. Multivariable competing risk regression identified the following 2 independent factors associated with CAVV replacement: the initial CAVV plasty performed before the Glenn procedure (HR, 3.13; 95% CI, 1.19-8.19; P = .020) and severe CAVV regurgitation at the initial CAVV plasty (HR, 2.84; 95% CI, 1.07-7.59; P = .037).Figure 3The cumulative incidence of CAVV replacement (A). Kaplan–Meier curves of survival of patients who underwent CAVV replacement (group R(+)) or CAVV plasty only (group R(−)) (B). The cumulative incidence of Fontan completion in patients who underwent CAVV replacement (group R(+)) or CAVV plasty only (group R(−)) (C). Shaded areas indicate 95% CIs.Table 3Results of the multivariable competing risk regression analysis of common atrioventricular valve replacementVariableUnivariableMultivariableHR (95% CI)P valueHR (95% CI)P valueInitial CAVV plasty before Glenn4.10 (1.62-10.4).0033.13 (1.19-8.19).020Severe CAVV regurgitation at initial CAVV plasty3.87 (1.57-9.57).0032.84 (1.07-7.59).037Body weight at initial CAVV plasty, kg0.86 (0.68-1.09).22Age at initial CAVV plasty, y0.96 (0.95-1.05).96Age, body weight, and severe CAVV regurgitation at the initial CAVV plasty and the initial CAVV plasty before the Glenn procedure were included. HR, Hazard ratio; CAVV, common atrioventricular valve. The cumulative incidence of CAVV replacement (A). Kaplan–Meier curves of survival of patients who underwent CAVV replacement (group R(+)) or CAVV plasty only (group R(−)) (B). The cumulative incidence of Fontan completion in patients who underwent CAVV replacement (group R(+)) or CAVV plasty only (group R(−)) (C). Shaded areas indicate 95% CIs. Table 3Results of the multivariable competing risk regression analysis of common atrioventricular valve replacementVariableUnivariableMultivariableHR (95% CI)P valueHR (95% CI)P valueInitial CAVV plasty before Glenn4.10 (1.62-10.4).0033.13 (1.19-8.19).020Severe CAVV regurgitation at initial CAVV plasty3.87 (1.57-9.57).0032.84 (1.07-7.59).037Body weight at initial CAVV plasty, kg0.86 (0.68-1.09).22Age at initial CAVV plasty, y0.96 (0.95-1.05).96Age, body weight, and severe CAVV regurgitation at the initial CAVV plasty and the initial CAVV plasty before the Glenn procedure were included. HR, Hazard ratio; CAVV, common atrioventricular valve. Results of the multivariable competing risk regression analysis of common atrioventricular valve replacement Age, body weight, and severe CAVV regurgitation at the initial CAVV plasty and the initial CAVV plasty before the Glenn procedure were included. HR, Hazard ratio; CAVV, common atrioventricular valve. Complications After Common Atrioventricular Valve Replacement The complications in patients who underwent CAVV replacement are shown in Table E2. Among the 19 patients who underwent CAVV replacement, extracorporeal membrane oxygenation cannulation was required in 3 patients (16%), and pacemaker implantation was performed in 6 patients (32%). Only 1 case of intracranial hemorrhage was observed after CAVV replacement. Reoperation for CAVV replacement was required in 7 patients, with a total of 9 procedures performed. The median time to reoperation for CAVV replacement after the initial CAVV replacement was 2.4 years (IQR, 0.8-13.0 years). The causes of reoperation for CAVVR included thrombosis in 5 patients, prosthetic valve malfunction in 2 patients, and pannus formation in 2 patients. Freedom from re-CAVV replacement was 89% and 80% after 1 and 5 years, respectively (Figure E2). Survival The median follow-up period was 10 years (IQR, 2.2-18 years). Fourteen patients died (9 and 5 patients in groups R(−) and R(+), respectively). In group R(−), death was caused by pneumonia (1 patient), low-output syndrome (2 patients), acute heart failure (1 patient), sudden death (1 patient), and sepsis (4 patients). In group R(+), death was caused by pneumonia (1 patient), low-output syndrome (1 patient), sudden death (1 patient), and sepsis (1 patient). The survivals in group R(−) were 87% (95% CI, 68-96), 76% (95% CI, 60-89), and 73% (95% CI, 51-86) after 1, 5, and 10 years, respectively, and survivals in group R(+) were 95% (95% CI, 68-99), 78% (95% CI, 51-91), and 78% (95% CI, 51-91) after 1, 5, and 10 years, respectively (Figure 3, B). In a Cox proportional hazards model treating CAVV replacement as a time-dependent covariate, CAVV replacement was not significantly associated with mortality (HR, 0.90, 95% CI, 0.17-4.94, P = .91). The survival in group R(+) after CAVV replacement is shown in Figure E2. Fontan Completion In groups R(−) and R(+), 20 patients (67%) and 10 patients (53%), respectively, underwent Fontan completion at the final follow-up. The cumulative incidence rates of Fontan completion in group R(−) were 45% (95% CI, 29-65), 68% (95% CI, 49-85), and 73% (95% CI, 54-89) after 1, 2, and 3 years, respectively, and those of group R(+) were 21% (95% CI, 8-47), 47% (95% CI, 27-77), and 66% (95% CI, 23-80) after 1, 2, and 3 years, respectively (Figure 3, C). In a Cox proportional hazards model treating CAVV replacement as a time-dependent covariate, CAVV replacement was not significantly associated with Fontan completion (HR, 1.92, 95% CI, 0.41-9.09, P = .41). The median follow-up period after Fontan completion was 12 years (IQR, 5.5-16.7 years). An analysis of patients in both groups who underwent Fontan completion indicated no significant differences in the Fontan pressure (group R(−) vs group R(+): 10 mm Hg vs 10 mm Hg; P = .27), ventricular end-diastolic pressure (group R(−) vs group R(+): 8 mm Hg vs 7 mm Hg; P = .97), and cardiac index (group R(−) vs group R(+): 2.9 L/min/m2 vs 3.1 L/min/m2; P = .60) at the final follow-up. Furthermore, no significant differences in the incidence rates of protein-losing enteropathy (group R(−) vs group R(+): 5% vs%; P = .63) and arrhythmia (group R(−) vs group R(+): 11% vs 10%, P = .97) were observed between patients in both groups who underwent Fontan completion during the follow-up period (Table 4).Table 4Patient characteristics after Fontan completionVariableAll (n = 30)Group R(−) (n = 20)Group R(+) (n = 10)P valueCardiac catheterization Fontan pressure, mm Hg10 (9-12)10 (9-11)10 (10-14).27 Ventricular end-diastolic pressure, mm Hg7 (5-9)8 (5-9)7 (4-15).98 Cardiac index, mL/min/m23.0 (2.6-3.5)2.9 (2.6-3.4)3.1 (2.5-3.8).60Complications Protein-losing enteropathy2 (7)1 (5)1 (10).63 Tachyarrhythmia3 (10)2 (11)1 (10).97 Pacemaker implantation4 (13)1 (5)3 (30).057 Bleeding2 (7)1 (5)1 (10).63Values are expressed as n (%) or median (IQR). Table 4Patient characteristics after Fontan completionVariableAll (n = 30)Group R(−) (n = 20)Group R(+) (n = 10)P valueCardiac catheterization Fontan pressure, mm Hg10 (9-12)10 (9-11)10 (10-14).27 Ventricular end-diastolic pressure, mm Hg7 (5-9)8 (5-9)7 (4-15).98 Cardiac index, mL/min/m23.0 (2.6-3.5)2.9 (2.6-3.4)3.1 (2.5-3.8).60Complications Protein-losing enteropathy2 (7)1 (5)1 (10).63 Tachyarrhythmia3 (10)2 (11)1 (10).97 Pacemaker implantation4 (13)1 (5)3 (30).057 Bleeding2 (7)1 (5)1 (10).63Values are expressed as n (%) or median (IQR). Patient characteristics after Fontan completion Values are expressed as n (%) or median (IQR). Discussion Undergoing the initial CAVV plasty before the Glenn procedure and severe CAVV regurgitation at the initial CAVV plasty were associated with CAVV replacement after CAVV plasty in patients with FSV. Additionally, the survival, Fontan completion rates, and hemodynamic parameters of patients with FSV who underwent CAVV replacement after CAVV plasty were not inferior to those of patients whose regurgitation was controlled by CAVV plasty alone. CAVV regurgitation is associated with increased mortality, whereas CAVV plasty improves survival in patients with FSV by preventing progressive ventricular dysfunction. Numerous studies have demonstrated that the development of atrioventricular valve regurgitation and systemic ventricular dysfunction are closely related.3,5,10,18 Therefore, it is essential to maintain CAVV regurgitation at a level below moderate by performing CAVV plasty. Previous studies have consistently shown that moderate or greater AVV regurgitation is associated with increased mortality and a higher risk of reoperation.19,20 Early CAVV plasty is crucial in patients with FSV. Ono and colleagues2 suggested that early intervention, particularly annuloplasty, preserved and improved ventricular function because annular and ventricular dilatation were associated with ventricular dysfunction. Early CAVV intervention also improves survival in patients with the Fontan circulation.3 However, CAVV repair in patients with FSV remains technically challenging. In particular, the outcomes of AVV surgery before the Glenn procedure are often suboptimal due to low body weight, unfavorable valve anatomy, or abnormalities that involve a complex mechanism of valve regurgitation.10,21,22 Consequently, CAVV regurgitation may remain uncontrollable, leading to increased mortality and failure of Fontan circulation. Additionally, ventricular dysfunction was associated with poor outcomes. Most patients who required CAVV plasty before the Glenn procedure had significant CAVV regurgitation at birth; therefore, annular dilatation was progressive and ventricular function gradually worsened. Several reports have suggested this hypothesis.5,22,23 Therefore, early control of atrioventricular valve regurgitation, particularly before the Glenn procedure, may be essential. When CAVV regurgitation cannot be adequately controlled with CAVV plasty, CAVV replacement may be required. At our institution, CAVV replacement was performed in patients who required CAVV plasty before the Glenn procedure and in those with uncontrollable severe CAVV regurgitation after CAVV plasty. CAVV replacement is generally the final option for patients with FSV because the survival after Fontan completion is significantly lower after CAVV replacement than after CAVV plasty.5 Brown and colleagues15 also reported that FSV is a predictor of mortality for patients who undergo CAVV replacement. Nevertheless, several studies have suggested an acceptable 5-year mortality rate of approximately 65% after CAVV replacement.12, 13, 14, 15, 16 In this study, the survival of patients who underwent CAVV replacement after CAVV plasty was similar to that of patients who underwent CAVV plasty alone, thus supporting the latter results. Moreover, Fontan completion and outcomes after the Fontan procedure were acceptable in this study. Previous studies have reported that the probability of Fontan completion is approximately 60%.1,7,17 Although unbalanced atrioventricular septal defects, CAVV regurgitation, and ventricular dysfunction are known risk factors for unsuccessful Fontan completion,2,24 our findings showed that the Fontan completion rate in patients who underwent CAVV replacement after CAVV plasty, as well as their post-Fontan hemodynamic parameters—including Fontan pressure, ventricular end-diastolic pressure, and cardiac index—and the complications they experienced were comparable to those of patients who underwent CAVV plasty alone. Therefore, if CAVV regurgitation remains uncontrolled—particularly when it exceeds a moderate degree—even after repeated CAVV plasties using various techniques, CAVV replacement may be a reasonable option in patients with FSV. When concomitant volume reduction can be performed, particularly at the time of the Glenn procedure, reattempting valve plasty may be preferable. In contrast, in cases of hemodynamic instability, such as preoperative extracorporeal membrane oxygenation support, a history of cardiopulmonary resuscitation, or clinical signs of heart failure, valve replacement may warrant consideration even before the Glenn procedure. Several complications are associated with CAVV replacement, including atrioventricular block, prosthetic valve failure, and bleeding events. Complete atrioventricular block is an important cause of morbidity after CAVV replacement. Previous reports have shown that the incidence of postoperative pacemaker implantation ranged from 9% to 44%.12,13 Additionally, heterotaxy and smaller valve size implantation were identified as risk factors.13 In this study, no patient in the CAVV plasty group required postoperative permanent pacemaker implantation. In contrast, in the CAVV replacement group, 6 patients (32%) required permanent pacemaker implantation in the early postoperative period, which is comparable to previous reports. However, a relatively high rate of pacemaker implantation was observed in our cohort because 3 patients required implantation of smaller prosthetic valves (16-18 mm), and 2 of the 3 patients had asplenia. Regarding bleeding events, Moon and colleagues25 reported that the rate of freedom from hemorrhagic and thromboembolic complications at 10 years was 79%. In this study, few hemorrhagic events occurred after CAVV replacement. However, valve thrombosis was frequently observed. Reoperation within 5 years after the initial CAVV replacement was required in 4 patients, with a total of 6 procedures performed. Most reoperations were due to thrombosis. In general, prosthetic valve thrombosis may be caused by difficulty in maintaining adequate anticoagulation control and turbulent blood flow resulting from a mismatch between the effective orifice area of the prosthetic valve and low cardiac output. In our cohort, most patients who required early re-replacement had small prosthetic valves (16-20 mm). Although anticoagulation was managed with warfarin, difficulty in maintaining therapeutic anticoagulation may have also contributed to valve thrombosis. Limitations This study had some limitations. First, the statistical power was reduced because of the small sample size. Second, this was a nonrandomized, retrospective, single-center study. Third, because the patients in this study were relatively young and small, cardiac function was assessed solely using echocardiography and catheterization. Fourth, patients in the replacement group exhibited more severe baseline conditions. Regarding the timing of intervention, CAVV surgery was performed at multiple stages of single-ventricle palliation. Consequently, although propensity score matching would have been warranted, it was not feasible because of the limited sample size. Further studies with larger sample sizes are required to confirm these findings. Conclusions Despite severe preoperative conditions, the survival and Fontan completion rate and hemodynamic parameters of patients with FSV who underwent CAVV replacement after CAVV plasty were acceptable in selected patients. These findings suggest that CAVV replacement may be a reasonable therapeutic option for uncontrolled CAVV regurgitation in patients with FSV. Data Availability All data obtained during this study are incorporated in the article. Data are available upon request from the corresponding author. Conflict of Interest Statement The authors reported no conflicts of interest. The Journal policy requires editors and reviewers to disclose conflicts of interest and to decline handling or reviewing manuscripts for which they may have a conflict of interest. The editors and reviewers of this article have no conflicts of interest.