Abstract
Background
Total anomalous pulmonary venous connection (TAPVC) is a fatal congenital cardiac anomaly that requires urgent surgical intervention. The development of postoperative pulmonary vein obstruction (PVO) negatively impacts long-term survival.
Objectives
The present study aimed to evaluate the surgical outcomes of TAPVC repair and risk factors associated with postoperative PVO.
Methods
Patients who underwent primary TAPVC repair at our institute between 2004 and 2022 were retrospectively enrolled, and those with right atrial isomerism and single ventricle physiology were excluded. Factors associated with survival and postoperative PVO were analyzed.
Results
A total of 116 patients were enrolled in the present study. The early mortality rate was 6.9%. Nineteen patients (16.4%) developed postoperative PVO within a median time of 59 days of the primary repair, of whom 10 were successfully relieved without any recurrent obstruction. In long-term follow-up, patients with postoperative PVO had significantly lower long-term survival rates than those without postoperative PVO [57.9%, 95% confidence interval (CI) = 34.8-79.5%; vs. 90.4%, 95% CI = 83-96.6% at 10 years, p < 0.001]. Risk factors for postoperative PVO development included lower body weight, younger age, preoperative mechanical ventilation, preoperative inotrope use, and emergency operation.
Conclusions
Postoperative PVO was significantly associated with a higher long-term mortality rate after primary TAPVC repair, with the risk being higher in patients with critical preoperative status. The long-term outcome was good for patients in whom the obstruction was successfully relieved. Early detection and prompt intervention for postoperative PVO after TAPVC repair can improve overall survival in these patients.
Keywords: Congenital heart surgery, Pulmonary venous obstruction, Total anomalous pulmonary vein connection
Abbreviations
ACC, Aortic cross-clamp
ASD, Atrial septal defect
BW, Body weight
CI, Confidence interval
CoA, Coarctation of the aorta
CPB, Cardiopulmonary bypass
CT, Computed tomography
ECMO, Extracorporeal membrane oxygenation
HR, Hazard ratio
IQR, Interquartile range
LA, Left atrium
OR, Odds ratio
PV, Pulmonary vein
PVO, Pulmonary vein obstruction
RA, Right atrium
TAPVC, Total anomalous pulmonary venous connection
VSD, Ventricular septal defect
INTRODUCTION
Total anomalous pulmonary venous connection (TAPVC) is a fatal congenital heart disease which requires urgent surgical repair. The reported surgical mortality rate ranges from 2.7%-14.3%.1-5 According to The Society of Thoracic Surgeons (STS)–European Association for Cardio-Thoracic Surgery (EACTS) Congenital Heart Surgery Mortality Categories (STAT Category), TAPVC repair is assigned to mortality category 4, which indicates a high risk of in-hospital mortality.6 Pulmonary vein obstruction (PVO) is an important cause of late morbidity and mortality in patients who survive TAPVC repair. Currently, surgery is the most effective option to relieve the obstruction.2,4,5,7-11
This study aimed to evaluate the outcomes after surgical correction of TAPVC at our institute, and to assess the risk factors associated with mortality and postoperative PVO.
METHODS
Patients who underwent primary TAPVC repair at our institute between January 2004 and January 2022 were retrospectively enrolled. Patient profiles and follow-up data were retrieved from electronic medical records and telephonic interviews. Those with right atrial isomerism and single ventricle physiology were excluded. This study was approved by the National Taiwan University Hospital Research Ethics Committee (202204031RINA).
The subtypes of TAPVC were defined according to the report by Darling et al.,12 and the diagnosis was based on preoperative echocardiography and electrocardiogram-gated computed tomography (CT). The following clinical variables were included for analysis: TAPVC type, age, body weight (BW), sex, preoperative PVO, prematurity (defined as gestational age < 37 weeks), emergency operation (defined as life-saving surgery performed within 24 hours of diagnosis), preoperative mechanical ventilation, cardiopulmonary bypass (CPB) time, aortic cross-clamp (ACC) time, and associated cardiac anomalies requiring concomitant corrective procedures.
Preoperative PVO was defined as a non-phasic pulmonary vein peak flow velocity > 1.8 m/s or a restrictive ASD < 3 mm11 on echocardiography or diagnosed on preoperative CT, and on the basis of intraoperative findings. Postoperative PVO was defined as an abnormal pulmonary venous flow on echocardiography, as described above, and was further confirmed on CT imaging.
Early mortality was defined as any death occurring within 30 days after primary TAPVC repair, while late mortality was defined as death occurring after 30 days and during the follow-up period.
Surgical technique
TAPVC repair was performed through median sternotomy. The patent ductus arteriosus was ligated. A standard CPB was established, and the patients were routinely cooled to 20 °C, except in cardiac type TAPVC. Under deep hypothermic circulatory arrest, the confluence of the pulmonary veins (PVs) was identified and opened. The incision was extended from the central confluence to the bilateral PV bifurcation, with a provision to further extend into individual PVs in patients with small confluence and PVs. The left atrium (LA) was opened from the posterior wall toward the LA appendage according to the size of the PV confluence. The PV confluence and the LA were then anastomosed. For the supracardiac type, the vertical vein was incised together with the PV confluence and anastomosed to the LA. For the infracardiac type, the vertical vein was usually divided at the level of the diaphragm, and an incision was made from the vertical vein upward to the PV confluence. After LA-PV anastomosis, the right atrium (RA) was opened. The patency of the four PVs was examined through the atrial septal defect (ASD), after which the ASD was closed, usually with the fenestration left open.
In cardiac type TAPVC, mild hypothermia was induced by cooling the patient to 33 °C. The repair was performed through RA atriotomy. For the coronary sinus type, the septum between the coronary sinus ostium and fossa ovalis was excised and the coronary sinus was unroofed toward the LA. The large interatrial defect was repaired with a pericardial patch to redirect the PVs that drained into the left side. For patients with PV drainage into the RA, the intra-atrial septum was excised and a new intra-atrial baffle was reattached to the right side between the PVs and vena cava ostia; thus, the PVs could directly drain into the LA, as described by Hiramatsu et al.13
A sutureless repair was performed in two patients during the primary TAPVC repair by anastomosing the LA edge to the pericardium rather than anastomosing to the PV confluence directly.3,10,14-18
Intervention for postoperative PVO
Surgery for postoperative PVO was performed through a median sternotomy. After adhesiolysis, a standard CPB was established, and the patient was cooled to 20 °C. After identifying the PV and LA anastomosis site, ongoing techniques including scar release, patch augmentation for the stenotic confluence, direct unroofing or enlarging the stenotic PV ostium, or sutureless repair were performed depending on the intraoperative findings of the obstructed PVs. In addition to surgical repair, two patients received stent insertion and balloon dilatation to relieve the obstruction.
Statistical analysis
Categorical variables are presented as percentages and analyzed using the chi-square test or Fisher’s exact test, while continuous variables are presented as median and interquartile range (IQR) and analyzed using the Student’s t-test or the Mann-Whitney U-test depending on whether they were normally distributed or not. Survival outcomes were evaluated using the Kaplan-Meier method and compared using the log-rank test. Logistic regression analysis was conducted to evaluate the risk factors associated with early mortality, and Cox regression analysis was performed to evaluate the association between each independent factor and late mortality. Variables found to be associated with early and late mortality (p < 0.2) were further subjected to multivariable logistic regression and multivariable Cox regression analyses. Statistical significance was defined as a p value of < 0.05. All statistical analyses were performed using MedCalc® Statistical Software version 20 (MedCalc Software Ltd., Ostend, Belgium).
RESULTS
Between January 2004 and January 2022, 152 primary TAPVC repair surgeries were performed at our institute. After excluding 34 patients with right atrial isomerism and two with hypoplastic left heart syndrome, 116 patients were ultimately included in the study.
The median age at surgery in these patients was 10 days (IQR: 3-30 days), with a median bodyweight of 3100 g (IQR: 2676-3480 g). Sixty-five (56%) of the patients were male, and 18 (15.5%) were premature. Seven patients presented with other cardiac anomalies that required concomitant corrective procedures, including ventricular septal defect (VSD) requiring VSD repair (n = 2), coarctation of the aorta (CoA) requiring aortoplasty (n = 2), VSD combined with CoA requiring simultaneous VSD repair and aortoplasty (n = 2), and tetralogy of Fallot requiring Fallot total correction (n = 1). Regarding their preoperative condition, 57 patients (49.1%) required preoperative mechanical ventilation, 3 (2.6%) received preoperative extracorporeal membrane oxygenation (ECMO), and 36 (31%) required preoperative inotropic support. A total of 59 patients (50.9%) had preoperative PVO. The clinical characteristics of the patients are listed in Table 1.
Table 1. Clinical characteristics of 116 patients undergoing TAPVC repair.
| Variables | Number (%) |
| Male | 65 (56%) |
| Body weight (median, IQR, grams) | 3100 (2676-3480) |
| Low body weight (< 2500 grams) | 21 (18.1%) |
| Age at operation (median, IQR, days) | 10 (3-30) |
| Age < 30 days | 85 (73.3%) |
| Prematurity | 18 (15.5%) |
| Type | |
| Supracardiac | 44 (37.9%) |
| Cardiac | 44 (37.9%) |
| Infracardiac | 19 (16.4%) |
| Mixed* | 9 (7.8%) |
| Emergency operation | 53 (45.7%) |
| Preoperative mechanical ventilation | 57 (49.1%) |
| Preoperative PVO | 59 (50.9%) |
| Preoperative extracorporeal membrane oxygenation | 3 (2.6%) |
| Preoperative inotropes | 36 (31%) |
| Concomitant operations# | 7 (6%) |
| CPB time (median, IQR, minutes) | 109 (88-133) |
| ACC time (median, IQR, minutes) | 50 (39-60) |
ACC, aortic cross-clamp; CoA, coarctation of the aorta; CPB, cardiopulmonary bypass; IQR, interquartile range; PVO, pulmonary vein obstruction; TAPVC, total anomalous pulmonary venous connection; VSD, ventricular septal defect.
* The mixed type included cardiac + supracardiac type (n = 7), cardiac + infracardiac type (n = 1), and supracardiac + infracardiac type (n = 1).
# VSD requiring VSD repair (n = 2), CoA requiring aortoplasty (n = 2), combined VSD and CoA requiring simultaneous VSD repair and aortoplasty (n = 2), and tetralogy of Fallot requiring Fallot total correction (n = 1).
Pulmonary venous anatomy
TAPVC was classified as the supracardiac type in 37.9% (n = 44), cardiac type in 37.9% (n = 44), infracardiac type in 16.4% (n = 19), and mixed type in 7.8% (n = 9) of the patients.
The presence of cardiac anomalies was significantly higher in the supracardiac type than in the other types (6/44, 13.6%, p = 0.02). In contrast, the infracardiac type presented more frequently with preoperative PVO (15/19, 78.9%, p < 0.01) and the need for emergency operation (13/19, 68.4%, p < 0.01), while the mixed type required preoperative mechanical ventilation more frequently than the other types (7/9, 77.8%, p < 0.01) (Table 2).
Table 2. Preoperative characteristics of each TAPVC subtype.
| Supracardiac (n = 44) | Cardiac (n = 44) | Infracardiac (n = 19) | Mix (n = 9) | p value | |
| Low bodyweight | 11 (25%) | 5 (11.4%) | 3 (15.8%) | 2 (22.2%) | 0.40 |
| Preoperative inotropes | 15 (34.1%) | 10 (22.7%) | 7 (36.8%) | 4 (44.4%) | 0.44 |
| Preoperative mechanical ventilation | 26 (59.1%) | 13 (29.5%) | 11 (57.9%) | 7 (77.8%) | < 0.01 |
| Emergency operations | 27 (61.4%) | 8 (18.2%) | 13 (68.4%) | 5 (55.6%) | < 0.01 |
| Preoperative PVO | 28 (63.6%) | 10 (22.7%) | 15 (78.9%) | 6 (66.7%) | < 0.01 |
| Associated cardiac anomalies | 6 (13.6%) | 0 | 0 | 1 (11.1%) | 0.02 |
PVO, pulmonary vein obstruction; TAPVC, total anomalous pulmonary venous connection.
Early mortality
Eight early deaths (6.9%) were recorded in the study population, the causes of which included sepsis with multiorgan failure (n = 3), severe left ventricular failure with profound cardiogenic shock (n = 4) and intracranial hemorrhage (n = 1).
The early mortality rate was 13.6% (6/44) for the supracardiac type, 2.3% (1/44) for the cardiac type, 0% for the infracardiac type, and 11.1% (1/9) for the mixed type. The early mortality rate was not significantly different between subtypes (p = 0.10).
In univariable logistic regression analysis, the significant risk factors for early mortality were as follows: low BW [< 2500 g, odds ratio (OR) = 5.35, 95% confidence interval (CI) = 1.22-23.5, p = 0.03], preoperative inotrope use (OR = 4.14, 95% CI = 0.93-18.38, p = 0.049), supracardiac type TAPVC (OR = 5.53, 95% CI = 1.06-28.72, p = 0.02), and associated cardiac anomalies requiring concomitant surgery (OR = 43.5, 95% CI = 6.6-287.44, p < 0.01). However, only patients with associated cardiac anomalies requiring concomitant surgery remained an independent risk factor for early mortality in the multivariable logistic regression analysis (OR = 27.41, 95% CI = 3.24-231.85, p < 0.01) (Table 3).
Table 3. Logistic-regression analysis for risk factors associated with early mortality.
| Variables | Early mortality | |||
| Univariable | Multivariable | |||
| OR (95% CI) | p value | OR (95% CI) | p value | |
| Male | 2.49 (0.48-12.9) | 0.25 | ||
| Age ≤ 30 days at presentation | 1.42 (0.34-5.83) | 0.63 | ||
| Low bodyweight* | 5.35 (1.22-23.5) | 0.03 | 2.32 (0.34-15.83) | 0.39 |
| Prematurity | 0.76 (0.09-6.62) | 0.8 | ||
| Emergency | 3.89 (0.75-20.17) | 0.08 | ||
| Preoperative mechanical ventilation | 1.79 (0.41-7.89) | 0.43 | ||
| Preoperative inotropes* | 4.14 (0.93-18.38) | 0.049 | 2.15 (0.31-15.18) | 0.44 |
| Preoperative pulmonary vein obstruction | 3.11 (0.6-16.12) | 0.15 | 1.94 (0.2-18.59) | 0.56 |
| Concomitant operations* | 43.5 (6.6-287.44) | < 0.01 | 27.41 (3.24-231.85) | < 0.01 |
| TAPVC type | ||||
| Supracardiac* | 5.53 (1.06-28.72) | 0.02 | 2.23 (0.16-30.34) | 0.55 |
| Cardiac | 0.22 (0.03-1.82) | 0.09 | 0.3 (0.04-4.35) | 0.71 |
| Infracardiac | None | |||
| Mixed | 1.79 (0.19-16.37) | 0.63 |
CI, confidence interval; OR, odds ratio; TAPVC, total anomalous pulmonary venous connection.
* Significant risk factors (p < 0.05).
Postoperative pulmonary vein obstruction
Incidence and risk factors for PVO
Postoperative PVO was confirmed in 19 patients (19/116; incidence rate = 16.4%) after primary TAPVC repair. Comparing the patients with and without postoperative PVO, those with PVO had a significantly higher frequency of emergency operations (78.9% vs. 39.2%, p < 0.01), preoperative mechanical ventilation (68.4% vs. 45.4%, p = 0.04), and preoperative inotropic support (52.6% vs. 26.8%, p = 0.02). The incidence of preoperative PVO (68.4% vs. 47.4%, p = 0.09) tended to be higher in the patients with postoperative PVO. In addition, compared to the patients without postoperative PVO, those with obstructed PVs had a lower age (3 vs. 11 days, p < 0.01), lower BW (2960 vs. 3100 g, p < 0.01), longer CPB time (144 vs. 106 minutes, p < 0.01), and longer ACC time (67 vs. 49 minutes, p < 0.01) (Table 4).
Table 4. Comparison of clinical characteristics between patients with and without postoperative PVO.
| Variables | Without postoperative PVO (n = 97) | With postoperative PVO (n = 19) | p value |
| Connection type | |||
| Supracardiac | 37 (38.1%) | 7 (36.8%) | 0.91 |
| Cardiac | 40 (41.2%) | 4 (21.1%) | 0.09 |
| Infracardiac | 13 (13.4%) | 6 (31.6%) | 0.051 |
| Mixed | 7 (7.2%) | 2 (10.5%) | 0.62 |
| Male | 51 (52.6%) | 14 (73.7%) | 0.09 |
| Age (median, days, IQR)* | 11 (4-30) | 3 (1-26.25) | < 0.01 |
| Bodyweight (median, gram, IQR)* | 3100 (2730-3506.25) | 2960 (2557.5-3388.5) | < 0.01 |
| Prematurity | 14 (14.4%) | 4 (21.1%) | 0.47 |
| Emergency surgery* | 38 (39.2%) | 15 (78.9%) | < 0.01 |
| Preoperative mechanical ventilation* | 44 (45.4%) | 13 (68.4%) | 0.04 |
| Preoperative inotropes* | 26 (26.8%) | 10 (52.6%) | 0.02 |
| Preoperative PVO | 46 (47.4%) | 13 (68.4%) | 0.09 |
| Concomitant operation | 6 (6.2%) | 1 (5.3%) | 0.86 |
| Cardiopulmonary bypass time (median, IQR, minutes)* | 106 (86-132) | 144 (113-190) | < 0.01 |
| Aortic cross-clamp time (median, IQR, minutes)* | 49 (38-59) | 67 (47-93) | 0.02 |
| Postoperative extracorporeal membrane oxygenation | 7 (7.2%) | 1 (5.3%) | 0.76 |
| Postoperative arrythmias | 14 (14.4%) | 4 (21.1%) | 0.47 |
IQR, interquartile range; PVO, pulmonary vein obstruction.
* Significantly different parameters (p < 0.05).
Among the subtypes, the infracardiac type had the highest incidence rate of postoperative PVO (6/19, 31.6%), followed by the mixed type (2/9, 22.2%), supracardiac type (7/44, 15.9%), and cardiac type (4/44, 9.1%). Infracardiac type TAPVC tended to be more frequently associated with postoperative PVO (31.6% vs. 13.4%, p = 0.051) than the other types.
Time to develop postoperative PVO
After TAPVC repair, the time to develop postoperative PVO ranged from 31 to 284 days, with a median time of 59 days (IQR: 42-73 days). Among the patients with postoperative PVO, 14 (14/19, 73.7%) developed this complication within 2 months of primary repair (Figure 1), of whom 3 (21.4%) died after the primary repair and 7 (50%) developed recurrent PVO after the first reoperation. The mortality rate for the patients who developed postoperative PVO within 2 months of primary TAPVC repair was significantly higher than that of the patients who developed postoperative PVO ≥ 2 months after the repair (64.3% vs. 0%, p < 0.001).
Figure 1.
Kaplan-Meier analysis for the overall survival probability of 19 patients who presented with postoperative pulmonary vein obstruction (PVO) according to the time of presentation since primary total anomalous pulmonary vein connection (TAPVC) repair. Among the 19 patients, 14 (73.7%) developed postoperative PVO within 2 months, with a mortality rate of 64.3 %. No mortality occurred in the patients who developed postoperative PVO after 2 months.
Management for postoperative PVO
All of the 19 patients who developed postoperative PVO underwent surgery to relieve the obstruction. A flowchart of the management of postoperative PVO is presented in Figure 2.
Figure 2.
Summary of the surgical outcomes of 116 patients after primary total anomalous pulmonary venous connection (TAPVC) repair and the management strategies for postoperative pulmonary vein obstruction (PVO).
The success rate for the intervention was 47.4% (9/19) after the first reoperation without any recurrent PVO during the follow-up period. Among the remaining 10 patients, 3 died after the reoperation, and 7 developed recurrent obstruction. One of the patients underwent balloon dilatation (4 × 20 mm and 6 × 20 mm, SterlingTM, Boston Scientific, Marlborough, Massachusetts, United States) and stent insertion (7 × 16 mm to the right middle PV and 7 × 16 m to the right lower PV, FormulaTM, COOK Medical, Bloomington, Indiana, United States); however, the patient succumbed to pulmonary hemorrhage on the second day after stent insertion.
For the remaining 6 patients with recurrent PVO after the first reoperation, a second reoperation was performed. One (1/6, 16.7%) of these patients survived without any recurrent PVO, 2 died after the second reoperation (2/6, 33.3%), while 3 (3/6, 50%) developed re-obstruction at 1, 2, and 5 months of the surgery, respectively, necessitating a third reoperation. Of these 3 patients, 2 ultimately died 2 and 3 months after the surgery, respectively. The single surviving patient developed recurrent obstruction after 2 months of the third reoperation followed by a fourth reoperation, with intraoperative stent insertion to the right upper PV (FormulaTM, COOK Medical, Bloomington, Indiana, United States). However, it was complicated by a recurrent obstruction within 2 weeks; balloon dilatation (Ultra-softTM SV and MaverickTM 4 × 15 mm, Boston Scientific, Marlborough, Massachusetts, United States) was performed twice but it was not effective. In addition, stent collapse and intimal hyperplasia causing severe in-stent restenosis were noted. A fifth salvage reoperation was performed, but in vain.
Regarding the surgical intervention techniques employed for postoperative PVO, scar release and fibrotic tissue excision were performed in 2 patients without any signs of re-obstruction during follow-up. Direct ostial enlargement with or without patch enlargement was performed in 12 patients, of whom 4 developed re-obstruction requiring further surgery. Sutureless repair was performed in 5 patients. One of these patients (1/5, 20%) survived without any signs of recurrent obstruction, while one (1/5, 20%) died after the first reoperation. The remaining 3 patients (3/5, 60%) developed recurrent obstruction necessitating further sutureless repair for PVO; however, all of them succumbed to the surgery.
Late mortality
The median follow-up time was 4.67 years (range from 0 to 18 years, IQR: 1.58-8.12 years), with an overall postoperative survival rate of 84.7% (95% CI = 79.1-92.3%) at 1 year, 84.7% (95% CI = 79.1-92.3%) at 5 years, and 84.7% (95% CI = 79.1-92.3%) at 10 years (Figure 3A).
Figure 3.
Kaplan-Meier analysis for the overall survival probability for all the patients (A) and patient with and without postoperative pulmonary vein obstruction (PVO) (B).
Overall, 9 deaths occurred after 30 days of the primary TAPVC repair, all of which were within 1 year due to postoperative PVO. Except for 1 patient who required permanent pacemaker insertion due to atrioventricular block, none required a revision surgery during the follow-up period.
In terms of TAPVC subtypes, the cardiac type had the highest overall postoperative survival rate at 10 years after TAPVC repair (88.3%, 95% CI = 78-97.6%) compared to the infracardiac (88.2%, 95% CI = 73.9-99.5%), supracardiac (81.5%, 95% CI = 70.3-93%), and mixed type (75%, 95% CI = 44-99.5%); however, there was no significant difference between the subtypes (p = 0.63) (Figure 4).
Figure 4.
Kaplan-Meier analysis for the overall survival probability between different total anomalous pulmonary venous connection (TAPVC) types.
Univariable Cox regression analysis revealed that among all of the preoperative factors, only inotrope use [hazard ratio (HR) = 4.97, 95% CI = 1.24-19.87, p = 0.02] was significantly associated with late mortality. However, none of the preoperative factors had a significant impact on late mortality in multivariable analysis.
In the patients complicated by postoperative PVO, the overall survival rate was significantly lower than that in the patients without postoperative PVO (57.9%, 95% CI = 34.8-79.5% at 10 years vs. 90.4%, 95% CI = 83-96.6% at 10 years; p < 0.001) (Figure 3B). Cox-regression analysis revealed that postoperative PVO (HR = 6.92, 95% CI = 2.63-18.2, p < 0.01) was a significant postoperative risk factor associated with late mortality (HR = 19.92, 95% CI = 4.12-96.27, p < 0.01) (Table 5).
Table 5. Cox-regression analysis for risk factors associated with late mortality.
| Preoperative factors | Late mortality | |||
| Univariable | Multivariable | |||
| Hazard ratio (95% CI) | p value | Hazard ratio (95% CI) | p value | |
| Male | 2.81 (0.58-13.51) | 0.19 | 2.83 (0.57-13.99) | 0.2 |
| Age ≤ 30 days at presentation | 0.99 (0.98-1) | 0.48 | ||
| Low body weight | 2.9 (0.73-11.64) | 0.13 | 1.98 (0.48-8.15) | 0.34 |
| Prematurity | 1.93 (0.4-9.33) | 0.41 | ||
| Emergency operation | 2.63 (0.66-10.49) | 0.17 | 1.12 (0.24-5.28) | 0.88 |
| Preoperative mechanical ventilation | 3.69 (0.77-17.74) | 0.10 | 1.67 (0.26-10.78) | 0.59 |
| Preoperative inotropes* | 4.97 (1.24-19.87) | 0.02 | 3.31 (0.67-16.46) | 0.14 |
| Preoperative PVO | 2.05 (0.51-8.2) | 0.31 | ||
| Concomitant operations | None | |||
| TAPVC type | ||||
| Supracardiac | 0.53 (0.11-2.55) | 0.4 | ||
| Cardiac | 1.19 (0.32-4.45) | 0.79 | ||
| Infracardiac | 1.33 (0.28-6.43) | 0.72 | ||
| Mixed | 1.6 (0.2-12.82) | 0.68 | ||
| Postoperative factors | ||||
| Postoperative arrythmias | 0.61 (0.08-4.89) | 0.64 | 0.44 (0.05-3.55) | 0.44 |
| Postoperative PVO* | 18.94 (3.93-91.27) | < 0.01 | 19.92 (4.12-96.27) | < 0.01 |
CI, confidence interval; PVO, pulmonary vein obstruction; TAPVC, total anomalous pulmonary venous connection.
* Significant risk factors (p < 0.05).
DISCUSSION
This study analyzed the surgical outcomes of 116 patients who underwent primary TAPVC repair, and demonstrated a 6.9% early mortality rate and 84.7% overall survival rate 10 years after primary repair. Although postoperative PVO was significantly associated with late mortality, the long-term outcome was good for patients who survived the reoperation to relieve PVO.
Early mortality
The early mortality rate after primary TAPVC repair varies among reported studies. An international collaborative population-based study by Seale et al. reported an early mortality rate of 14.3% in 422 patients,1 Husain et al. reported a 9.8% mortality rate in 51 patients,2 Shi et al. reported a 5% mortality rate in 768 patient,4 and Harada et al. reported a rate of 2.7% in 256 patients.5
The early mortality rate in the present study was 6.9%. Previous studies have reported that risk factors including younger age,1 infracardiac type,2,4 mixed type,4 and preoperative PVO2,4 were significantly associated with early mortality. However, in the present study, the only independent risk factor related to early death in the multivariable analysis was the presence of associated cardiac anomalies requiring concomitant corrective procedures (OR = 27.4, 95% CI = 3.23-231.85, p < 0.01). Undertaking concomitant surgeries for associated cardiac anomalies during TAPVC repair increases disease complexity and is technically challenging for surgeons, leading to an increased surgical risk. To address this, perioperative care for patients with TAPVC and other cardiac anomalies should be managed meticulously.
Postoperative PVO
Incidence and risk factors for postoperative PVO
The incidence rates of postoperative PVO reported in previous studies include 17.5% by Seale et al.,1,7 14.4% by Shi et al.,4 and 13.3% by Harada et al.5 Several preoperative and anatomical risk factors have been reported in the literature.
Regarding preoperative and intraoperative factors, preoperative PVO (cause-specific HR = 2.307, 95% CI = 1.598-3.330, p < 0.001) and longer CPB time (cause-specific HR = 1.006, 95% CI = 1.003-1.009, p < 0.001) were reported to be significant risk factors for developing postoperative PVO in a study by Shi et al.4 White et al. found that the severity of preoperative pulmonary obstruction had a significant impact on postoperative PVO in patients with isolated TAPVC (severe vs. mild or less severe preoperative PVO, HR = 10.25, 95% CI = 1.06-98.63, p = 0.044).19 Harada et al. reported that a younger age (p = 0.001), lower body weight (p < 0.001) at the time of surgical repair, and emergency surgery (p = 0.004) were risk factors for postoperative PVO.5
Regarding anatomical factors, Shi et al. reported that mixed type TAPVC (cause-specific HR = 2.020, 95% CI = 1.155-3.533, p = 0.013) and infracardiac-type TAPVC (cause-specific HR = 2.900, 95% CI = 1.632-5.151, p < 0.001) were significantly associated with postoperative PVO.4 Seale et al. found that diffusely small PVs (HR = 6.54, 95% CI = 2.5-17.07, p < 0.001) and the number of lung segments involved (HR = 1.70, 95% CI = 1.03-2.82, p = 0.038) had a significant impact on developing postoperative PVO.7 Interestingly, mixed type TAPVC was also reported to be a significant anatomical risk factor by Seale et al. (OR = 3.15, 95% CI = 1.38-7.22, p = 0.004 in univariable analysis) and White et al. (HR = 7.42, 95% CI = 2.25-24.44, p = 0.001).4,19
In this study, the incidence rate of postoperative PVO was 16.4%. The factors associated with PVO included the need for emergency operations, preoperative mechanical ventilation, preoperative inotropic support, younger age, lower body weight, longer CPB time, and longer ACC time, which is consistent with those reported previously. In addition, preoperative PVOs were more frequently observed in patients who presented with postoperative PVO. These risk factors indicate a critical preoperative status, which is reflective of relatively smaller PVs in these patients; therefore, a certain degree of cardiopulmonary instability is expected prior to the operation. Meanwhile, longer CPB and ACC times in these patients reflect surgical complexity. Although these factors were not found to be significantly associated with early mortality, patients presenting with such critical preoperative conditions should be carefully identified, as they have a higher risk of developing postoperative PVO.
Regarding anatomical factors in our patients, postoperative PVO developed most frequently in the infracardiac type (31.6% vs. 13.4%, p = 0.051). The vertical vein in the infracardiac type was usually long with a smaller PV confluence. In addition, variable confluence configurations could be identified in the patients with infracardiac TAPVC, which may have had a significant impact on the development of postoperative PVO. According to Shi et al., an antler-type PV confluence (defined as left or right upper and inferior PVs joined to form a common vein, with both common veins draining to form a confluence as diagnosed by CT angiography) was significantly related to postoperative PVO compared to an inverted Christmas tree configuration (defined as left or right upper and inferior PVs joined separately along the line of the vertical vein in a symmetric or asymmetric way), with an HR of 2.12 (95% CI = 1.03-5.47, p = 0.002).20 In our patients, the infracardiac type had a significantly higher percentage of preoperative PVO (78.9%) and need for emergency operations (68.4%) compared to the other subtypes, which implies that the PVs in these patients were smaller, longer, and prone to obstruction before the operation. Although the extent of PV confluence was created as large as possible (from the vertical vein upward into PV confluence, and possibly extending into individual PVs) during repair of infracardiac type TAPVC, 31.6% of the patients still developed postoperative PVO, indicating that individual anatomical factors may have had a large impact on the development of postoperative PVO in these patients.
Before the advent of routine preoperative CT imaging, the anatomic details obtained from medical records were insufficient to analyze the anatomical predictive factors for postoperative PVO development in the present study. Further studies are required to identify the possible correlation between individual anatomical factors and postoperative PVO in patients with infracardiac type TAPVC.
Time to develop postoperative PVO
Regarding the timeline of postoperative PVO development, Seale et al. reported a median of 49 days (range from 0 days to 5.2 years) for the diagnosis of postoperative PVO,7 and Harada et al. reported a median of 105 days (IQR: 45-300 days) from the primary repair to initial reoperation for PVO.5 The early presentation of postoperative PVO, especially within 6 months after primary TAPVC repair, has been associated with progressive disease and an unfavorable prognosis, owing to the progressive nature of PVO.5,7,11
In this study, the median time to develop postoperative PVO was 59 days after the primary TAPVC repair. Among the 19 patients who developed postoperative PVO, 14 (73.7%) presented within 2 months of the repair. Moreover, among the patients who developed recurrent obstruction after the first reoperation, all presented with PVO within 2 months after the primary TAPVC repair. The intra-operative findings of these patients revealed diffusely small and stenotic PVs. The overall postoperative mortality rate in these patients was significantly lower than that in the patients who developed PVO ≥ 2 months after primary TAPVC repair (57.1% vs. 0%, p < 0.001). These results support Seale et al.’s hypothesis that diffusely small PVs and shorter time to presentation with obstruction are both risk factors for death.7
Although the outcomes of the patients with early postoperative PVO were poor, 1 patient (16.7%) survived the second reoperation and did not show any evidence of recurrent obstruction. Our study supports the findings described by Seale et al., Ricci et al., and Harada et al. that the early detection and prompt intervention for postoperative PVO5,7,11 are crucial in patients with TAPVC. Aggressive surgery for recurrent obstruction can potentially save lives.7
Management for postoperative PVO
Surgical intervention is the most effective method to relieve postoperative PVO. Several surgical techniques, including sutureless repair, have been reported.2-4,10,11,14-17
The surgical techniques performed in our study included direct ostium unroofing and enlargement, scar and fibrotic tissue excision, patch augmentation, and sutureless repair. The success rate for resolving postoperative PVO after the first reoperation was 47.4%, which is comparable to that reported by Harada et al. (47.6%)5 and Seale et al. (36.3%).7 For patients who successfully survived reoperation without developing recurrent PVO, the long-term outcome was very good.
Sutureless repair, which incorporates the concept of avoiding direct contact between suture lines and the edge of the PV confluence to prevent mechanical stimulation of the suture line, has been shown to be an effective method to relieve postoperative PVO compared to conventional repair.14,15 When applied to primary TAPVC repair, Lo Rito et al. reported that sutureless repair had a lower incidence of developing moderate to severe PVO than conventional repair (2.9% vs. 11.3%, respectively; p = 0.05).3 Shi et al. also demonstrated a significantly lower incidence rate of postoperative PVO in the sutureless group for patients with preoperative PVO (14.3% vs. 33.7%, respectively; p < 0.05),4 especially for patients with the infracardiac type.20 However, progressive peripheral restenosis can still occur in patients undergoing sutureless repair.3
In our study, only 2 patients underwent sutureless repair for primary TAPVC repair. One patient did not develop postoperative PVO and was alive during the follow-up period, while the other developed postoperative PVO and died after reoperation using a sutureless technique. Among the 5 patients who underwent sutureless repair as an intervention for postoperative PVO, the PVO was only successful relieved in 1 (1/5, 20%) without any recurrent obstruction, while the remaining 4 died. However, owing to a limited number of patients, an inference on the efficacy of sutureless techniques for primary repair or to relieve postoperative PVO cannot be made.
Catheter interventions for PVO have also been described previously; however, the outcomes were unsatisfactory.7,10,21-23 In our study, both patients who underwent PV stent insertion and percutaneous balloon dilatation died. The intraoperative findings in 1 patient revealed stent collapse and intimal hyperplasia causing total in-stent occlusion. We believe that surgery should always be considered the first-line treatment for all patients who develop postoperative PVO.
Late mortality
The overall survival rates after primary TAPVC repair were 84.7% at 1 year, 84.7% at 5 years, and 84.7% (95% CI = 79.1-92.3%) at 10 years. All patients who died during the follow-up period died within 1 year of the primary repair, and the cause of late death during the follow-up period was attributed to heart failure resulting from recurrent PVO. The long-term survival in our study was comparable to that of Lo Rito et al., who reported an 83.1% survival rate at 5 years after both conventional repair and sutureless repair (p = 0.73).3 Harada et al. also reported survival rates of 89.1% (95% CI = 84-93%) at 1 year, 86.6% (95% CI = 82-92%) at 5 years, and 85.3% (95% CI = 80-90%) at 20 years.5 The associated risk factors for late mortality included associated cardiac anomalies, age at surgery, and operations conducted before March 1998.5
In the present study, the only independent risk factor associated with late mortality was postoperative PVO (HR = 19.92, 95% CI = 4.12-96.27, p < 0.01). For the patients with postoperative PVO, the survival rate was significantly lower than that in the patients without postoperative PVO (57.9%, 95% CI = 34.8-79.5% vs. 90.4%, 95% CI = 83-96.6% at 10 years, p < 0.001). Although surgical intervention for postoperative PVO is challenging, reducing the incidence rate and successfully relieving the obstruction remains important to improve the overall survival of patients with TAPVC. Surgical techniques to improve the success rate of postoperative PVO should be further studied.
Our study has several limitations. First, it was a single-center retrospective study with a limited number of patients. Second, the surgeries were performed by multiple surgeons, therefore some variations in the surgical techniques are likely. Finally, during the study period, the knowledge and experience of perioperative care for critically ill pediatric patients evolved, as did the diagnostic imaging and surgical techniques. However, their influence on the surgical outcomes could not be quantified in this study.
CONCLUSIONS
Postoperative PVO was significantly associated with a higher long-term mortality rate after primary TAPVC repair, especially in the patients who presented with postoperative PVO within 2 months. The presence of a critical preoperative condition was significantly associated with the development of postoperative PVO. The long-term outcome was very good for patients in whom postoperative PVO was successfully relieved. Early detection and prompt intervention for postoperative PVO after primary TAPVC repair are crucial to mitigate disease progression and improve overall survival in these patients.
DECLARATION OF CONFLICT OF INTEREST
All the authors declare no conflicts of interest.
REFERENCES
- 1.Seale AN, Uemura H, Webber SA, et al. Total anomalous pulmonary venous connection: morphology and outcome from an international population-based study. Circulation. 2010;122:2718–2726. doi: 10.1161/CIRCULATIONAHA.110.940825. [DOI] [PubMed] [Google Scholar]
- 2.Husain SA, Maldonado E, Rasch D, et al. Total anomalous pulmonary venous connection: factors associated with mortality and recurrent pulmonary venous obstruction. Ann Thorac Surg. 2012;94:825–832. doi: 10.1016/j.athoracsur.2012.04.026. [DOI] [PubMed] [Google Scholar]
- 3.Lo Rito M, Gazzaz T, Wilder T, et al. Repair type influences mode of pulmonary vein stenosis in total anomalous pulmonary venous drainage. Ann Thorac Surg. 2015;100:654–662. doi: 10.1016/j.athoracsur.2015.04.121. [DOI] [PubMed] [Google Scholar]
- 4.Shi G, Zhu Z, Chen J, et al. Total anomalous pulmonary venous connection: the current management strategies in a pediatric cohort of 768 patients. Circulation. 2017;135:48–58. doi: 10.1161/CIRCULATIONAHA.116.023889. [DOI] [PubMed] [Google Scholar]
- 5.Harada T, Nakano T, Oda S, Kado H. Surgical results of total anomalous pulmonary venous connection repair in 256 patients. Interact Cardiovasc Thorac Surg. 2019;28:421–426. doi: 10.1093/icvts/ivy267. [DOI] [PubMed] [Google Scholar]
- 6.O'Brien SM, Clarke DR, Jacobs JP, et al. An empirically based tool for analyzing mortality associated with congenital heart surgery. J Thorac Cardiovasc Surg. 2009;138:1139–1153. doi: 10.1016/j.jtcvs.2009.03.071. [DOI] [PubMed] [Google Scholar]
- 7.Seale AN, Uemura H, Webber SA, et al. Total anomalous pulmonary venous connection: outcome of postoperative pulmonary venous obstruction. J Thorac Cardiovasc Surg. 2013;145:1255–1262. doi: 10.1016/j.jtcvs.2012.06.031. [DOI] [PubMed] [Google Scholar]
- 8.Hancock Friesen CL, Zurakowski D, Thiagarajan RR, et al. Total anomalous pulmonary venous connection: an analysis of current management strategies in a single institution. Ann Thorac Surg. 2005;79:596–606. doi: 10.1016/j.athoracsur.2004.07.005. [DOI] [PubMed] [Google Scholar]
- 9.Kelle AM, Backer CL, Gossett JG, et al. Total anomalous pulmonary venous connection: results of surgical repair of 100 patients at a single institution. J Thorac Cardiovasc Surg. 2010;139:1387–1394. doi: 10.1016/j.jtcvs.2010.02.024. [DOI] [PubMed] [Google Scholar]
- 10.Caldarone CA, Najm HK, Kadletz M, et al. Relentless pulmonary vein stenosis after repair of total anomalous pulmonary venous drainage. Ann Thorac Surg. 1998;66:1514–1520. doi: 10.1016/s0003-4975(98)00952-7. [DOI] [PubMed] [Google Scholar]
- 11.Ricci M, Elliott M, Cohen GA, et al. Management of pulmonary venous obstruction after correction of TAPVC: risk factors for adverse outcome. Eur J Cardiothorac Surg. 2003;24:28–36. doi: 10.1016/s1010-7940(03)00180-5. [DOI] [PubMed] [Google Scholar]
- 12.Darling RC, Rothney WB, Craig JM. Total pulmonary venous drainage into the right side of the heart; report of 17 autopsied cases not associated with other major cardiovascular anomalies. Lab Invest. 1957;6:44–64. [PubMed] [Google Scholar]
- 13.Hiramatsu T, Takanashi Y, Imai Y, et al. Atrial septal displacement for repair of anomalous pulmonary venous return into the right atrium. Ann Thorac Surg. 1998;65:1110–1114. doi: 10.1016/s0003-4975(97)01419-7. [DOI] [PubMed] [Google Scholar]
- 14.Yun TJ, Coles JG, Konstantinov IE, et al. Conventional and sutureless techniques for management of the pulmonary veins: evolution of indications from postrepair pulmonary vein stenosis to primary pulmonary vein anomalies. J Thorac Cardiovasc Surg. 2005;129:167–174. doi: 10.1016/j.jtcvs.2004.08.043. [DOI] [PubMed] [Google Scholar]
- 15.Najm HK, Caldarone CA, Smallhorn J, Coles JG. A sutureless technique for the relief of pulmonary vein stenosis with the use of in situ pericardium. J Thorac Cardiovasc Surg. 1998;115:468–470. doi: 10.1016/S0022-5223(98)70294-6. [DOI] [PubMed] [Google Scholar]
- 16.Azakie A, Lavrsen MJ, Johnson NC, Sapru A. Early outcomes of primary sutureless repair of the pulmonary veins. Ann Thorac Surg. 2011;92:666–672. doi: 10.1016/j.athoracsur.2011.04.024. [DOI] [PubMed] [Google Scholar]
- 17.Honjo O, Atlin CR, Hamilton BC, et al. Primary sutureless repair for infants with mixed total anomalous pulmonary venous drainage. Ann Thorac Surg. 2010;90:862–868. doi: 10.1016/j.athoracsur.2010.05.007. [DOI] [PubMed] [Google Scholar]
- 18.Hickey EJ, Caldarone CA. Surgical management of post-repair pulmonary vein stenosis. Semin Thorac Cardiovasc Surg Pediatr Card Surg Annu. 2011;14:101–108. doi: 10.1053/j.pcsu.2011.01.007. [DOI] [PubMed] [Google Scholar]
- 19.White BR, Ho DY, Faerber JA, et al. Repair of total anomalous pulmonary venous connection: risk factors for postoperative obstruction. Ann Thorac Surg. 2019;108:122–129. doi: 10.1016/j.athoracsur.2019.02.017. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Shi G, Zhu F, Wen C, et al. Single-institution outcomes of surgical repair of infracardiac total anomalous pulmonary venous connection. J Thorac Cardiovasc Surg. 2021;161:1408–1417. doi: 10.1016/j.jtcvs.2020.06.023. [DOI] [PubMed] [Google Scholar]
- 21.Seale AN, Daubeney PE, Magee AG, Rigby ML. Pulmonary vein stenosis: initial experience with cutting balloon angioplasty. Heart. 2006;92:815–820. doi: 10.1136/hrt.2005.073973. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Driscoll DJ, Hesslein PS, Mullins CE. Congenital stenosis of individual pulmonary veins: clinical spectrum and unsuccessful treatment by transvenous balloon dilation. Am J Cardiol. 1982;49:1767–1772. doi: 10.1016/0002-9149(82)90257-0. [DOI] [PubMed] [Google Scholar]
- 23.Cullen S, Ho SY, Shore D, et al. Congenital stenosis of pulmonary veins—failure to modify natural history by intraoperative placement of stents. Cardiol Young. 1994;4:395–398. [Google Scholar]