Total anomalous pulmonary venous drainage: Overall results, comparison of the sutureless versus the conventional approach, and predictors of adverse events – A 17-year single-center experience

Abstract

Introduction:

Total anomalous pulmonary venous drainage (TAPVD) constitutes approximately 1.5% of all congenital heart defects. Surgical correction has traditionally been performed using the conventional technique; however, the sutureless approach has emerged as a recent alternative. This study aims to compare the surgical outcomes of the two techniques and to identify the variables influencing them.

Subjects and Methods:

This was a retrospective study over a 17-year period from 2007 to 2024. There were 55 patients of TAPVD: supracardiac (n = 32), intracardiac (n = 18), infracardiac (n = 3), and mixed (n = 2).

Results:

The median age of surgical intervention was 43 days (interquartile range [IQR]: 30–120), and the median weight was 3.40 kg (IQR: 3.00–4.30). Prematurity was present in 3 (5.5%) patients, 11 (19.6%) had preoperative pulmonary vein obstruction, 2 (3.6%) had atrial septal defect obstruction, and 23 (41.1%) had preoperative pulmonary hypertension. Twelve (21.4%) patients required preoperative intubation, 4 (7.1%) required inhaled nitric oxide, and 7 (12.5%) had hemodynamic instability/cardiopulmonary resuscitation (CPR). Surgical correction (supracardiac and infracardiac) was by conventional repair in 20 patients and sutureless in 15 patients. When comparing sutureless and conventional TAPVD repair techniques, no statistically significant differences were observed in rates of pulmonary hypertension crisis, hospital stay duration, noninvasive ventilation duration, postoperative infections, or chest re-exploration rates. Similarly, the incidence of pulmonary venous obstruction (PVO) and early mortality was comparable between the two groups. The predictors of early mortality were seen in patients with preoperative pulmonary hypertension (odds ratio [OR]: 5.63, P = 0.048), preoperative intubation (OR: 9.29, P = 0.008), and preoperative CPR (OR: 36.67, P ≤ 0.001). Incidence of postoperative low cardiac output syndrome was strongly associated with preoperative pulmonary hypertension (OR: 3.90, P = 0.02), while predictors of postoperative extracorporeal membrane oxygenation predominant in patients with preoperative PVO (OR: 24.57, P = 0.007), associated pulmonary abnormality (OR: 4.57, P = 0.031), and preoperative intubation (OR: 3.88, P = 0.05).

Conclusions:

Early postoperative outcomes were found to be closely associated with the patients’ preoperative clinical status. Overall, in our cohort, no difference was seen between the conventional and sutureless technique of TAPVD repair. These findings may be influenced by the study’s limited sample size, potentially reducing its statistical power.

INTRODUCTION

Total anomalous pulmonary venous drainage (TAPVD) is a rare congenital anomaly, accounting for approximately 1.5% of all congenital heart defects, in which the pulmonary veins fail to connect to the left atrium and instead drain into the systemic venous circulation. This anatomical defect necessitates surgical correction early in life, as untreated TAPVD is fatal due to progressive hypoxemia and right heart failure caused by pulmonary venous obstruction (PVO) and pulmonary hypertension.[1,2]

Historically, the conventional surgical repair involves directly anastomosing the pulmonary venous confluence to the left atrium [Figure 1]. While effective in re-establishing pulmonary venous return, this technique carries the risk of suture-line stenosis, particularly in neonates and in patients with small or hypoplastic veins. Postoperative PVO is a serious complication that can arise from this technique, with reported incidence ranging between 8% and 21% and a strong association with increased morbidity and mortality.[4,5]

Figure 1.

Medical illustration comparing infracardiac total anomalous pulmonary venous drainage anatomy, conventional repair, and sutureless repair.[3] TAPVC: Total anomalous pulmonary venous connection

In contrast, the sutureless technique [Figure 1] avoids direct suturing of the pulmonary veins by instead anchoring the left atrial wall to the pericardium surrounding the venous confluence, thereby reducing the likelihood of mechanical distortion and subsequent stenosis.[6,7] Several meta-analyses have supported the benefits of the sutureless method, demonstrating lower rates of postoperative PVO and reintervention.[8,9] Moreover, the technique may offer advantages in early mortality, especially in patients presenting with preoperative PVO, mixed-type anatomy, or severe pulmonary hypertension.[10]

However, the sutureless approach is not without drawbacks. It can lead to peripheral PVO, which, while less severe, may increase the risk of pericardial bleeding or technical complexity in cases with distorted mediastinal anatomy.[11] Furthermore, the long-term implications of using noncontractile pericardial tissue in atrial reconstruction remain unclear, including its impact on left atrial compliance and pulmonary venous hemodynamics.

Given these evolving surgical strategies, there is an ongoing need to define the optimal repair technique based on patient risk profiles and anatomical subtypes. Thus, this study reviews a 17-year single-center experience comparing clinical outcomes between the conventional and sutureless techniques in TAPVD repair. It focuses particularly on rates of PVO, reintervention, perioperative complications, and predictors of early mortality, aiming to identify which surgical method confers better outcomes in specific clinical scenarios.

SUBJECTS AND METHODS

This is a retrospective study which collected and recorded clinical information on patients who underwent TAPVD repair from 2007 to 2024. The overall sample size collected in this study was 55. Baseline characteristics were summarized as means with standard deviations for continuous variables and as counts with percentages for categorical variables. Categorical variables were analyzed using the Chi-square test, whereas t-test was carried out to assess the distribution of continuous variables. Risk factors against mortality were analyzed using univariate and multivariate analysis. Correlation was used for analysis between continuous data. All analyses were done using IBM SPSS Statistics 27.0 (Statistical Package for the Social Sciences, IBM Corp., Armonk, NY).

RESULTS

The median age of patients was 43 weeks with an almost equal gender distribution (51% male and 49% female). 7.3% of patients were syndromic [Table 1]. The majority of the patients were of the supracardiac type of 57.2%, followed by intracardiac, 32.2%, with infracardiac and mixed types being less common. 19.6% of patients had preoperative PVO, 41.1% had pulmonary hypertension, and 21.4% required intubation before surgery. Associated cardiac anomalies were present in 16.1% patients while 8.9% of patients had pulmonary conditions [Table 2].

Table 1.

Demographics (n=55)

		Value, n (%)
Age (n)
2 weeks and below		7
2 weeks to 1 month		14
1 month to 3 months		18
3 months to 5 months		7
5 months to 1 year		6
Above 1 year		3
Median age (weeks)		43 (30–120)
Gender (n)
Male		28 (51.0)
Female		27 (49.0)
Race (n)
Malay		39 (71.0)
Chinese		5 (9.1)
Indian		11 (20.0)
Others		0
Median weight (kg) (IQR)		3.40 (3.00–4.30)
Prematurity (n)
Yes		3 (5.5)
No		52 (94.5)
Gestational age (n)
<36 weeks		1 (1.8)
36 weeks to 36 weeks 6 days		2 (3.6)
37 weeks to 37 weeks 6 days		15 (27.3)
38 weeks to 38 weeks 5 days		31 (56.4)
39 weeks and above		6 (10.9)
Syndromic (n)
Yes		4 (7.3)
No		51 (91.1)

Table 2.

Preoperative data (n=55)

		Value, n (%)
Anatomical type (n)
Supracardiac		32 (57.2)
Intracardiac		18 (32.2)
Infracardiac		3 (5.4)
Mixed		2 (3.6)
Preoperative PVO (n)		11 (19.6)
Restrictive ASD (n)		2 (3.6)
Pulmonary hypertension (n)		23 (41.1)
Single ventricle (n)		1 (1.8)
Isomerism (n)		1 (1.8)
Associated cardiac anomalies		9 (16.1)
Associated pulmonary anomalies		5 (8.9)
Mean LVEF (%)		69.81±8.42
Preoperative intubation		12 (21.4)
Preoperative NIV		4 (7.1)
Preoperative nitric oxide		4 (7.1)
Preoperative ECMO		0
Preoperative CPR/haemodynamically instability		7 (12.5)
Preoperative catheterization		4 (7.1)

ASD: Atrial septal defect, LVEF: Left ventricular ejection fraction, ECMO: Extracorporeal membrane oxygenation, CPR: Cardiopulmonary resuscitation, PVO: Pulmonary vein obstruction, NIV: Noninvasive ventilation

The mean cardiopulmonary bypass (CPB) time was 137 min with an average cross-clamp time of 84 min. Only one patient required intraoperative extracorporeal membrane oxygenation (ECMO) support [Table 3]. Postoperatively, patients required a median of 6 days of invasive ventilation, and the median intensive care unit (ICU) stay was 9 days, with an overall hospital stay averaging 19 days. 30.4% of patients developed infections, including surgical site infections (SSIs), 25% experienced arrhythmias, effusion affected 19.6% of patients, chylothorax in 5.4%, and 33.9% experienced low cardiac output syndrome (LCOS). Pulmonary hypertension was observed in 66.1% of patients, while 12.5% of patients had diaphragmatic paralysis and 3.6% suffered vocal cord palsy [Table 4].

Table 3.

Intra-operative data (n=55)

		Value, n (%)
Mean CPB time (min)		136.64±41.99
Mean cross-clamp time (min)		84.24±28.47
Mean circulatory arrest time (min)		0
Redo sternotomy (n)		0
Type of repair (For supracardiac and infracardiac TAPVD only) (n) (n=35)
Conventional		20 (57.14)
Sutureless		15 (42.86)
PFO left open (n)		31 (55.4)
Delayed chest closure		32 (57.1)
Wean off CPB with ECMO		1 (1.8)

CPB: Cardiopulmonary bypass, PFO: Patent foramen ovale, ECMO: Extracorporeal membrane oxygenation, TAPVD: Total anomalous pulmonary venous drainage

Table 4.

Postoperative data (n=55)

		Value, n (%)
Median postoperative invasive ventilation duration (days) (IQR)		6 (3–11)
Mean postoperative NIV duration (days)		7.84±21.33
Median ICU stay (days) (IQR)		9 (6–15)
Median length of hospital stay (days) (IQR)		19 (11–30)
Median duration chest open (days) (IQR)		1.00 (0–2)
Infection/SSI (n)		17 (30.4)
Chest reopen (n)		4 (7.1)
Arrythmia (n)		14 (25.0)
Effusion (n)		11 (19.6)
Chylothorax (n)		3 (5.4)
Diaphramatic paralysis		7 (12.5)
Vocal cord palsy		2 (3.6)
LCOS		19 (33.9)
Postoperative ECMO		5 (8.9)
Pulmonary hypertension crisis		37 (66.1)

NIV: Noninvasive ventilation, ECMO: Extracorporeal membrane oxygenation, IQR: Interquartile range, ICU: Intensive care unit, SSI: Surgical site infections, LCOS: Low cardiac output syndrome

The early mortality rate was 14.3%, with mean time to death following surgery of 10.9 days. A proportion of patients (10.7%) were discharged with continued respiratory support, such as noninvasive ventilation (NIV) or endotracheal tube to the respective referring hospital for further care [Table 5]. Reintervention was necessary in 7.1% of patients due to postoperative PVO, all managed surgically rather than via catheter-based interventions [Table 6].

Table 5.

Mortality and follow up (n=55)

		Value, n (%)
Early mortality (n)		8 (14.3)
Mean time to death (days)		10.88±6.31
Discharge to tertiary/referring hospital with NIV/Edotracheal intubation (ETT)		6 (10.7)

NIV: Noninvasive ventilation, ETT: Endotracheal tube

Table 6.

Postoperative pulmonary vein obstruction (n=55)

		Value, n (%)
Postoperative PVO (n)		4 (7.1)
Reintervention for PVO (transcath/surgical) (n)		4 (7.1)
Surgical		4
Transcathether		0
Surgical/transcathether intervention for asc/dec vein (n)		1 (1.8)
Surgical/transcathether intervention for ASD closure/widening (n)		1 (1.8)

PVO: Pulmonary vein obstruction, ASD: Atrial septal defect

The conventional or sutureless technique was performed on two groups of patients, which are the supracardiac and infracardiac types. 57.4% of these patients (supracardiac and infracardiac) had conventional repair done, while 42.8% of patients had the sutureless repair. When comparing the two primary repair techniques, there was no statistically significant difference in mean CPB times between the two techniques (139.93 min vs. 142.95 min). The ICU stay, hospital stay, and postoperative noninvasive ventilatory duration were also not statistically significant. Patients that underwent sutureless TAPVD repair experienced lower incidence of LCOS (26.7% vs. 40%), postoperative ECMO (13.3% vs. 15%), and pulmonary hypertension crisis (60% vs. 70%), albeit no statistical significance was demonstrated [Table 7].

Table 7.

Comparision between conventional and sutureless technique for supracardiac and infracardiac total anomalous pulmonary venous drainage

Variable		Sutureless (n=15), n (%)		Conventional (n=20), n (%)		P
Mean CPB time (min)		138.9±21.4		142.9±47.1		0.761
Mean aortic cross-clamp time (min)		95.9±18.8		83.7±26.5		0.138
Mean postoperative invasive ventilation (days)		8.5±8.8		11.1±12.8		0.519
Mean postoperative NIV (days)		7.1±12.5		11.6±33.7		0.628
Mean ICU stay (days)		11.3±10.2		14.0±13.5		0.524
Mean hospital stay (days)		26.5±20.8		30.6±38.8		0.719
Infection/SSI (n)		6 (40)		8 (40)		0.467
Chest re-open (n)		1 (6.7)		3 (15)		0.619
Arrhythmia (n)		2 (13.3)		6 (30)		0.419
Pleural/pericardial effusion (n)		4 (26.7)		5 (25)		N.S
Chylothorax (n)		1 (6.7)		2 (10)		N.S
Diaphragmatic paralysis (n)		1 (6.7)		2 (10)		N.S
Vocal-cord palsy (n)		0		0		N.S
LCOS (n)		4 (26.7)		8 (40)		0.411
Postoperative ECMO (n)		2 (13.3)		3 (15)		N.S
Pulmonary-hypertension crisis (n)		9 (60)		14 (70)		0.537
Early mortality (n)		3 (20)		4 (20)		N.S
Postoperative PVO (n)		1 (6.7)		1 (5)		N.S
Re-intervention for PVO (n)		1 (6.7)		1 (5)		N.S
Ascending/descending vein intervention (n)		0		0		N.S
ASD closure/widening (n)		0		0		N.S

CPB: Cardiopulmonary bypass, LCOS: Low cardiac output syndrome, PVO: Pulmonary vein obstruction, N.S: Not significant, ASD: Atrial septal defect, ICU: Intensive care unit, SSI: Surgical site infections, ECMO: Extracorporeal membrane oxygenation, NIV: Noninvasive ventilation

DISCUSSION

A total of 55 patients were analyzed in this study. The median patient age was 43 weeks. A similar study by Yong et al. yielded a median age of 18 days (1 day to 1 year), which highlighted the need for early TAPVD treatment.[12] As in our series, we believed that the delay in operation was mainly due to the delayed diagnosis. The majority of patients were probably asymptomatic at the initial stage. However, once they started becoming symptomatic, that is when the majority of the patients’ diagnoses were established and thus referred to a tertiary center for surgical interventions. The male-to-female ratio (51% male, 49% female) was similar; however, male predominance was noted in some studies, but data remained inconclusive across broader demographics.[13] The study had 71% Malay patients, reflecting the local demographic profile of patients in Malaysia, where the majority of the population are Malays. This cohort’s TAPVD types supracardiac (57.2%), intracardiac (32.2%), infracardiac (5.4%), and mixed (3.6%) match existing data, where supracardiac TAPVD is the most prevalent type such as that by Al-Radi et al.[14]

CPB and cross-clamp time averaged 136.64 and 84.24 min, respectively, in this group of patients. The median CPB time for TAPVD repair typically ranges between 60 and 120 min, while longer CPB times are often associated with complex cases such as infracardiac and mixed TAPVD due to technical challenges.[12,15] Preoperative PVO significantly impacts CPB time, as additional procedures to relieve obstruction and ensure adequate pulmonary venous flow are necessary.[13] Less complex cases, such as the supracardiac and intracardiac TAPVD, generally have shorter CPB times of around 70–90 min due to their less challenging anatomical features. In contrast, infracardiac and mixed TAPVD require more extensive dissection and complex anastomoses, leading to longer CPB durations, often exceeding 100 min.[16]

In our series, 66.1% of patients developed postoperative pulmonary hypertension crisis. Across literature, pulmonary hypertension is the most common and dreaded postoperative problem occurring in 40%–66% of cases.[15] Due to high pulmonary vascular resistance, patients subsequently develop acute right heart failure and systemic hypoperfusion. Nitric oxide, cautious mechanical ventilation, and ECMO for refractory cases enhance outcomes in high-risk patients.[17] 33.9% of patients developed LCOS, which is similar to numbers from other series.[18] We believe that the reason behind patients developing LCOS are due to myocardial dysfunction, prolonged CPB periods, and inflammatory reactions. The mainstay of treatment for these groups of patients is mainly inotropic support, fluid management, and ECMO whenever necessary.

Arrhythmias occurred in 25% of patients where atrial flutter, junctional ectopic tachycardia, and heart block often develop due to surgery itself, inflammation, or electrolyte imbalances. The overall incidence of sepsis is 30.4%. Infections following TAPVD repair are relatively rare but may occur due to factors such as prolonged surgical time, poor nutritional status, or underlying comorbidities. We believe that the higher rates of sepsis (30.4%) are likely due to the fact that almost half of our patients’ chest were closed in a delayed manner. Incidence of preoperative pulmonary hypertension was 41.1% and 21.4% of preoperative intubation; we decided to keep the chest open postoperatively to manage the pulmonary hypertension and right ventricular dysfunction. Therefore, once these issues have been resolved, the patient will undergo delayed chest closure and thus, increasing the rate of sepsis. In addition, due to prolonged hospital stay, these patients are prone to an array of hospital-acquired infections. Superficial and deep wound infections have been reported, with rates varying based on institutions and surgical techniques. To minimize SSI, stringent aseptic methods, early antibiotic administration, and attentive wound care are essential.

Postoperative PVO, one of the most feared postoperative consequences, occurred in 7.1% of our patients, which is lower than literature’s range 8%–21% of patients undergoing TAPVD repair.[4] Most cases (83%) are usually diagnosed within the first 6 months after surgery, highlighting the importance of early and regular postoperative monitoring.[5] Preoperative PVO is a strong predictor of postoperative PVO, with odds ratios (ORs) as high as 5.27.[19] Anatomical subtypes such as mixed-type and infracardiac TAPVD are more prone to developing PVO postoperatively.[10] Factors, such as younger age, lower weight, later surgery, and prolonged CPB time, further increase the likelihood of PVO.[11] The presence of postoperative PVO significantly reduces survival rates, with a 10-year survival rate of 57.9% in patients with PVO compared to 90.4% in those without.[20] Interventions for PVO include surgical reinterventions and catheter-based therapies.

7.4% of patients who developed PVO postoperatively were managed by surgical intervention in our center. None of the patients underwent catheter-based intervention. In general, PVO reinterventions include balloon angioplasty, stenting, and surgery. Despite being less invasive, balloon angioplasty often causes restenosis.[21] Stenting relieves recurrent or severe PVO immediately; however, in-stent restenosis can occur, especially in younger groups of patients. For multifocal or severe restenosis, surgical revision, sometimes incorporating reanastomosis or sutureless conversion, is the gold standard.[10]

Operative mortality for TAPVD repair varies widely from 2.7% to 35%, depending on factors like the anatomical subtype and the presence of preoperative PVO.[22] This study demonstrated an early mortality rate of 14.1% with a mean time to death following surgery of 10.88 days. We believe that the higher mortality rate is likely due to the critical preoperative condition of the child. Nearly 50% of the children had preoperative pulmonary hypertension, with 7.1% requiring nitric oxide and 12.5% had a history of cardiopulmonary resuscitation (CPR). As the median age of diagnosis is 43 weeks in our series, the delay in diagnosis may be a contributing factor, leading to these critical preoperative states. Neonatal surgery is associated with higher mortality (7.9%–13%) compared to infants.[12] PVO is the leading cause of early and late mortality, with a survival rate of 38.6% at 20 years for patients with PVO, compared to 92.2% for those without.[4] Emergency surgery, low body weight (<2.5 kg), and prolonged CPB time (>90 min) significantly increase mortality risk. After effective repair, 10- and 20-year survival rates are 88%–90%.[12]

In this series, the predictors of early mortality were seen in patients with preoperative pulmonary hypertension (OR: 5.63, P = 0.048), preoperative intubation (OR: 9.29, P = 0.008), and preoperative CPR (OR: 36.67, P ≤ 0.001). Incidence of postoperative LCOS was strongly associated with preoperative pulmonary hypertension (OR: 3.90, P = 0.02) while predictors of need for postoperative ECMO were preoperative PVO (OR: 24.57, P = 0.007), associated pulmonary abnormality (OR: 4.57, P = 0.031), and preoperative intubation (OR: 3.88, P = 0.05).

Comparing between sutureless and conventional TAPVD repair in this series, the sutureless technique showed no statistical difference to the conventional group in terms of pulmonary hypertension crisis, hospital stay, and NIV duration. PVO rates, a major cause of reintervention, were similar in both groups. However, other studies have shown that the sutureless technique has a lower incidence of PVO (3%–5% vs. 13%–17%), reflecting the benefits of a tension-free anastomosis.[8] Conventional repairs are associated with central PVO, while sutureless repairs more commonly result in peripheral PVO, which is generally less severe.[19] Wu et al. also demonstrated a lower early mortality rate (3%–4% vs. 5%–11%) in the sutureless cohort.[9] While conventional repair remains viable for straightforward cases, its higher complication rates emphasize the need for careful patient selection and management. Yoshimura et al. advocate the sutureless techniques for high-risk cases such as preoperative PVO or mixed/infracardiac types, while the conventional technique is used for low-risk cases.[7]

The main drawback in this study is that the retrospective design and single-center setting restrict the study’s generalizability. The small sample size reduces statistical power, especially for surgical method subgroup analyses. Larger, multicenter, long-term follow-up studies are needed to improve surgical and postoperative treatment.

CONCLUSION

In this 17-year study involving 55 patients undergoing TAPVD repair, early postoperative outcomes were found to be closely associated with the patients’ preoperative clinical status. When comparing sutureless and conventional TAPVD repair techniques, no statistically significant differences were observed in rates of pulmonary hypertension crisis, hospital stay duration, NIV duration, postoperative infections, or chest re-exploration. Similarly, the incidence of PVO and early mortality was comparable between the two groups. These findings may be influenced by the study’s limited sample size, potentially reducing its statistical power.

However, preoperative factors such as CPR, mechanical ventilation, and pulmonary hypertension were identified as strong predictors of adverse outcomes, including the need for ECMO and the development of LCOS. These results highlight the critical importance of early risk stratification and suggest that while the sutureless technique may not show broad statistical benefits in this cohort, its selective use in high-risk patients could enhance surgical outcomes and reduce postoperative complications.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

Nil.