Outcomes of single ventricle palliation in infants with heterotaxy syndrome

Abstract

 

Open in new tabDownload slide

OBJECTIVES

Heterotaxy is a known risk factor for morbidity and mortality in single ventricle palliation. In this study, we examined our experience with this challenging group of patients.

METHODS

Records of patients born between 2001 and 2019 with heterotaxy, who needed staged single ventricle palliation were retrospectively analysed.

RESULTS

A total of 53 patients were included in this study. Thirty-five (66%) patients had a right ventricular dominance, common atrioventricular septal defect was present in 37 (70%) patients. Anomalous pulmonary venous drainage was present in 29 (55%) patients. Forty-six (86%) patients underwent first-stage palliation. Forty-one (77.3%) patients received a bidirectional cavopulmonary connection. Thirty-one (58%) patients received total cavopulmonary connection (TCPC). Overall survival rate was 92% at 1 month, 74% at 1 year and 68% at 10 years. Twelve (22.6%) patients died before second palliation stage. Four (10%) patients died before TCPC. No patient died after TCPC. Independent risk factors for mortality in the multivariate COX regression were a presence of restrictive pulmonary blood flow (HR 3.23; 95% CI 1.02–10.2; P = 0.05) and greater than mild atrioventricular valve regurgitation (HR 3.57; 95% CI 1.27–10.0; P = 0.02).

CONCLUSIONS

Mortality and morbidity in patients with single ventricle and heterotaxy are high. Restrictive pulmonary blood flow needing early modulation and greater than mild atrioventricular valve regurgitation at presentation are independent risk factors for mortality. Total anomalous pulmonary venous connection was not identified as a risk factor in the current era.

INTRODUCTION

Patients with heterotaxy exhibit a wide variety of cardiac and extracardiac malformations and surgical interventions are frequently required. While biventricular repair is possible in a small proportion of these patients, the majority will need a single ventricle palliation [1–3].

A successful single ventricle palliation needs to follow general principles and the following conditions must be achieved: good ventricular function, unobstructed cardiac outflow, unobstructed cardiac inflow, a competent atrioventricular (AV) valve and a stable atrial heart rhythm. In patients with heterotaxy, achieving these conditions is challenging due to the common presence of complex morphological features, such as anomalous systemic venous connections, anomalous pulmonary venous connections, pulmonary stenosis or atresia, AV valve dysfunction and arrhythmias.

Multiple risk factors for mortality in these patients have been identified. Most commonly they include obstruction of pulmonary veins, AV valve regurgitation, obstruction of pulmonary inflow and heart rhythm disturbances [1–9]. Measures such as avoiding excessive volume load, that can lead to AV valve regurgitation through placement of a small Blalock–Tausig shunt; optimizing pulmonary artery development through banding or shunt placement; optimizing pulmonary venous development through repair of total anomalous pulmonary venous connection; and avoiding systemic outflow obstruction through a Damus–Kaye–Stansel type procedure [3, 8], were proposed to achieve better outcomes.

In this study, we analysed our experience with this group of patients in the current era.

MATERIALS AND METHODS

Ethical statement

This study was approved by the institutional review board of the Technical University Munich. Due to the retrospective nature of the study, the need for individual patient consent was waived.

Patients and data collection

We reviewed the medical records of all patients with heterotaxy who underwent single ventricle palliation at the German Heart Center Munich between 2001 and 2019. Patient demographics and characteristics were recorded from medical charts. Operative variables for each stage included operative weight, cardiopulmonary bypass time, type of palliation and concomitant procedures. Postoperative data recorded for each stage included serious adverse events such as reoperations, interventions and the use of extracorporeal membrane oxygenation. Echocardiography and when available angiography data were obtained before and after each stage. The degree of AV valve regurgitation (none/mild versus moderate/severe) was assessed by echocardiography. Reduced ventricular function was determined as systemic ventricular ejection fraction <50%.

Classification and nomenclature

We have used the classification and nomenclature proposed by the International Working Group for Mapping and Coding of Nomenclatures for Paediatric and Congenital Heart Disease in International Paediatric and Congenital Cardiac Code for Heterotaxy: ‘Heterotaxy is synonymous with “visceral heterotaxy” and “heterotaxy syndrome”. Heterotaxy is defined as an abnormality where the internal thoraco-abdominal organs demonstrate abnormal arrangement across the left-right axis of the body. By convention, heterotaxy does not include patients with either the expected usual or normal arrangement of the internal organs along the left-right axis, also known as “situs solitus”, nor patients with complete mirror-imaged arrangement of the internal organs along the left-right axis also known as “situs inversus”’. [6]

We are aware that atrial isomerism as an anatomical term is erroneous and incorrect, as only atrial appendages can be isomeric. Although this term is used widely and loosely, and used by many nomenclature charts, we have maintained the terms of left and right atrial isomerism throughout in this manuscript. We have used the following definition proposed by Jacobs et al. [7].

Left atrial isomerism is defined as a subset of heterotaxy where some paired structures on opposite sides of the left–right axis of the body are symmetrical mirror images of each other, and have the morphology of the normal left-sided structures. This is commonly associated with polysplenia.

Right atrial isomerism is defined as a subset of heterotaxy where some paired structures on opposite sides of the left–right axis of the body are symmetrical mirror images of each other, and have the morphology of the normal right-sided structures. This is commonly associated with asplenia.

All patients had a functional single ventricle or 2 well-formed ventricles, not suitable for a biventricular repair, due to associated cardiac malformations.

Operative techniques

The systemic-pulmonary shunt procedures were performed with or without standard cardiopulmonary bypass on beating heart. A 3.0–5.0 mm Gore-Tex graft was used, depending on the weight of the child [10, 11].

The operative techniques for BCPS [12] and TCPC [13] are described in previous reports. In all patients, antegrade pulmonary blood flow was eliminated at the time of BCPS, if present.

Pulmonary artery banding was performed in a standard fashion through median sternotomy using polytetrafluoroethylene band as previously described [14].

The concomitant procedures included AV valve reconstruction, total anomalous pulmonary venous connection (TAPVC) repair, pulmonary artery enlargement and atrial septectomy. All were completed using cardiopulmonary bypass, with or without cardioplegic arrest and with deep hypothermic circulatory arrest when needed. For AV valve reconstruction, we employed individualized valvuloplasty techniques, most commonly local annuloplasty and commissuroplasty, described before [15]. We have primarily employed surgical AV valve intervention in patients with greater than moderate AV valve regurgitation.

For TAPVC repair, standard ‘left-sided’ or ‘right-sided’ approach with side-to-side anastomosis was used [16].

Follow-up data

The patients were followed-up with paediatric cardiologists in an outpatient setting, and follow-up times were defined per patient as the time from birth to last follow-up. For patients who died, censoring occurred at the time of death. The data were regularly tracked using our institutional single ventricle patient database system. Three patients after stage II were lost to follow-up.

Statistical analysis

Categorical variables are presented as absolute numbers and percentages. Continuous variables are expressed as means ± standard deviations or medians with interquartile ranges (IQR). Estimated survival was determined by the Kaplan–Meier method. The primary outcome was time to death after birth. For multivariable analysis of the primary outcome, we used Cox proportional hazards models to identify the variables at birth that were independently predictive of outcome. Variables with a significance level of <0.1 in univariable Cox modelling were entered into the multivariable Cox regression models. Hazard ratios with 95% confidence intervals were estimated. Data analysis was performed using Statistical Package for the Social Sciences (SPSS), version 25.0 for Windows (IBM, Ehningen, Germany).

RESULTS

Patient attributes and cardiac morphology

We identified 53 patients diagnosed with heterotaxy and functionally single ventricle. Patient characteristics are shown in Table 1. Thirty-two (60%) were male. Birth weight was 3.1 kg (2.7–3.2). Seven patients (13%) were born prematurely. Thirty-five (66%) patients had a right ventricular dominance, 5 (9%) patients had 2 balanced ventricles and the others (25%) had left ventricular dominance. Common AV septal defect was present in 37 (70%) patients, double outlet right ventricle in 19 (36%) and a discordant ventriculo-arterial connection in 35 (66%).

Mild or more than mild AV valve insufficiency at presentation was present in 10 (19%) of the patients (Table 1). Nine (17%) of these patients had a common AV valve.

Anomalous systemic venous drainage was present in 46 (87%) patients, TAPVC was present in 26 (49%) (Table 1). Nine (15%) patients had an obstruction of the pulmonary veins. The obstruction was present in 5 (9%) patients with infracardiac type of TAPVC and in 4 (7%) patients with supracardiac TAPVC. No patients with cardiac type of TAPVC showed obstruction. All patients with supra- and infracardiac type of TAPVC had right isomerism.

Restrictive pulmonary blood flow was found in 43 patients, 20 (37%) with pulmonary valve stenosis and 23 (43%) with pulmonary atresia. Thirty-five of these patients (66%) needed early modulation of pulmonary blood flow, receiving either a systemic to pulmonary shunt or ductal stenting. Seven (13%) patients with pulmonary stenosis had a balanced pulmonary blood flow. Aortic arch obstruction was present in 1 patient (2%). Four (7%) patients had an advanced congenital AV block.

Outcomes after first-stage palliation

First-stage palliation was performed in 46 (86%) patients (Table 2). Median weight was 3.2 kg (3–47) and median age was 23 (8–55) days. Most common palliation was a systemic-pulmonary shunt performed in 32 (70%) patients. Most common type of shunt was a modified Blalock–Tausig shunt in 20 (43%) patients. Twelve (26%) patients received a central aorto-pulmonary shunt. The size of the shunt was dependent on the weight of the child and ranged from 3.0 to 5.0 mm. Pulmonary artery banding in was performed in 8 (17%) patients. Two patients needed a Damus–Kaye–Stansel type procedure.

Twenty-four (52%) patients needed concomitant surgery. TAPVC repair was performed in 14 (30%) patients: Nine patients (17%) with supracardiac TAPVC and 5 (9%) with infracardiac TAPVC. Pulmonary artery enlargement was performed in 11 (24%) patients. Four (9%) patients received pacemaker implantation at first palliation.

Five (11%) patients needed extracorporeal membrane oxygenation support after surgery. Two patients could be weaned from extracorporeal membrane oxygenation. One died on the day after weaning, the second patient died late after BCPC because of liver failure.

After first-stage palliation, 19 (41%) patients needed a total of 25 reoperations and 8 (17%) catheter interventions. Most commonly shunt size adjustments with shunt replacement or ‘clipping’, in 10 (22%) patients. Four (9%) patients needed a catheter intervention due to shunt stenosis. Three (6%) of those patients received a balloon angioplasty at the site of proximal or distal shunt anastomosis. One (2%) patient received a stent implantation on the site of proximal shunt anastomosis.

AV valve reconstruction as reoperation was needed in 4 patients. Of the 3 patients with severe AV valve insufficiency, 1 patient with severe AV valve insufficiency and straddling of the left AV valve received an AV valve replacement. One patient with severe AV valve insufficiency at presentation, complete AV septal defect, and TAPVC, received AV valve reconstruction at the time of BCPC. And one as reoperation after first palliation.

Pulmonary artery enlargement or reconstruction was necessary in 3 patients. Closure of antegrade pulmonary blood flow was needed in 3 patients, and 2 patients needed a systemic outflow reconstruction with a Damus–Kaye–Stansel type procedure. Most of these procedures were combined procedures.

Twelve (22.6%) patients died before reaching second palliation stage. Seven (15%) died early (<30 days after surgery) and 5 (10%) died late (>30 days after surgery) after first-stage palliation (Table 2). All patients died because of cardiac causes.

For 7 (13%) patients, bidirectional cavopulmonary connection (BCPC) was the first palliation stage.

Outcomes after BCPC

Forty-one (77.3%) patients received a BCPC (Table 3). Median age at BCPC was 5.5 (3.3–8) months, and median weight 5.3 kg (4.8–6.5). Twenty-two (54%) patients received bilateral BCPC. Twelve (29%) received bilateral BCPC as stage 2 palliation, 2 patients (5%) had a bilateral BCPC as primary palliation and 8 (19%) patients with bilateral superior vena cava had azygos continuation and received Kawashima procedure. Fourteen (34%) patients in total received a Kawashima procedure. In one patient, a small contralateral superior vena cava was closed at the time of BCPC.

Twenty-nine (70%) had concomitant surgery. Most commonly pulmonary artery or shunt division in 24 (59%), followed by pulmonary artery enlargement in 13 (32%) and atrial septectomy in 8 (20%) patients. Two (5%) patients received AV valve reconstruction.

Six (15%) patients needed reoperation. Four (10%) early after BCPC (<30 days after surgery), and 2 (5%) late (>30 days after surgery) after BCPC. Two needed enlargement of the pulmonary vein connection. One of them received additional AV valve reconstruction. The following procedures were needed in 4 patients: AV valve reconstruction (n = 1), diaphragm plication (n = 1), pulmonary artery enlargement (n = 1), closure of the left persistent superior caval vein (n = 1) and pacemaker implantation (n = 1).

Ten patients (10%) needed catheter interventions. Three (7%) early after BCPC and 7 (17%) late after BCPC. Most common interventions were pulmonary collateral closures (n = 3) and, pulmonary artery dilatation and stenting (n = 2).

No patient died early after BCPC. Four (10%) patients died late after BCPC before reaching total cavopulmonary connection (TCPC). One patient died 2 months after BCPC because of heart failure, 1 patient died 5 months after BCPC because pulmonary vein stenosis, 1 patient was not suitable for TCPC and died because of liver failure 1 year after BCPC and 1 patient experienced sudden cardiac death at home 2 years after BCPC. Three (7%) patients are waiting and are suitable for TCPC.

Outcomes after TCPC

Thirty-one (58%) patients underwent TCPC (Table 3). Median age at TCPC was 2.6 years (2–3.1), median weight was 11.2 kg (10–12.6). Nine patients (29%) needed concomitant surgery. Most commonly AV valve surgery in 7 (22%) patients. Additional procedures were TAPVC correction (n = 1), pulmonary vein stenosis enlargement (n = 1) and pulmonary artery enlargement (n = 1). One patient needed reoperation—diaphragm plication, with closure of a lymphatic fistula. Three patients needed catheter intervention late after TCPC. All 3 patients needed pulmonary artery dilatation with stent implantation. No patients died after TCPC. The outcomes of individual patients are summarized in Fig. 1.

Risk factor analysis for mortality

Overall survival rate was 92% at 1 month, 74% at 1 year and 68% at 10 years (Fig. 2). Risk factors for mortality after birth were analysed for the entire cohort (n = 53). In univariable Cox regression, the risk factors with a P-value of <0.1 were a presence of restrictive pulmonary blood flow needing early modulation and greater than mild AV valve regurgitation after birth. Multivariable Cox regression showed that restrictive pulmonary blood flow needing early modulation (HR 3.23; 95% CI 1.02–10.2; P = 0.05) and a presence of greater than mild AV valve regurgitation at presentation (HR 3.57; 95% CI 1.27–10.0; P = 0.02) were independent predictors for mortality (Table 4). Common AV septal defect, dominant right ventricle and TAPVC were not identified as a risk factors for death.

DISCUSSION

Despite advances in operative and perioperative care, treating single ventricle patients associated with heterotaxy is still challenging. Compared with patients without heterotaxy, single ventricle palliation in patients with heterotaxy is known to have a higher morbidity and mortality [1–8]. In a large series from Australia, the outcomes in these patients are described as disappointing, and no improvement in survival could be shown with time [14]. A smaller study from Anagnostopoulos et al. [17], though showed improved results in current era.

Our study shows that despite advances in pre- and postoperative management, mortality after first palliation stage still remains high, with an overall mortality reaching 22.6%. The proportion of patients requiring a reintervention either surgical or catheter based after the first palliation stage is with 60% also very high. Most reinterventions were necessary to improve the pulmonary blood flow, either with shunt size adjustment, replacing the pulmonary artery band with a shunt and closing the antegrade pulmonary flow, or pulmonary artery enlargement. The next most commonly addressed morbidity was AV valve regurgitation.

The incidence of mild or more than mild AV valve regurgitation at presentation in our group of patients was, with 10 patients (19%) relatively high. Almost all (except one) patients with AV valve insufficiency at presentation had a complete AV septal defect. The incidence is comparable to other published studies [18]. We would usually address a severe AV valve insufficiency at the time of presentation, mild insufficiency would be addressed at each palliation stage, preferably at the time of BCPC. We could identify AV valve insufficiency more than mild at presentation as a risk factor for mortality.

Achieving an optimal pulmonary blood flow seems to be very important in these patients and can be very challenging especially in small patients. The high number of postoperative interventions (41%), where mostly pulmonary blood flow was addressed, shows how difficult it is to achieve a balanced pulmonary blood flow in these patients. We have identified restrictive pulmonary blood flow with a need for early modulation as a risk factor for mortality in our group of patients.

Several different studies have identified TAPVC repair as a significant risk factor for early and late morbidity and mortality [4, 8, 18]. Operative mortality as high as 75% in patients with obstructed TAPVC and right atrial isomerism has been reported [19]. On the contrary, Morales et al. [20] described improving results in patients with TAPVC, single ventricle and heterotaxy. A study from Nakayama et al. [21] has shown that patients with heterotaxy could be more susceptible to development of recurrent pulmonary venous stenosis. It is very difficult to assess the potential for pulmonary outflow obstruction in the setting of TAPVC and pulmonary artery stenosis or atresia. Performing an adequate shunt procedure in this setting can reveal a latent pulmonary outflow obstruction.

Anomalous pulmonary venous drainage was also very common in our group of patients. Most commonly supracardiac type (Table 1). Fifteen per cent of our patient exhibited an obstructive TAPVC. We have adopted a policy of correcting anomalous pulmonary venous connection whenever a potential for obstruction exists. For this reason, we have corrected all patients with extracardiac type of TAPVC. All but 1 patient with extracardiac TAPVC received TAPVC correction at the time of first-stage palliation. Only 2 patients needed reoperation after BCPC due to pulmonary vein stenosis. We feel that this aggressive strategy has contributed to the fact that TAPVC, obstructed or not, could not be identified as a risk factor for mortality in our group of patients.

Reported outcomes after BCPS in heterotaxy patients [22, 23] also show inferior results and higher mortality and morbidity in these patients compared to patients without heterotaxy.

We have observed no early mortality after BCPC. But morbidity included reoperation in 15% and catheter intervention in 24%. These findings show that even after stage II palliation, regulating and achieving an optimal pulmonary blood flow in these patients is the main goal. The frequency of reinterventions also shows that with aggressive treatment it is possible to prepare these patients for the next palliation stage. Most of our patients progressed to or are suitable for TCPC.

We can report that no patients died at or after (longest follow-up of 16 years) TCPC. We would usually aggressively address problems such as AV valve regurgitation, pulmonary artery or vein stenosis the latest at time of TCPC. Postoperative morbidity was low in only 1 patient needing a diaphragm plication and 3 patients needing a catheter reintervention.

Extracardiac manifestations of heterotaxy have been reported to influence mortality and morbidity in this group of patients. A study from Chiu et al. [9] has shown that patients with heterotaxy are at high risk for community-acquired severe bacterial infection and also have high mortality rate whether the spleen is present or not. Our institutional practice is antibiotic prophylaxis and vaccination of all patients with functional asplenia. We have not observed an increased incidence of infections in our group of patients and no patients died from septic complications.

Swisher et al. [24] have observed an increased postsurgical mortality and risk for respiratory complications as compared to control patients in patients with heterotaxy [25]. They have speculated that increased respiratory complications maybe due to ciliary dysfunction. These patients exhibit increased postoperative hospital stay and mechanical ventilation time. We have not observed an increase in respiratory problems in our group of patients, and no patients died because of respiratory complications. It is our institutional practice that all our patients receive postoperative respiratory physiotherapy as a standard.

Intestinal malrotation exhibited by heterotaxy patients has been associated with increased risk for volvulus development. A recent study from White et al. [26] has shown that the incidence of volvulus in the studied group is low and that the outcomes following cardiac surgery for patients with heterotaxy are not impacted by the presence of malrotation. No patients in our study have shown complications as a consequence of intestinal malrotation.

Limitations

This study was limited by its retrospective and single-centre design. Because of the small size of some subgroups, results should be interpreted cautiously because only large effects could be identified.

CONCLUSIONS

In our study, we have analysed a relatively large single-centre group of patients with heterotaxy, needing a staged single ventricle palliation. Consistent with previous studies, our study has shown higher morbidity and mortality in this group of patients. The overall mortality after first palliation stage was 22.6%, and reintervention was required in 41% of patients. The main challenge in these patients is to achieve a balanced pulmonary blood flow that will prepare them for the next palliation stage. We could identify restrictive pulmonary blood flow needing early modulation and more than mild AV valve regurgitation at presentation as risk factors for mortality. In our study, neither TAPVC nor obstructed TAPVC could be identified as risk factors. Aggressive approach to correction of TAPVC could contribute to this fact. We can report that no patients died after TCPC and morbidity was also very low. Addressing problems such as AV valve regurgitation and pulmonary vein stenosis at the time of TCPC is essential for achieving a well-functioning Fontan circulation.

Conflict of interest: none declared.

Author contributions

Janez Vodiskar: Conceptualization; Data curation; Formal analysis; Investigation; Methodology; Writing—original draft; Writing—review & editing. Takashi Kido: Formal analysis; Validation; Writing—review & editing. Martina Strbad: Data curation; Project administration. Julie Cleuziou: Conceptualization; Writing—review & editing. Alfred Hager: Conceptualization; Validation; Writing—review & editing. Peter Ewert: Conceptualization; Supervision. Jürgen Hörer: Conceptualization; Supervision; Validation; Writing—review & editing. Masamichi Ono: Conceptualization; Data curation; Supervision; Validation; Writing—review & editing.

Reviewer information

European Journal of Cardio-Thoracic Surgery thanks Makoto Ando, Adriano Carotti, Sven Dittrich and the other anonymous reviewers for their contribution to the peer review process of this article.

Presented at the 34th Annual Meeting of the European Association for Cardio-Thoracic Surgery, Barcelona, Spain, 8–10 October 2020.

REFERENCES

ABBREVIATIONS

 
• AV

  Atrioventricular

 
• BCPC

  Bidirectional cavopulmonary connection

 
• TCPC

  Total cavopulmonary connection