Abnormalities of Cardiac Situs and Heart Disease Diagnosed by Echocardiography in Patients with Biliary Atresia

,

the written consent form was signed, a transthoracic echocardiography was performed in all patients; in case of situs abnormalities, an electrocardiography was performed at the Department of Cardiology.

Inclusion criteria
Patients with BA under 18 years of age and followedup at the Pediatric Hepatology outpatient unit at HC-UFMG during the study period who agreed to take part in the research, regardless of the surgical approach (Kasai procedure or liver transplant), with an acceptable echocardiographic window to determine abdominal situs.

Exclusion criteria
All patients with BA and followed-up at the same unit who did not agree to take part in the research or without an acceptable echocardiographic window.

Data collection
Echocardiographic assessment was performed according to the local routine and to national and international guidelines 10,11 using TOSHIBA APLIO 400 equipment (Tustin, CA, USA). Situs was assessed through longitudinal and transverse subcostal scanning using Doppler ultrasound. The connection between the inferior vena cava (IVC) and the right atrium was investigated at the longitudinal scan. The blood vessel distribution around the spine was identified at the transverse scan (Table 1 and Figure 1).
Electrocardiographic assessment was performed according to national and international guidelines, 12,13 using EDAN SE-3 MHE (Shenzhen, P.R. China) or PHILIPS abdominal situs and estimated by the relative position of large vessels near the spine. 7,8 Several comorbidities of varying severity may be present in patients with LI, such as craniofacial, musculoskeletal, tracheoesophageal, cardiac, and genitourinary malformations as well as malformations of the central nervous system and digestive system (intestinal malrotation, BA). 7,8 In patients with LI, cardiac malformations are usually less significant than those found in cases of right atrial isomerism. Cyanogenic heart diseases are less common, and some cardiac defects may be only partially seen. 7,8 The absence of the right atrium and sinus node (a typical right atrial structure) is also part of the syndrome, resulting in risk of rhythm disorders.
The literature is scarce when it comes to the prevalence of heart disease among patients with BA and LI. Therefore, it is unknown if this population has particular heart conditions when compared to those with left atrial isomerism only.
The objective of this study was to investigate the prevalence of LI and structural heart disease among patients with BA followed-up by the Pediatric Hepatology team at Hospital das Clínicas (HC)/Universidade Federal de Minas Gerais (UFMG), considering that the diagnosis might have been underestimated.

Study design
This was a cross-sectional study involving patients diagnosed with BA at HC/UFMG from February 2016 to April 2020. Once the case was identified and after  PageWriter Trim III (Andover, MA, USA) equipment.
The following variables were reported and analyzed: identification data, patient situation regarding surgical interventions, which had been considered, situs classification, record of associated structural heart disease, and heart rhythm classification in patients with abnormal situs.

Statistical analysis
Data were analyzed using SPSS for Windows, version 16.0 (SPSS Inc., Chicago, IL, USA).
Our descriptive analysis used frequencies or median and interquartile range values for categorical and continuous variables, respectively. A Kolmogorov-Smirnov normality test with a cut-off of 0.05 was used to test for normality of distribution.

Ethical aspects
The present study is in accordance with Resolution 466/2012 of the National Health Council of the Brazilian Ministry of Health, meeting the required ethical standards, including free and informed consent of the participants. The study was approved by the institution's Research Ethics Committee, No. 77 0 203000-09.

Results
Fifty-eight patients were recruited, 3 of whom had been referred from the hospital admissions unit (young infants recently diagnosed with BA). The remaining patients were referred from the Pediatric Hepatology outpatient unit at the same institution.
All 58 recruited patients met the selection criteria and research protocol requirements. Patient data are shown in Table 2.
Data distribution revealed a non-normal pattern (Kolmogorov-Smirnov normality test with p = 0.021).
T h e p a t i e n t s ' m e d i a n a g e a t t h e t i m e o f echocardiography was 3.08 years (interquartile range: 0.83-9.25 years). There was a predominance of female patients, with a 1.5:1 ratio compared to male patients. The median age of patients with situs abnormalities was 5 months (data not shown in the table).
Forty-two patients (72.4%) were in the postoperative period of a Kasai procedure, 10 (17.2%) were in the postoperative period of a liver transplant, and the remaining 6 (10.3%) were in the preoperative period of a Kasai procedure.
Among patients with situs abnormalities, the overall prevalence of congenital heart disease, apart from anomalies of the IVC, was 80% (4/5). Pulmonary valve stenosis (PVS) was the only abnormality found in these patients, with 3/4 or 75% of mild forms (Doppler peak instantaneous gradient between 10mmHg and 35mmHg) and 1/4 or 25% of moderate forms (Doppler peak instantaneous gradient between 36mmHg and 64mmHg). The 4 patients with LI had interrupted IVC (agenesis of the suprahepatic segment) with no clinical implications.
All patients with a heart condition were on regular cardiology follow-ups.
Among patients with situs abnormalities, the overall prevalence of rhythm changes at electrocardiography was 80% (4/5). The most common change was low atrial rhythm (left atrium) (3/4 or 75%). The P-wave axis was shifted to the right (over 120°) in the patient with situs inversus.
Among patients with situs solitus, the overall prevalence of congenital heart disease was 1.9% (1/53): a case of persistent ductus arteriosus (small canal, with no clinical implications).

Discussion
Our sample (58 patients) is small. On the one hand, it reflects the small annual inflow of new cases (6 to 9 patients/year). On the other hand, it reflects the losses to follow-up due to death, transfer to other healthcare facilities, or to the Adult Hepatology unit of the same institution (for patients over 18 years old). Nevertheless, the number of cases is noteworthy as these are records from a single healthcare center.
Despite the availability of an echocardiogram database within the institution, the prospective record was chosen focusing on the diagnosis of situs, implying which implied in repeating the echocardiography in some patients. This strategy was aimed at optimizing the accuracy of the echocardiography in the definition of situs.
The definition of atrial situs depends on the correct identification of atrial appendages, which are structures more consistently related to atrial morphology (right or left). However, appendages are not easily accessed by transthoracic echocardiography, making the direct definition of atrial situs challenging. On the other hand, atrial situs is generally consistent with abdominal and thoracic situs. Even though thoracoabdominal situs discordance may be seen exceptionally, the identification of abdominal situs using conventional echocardiography allows the echocardiographer to infer atrial situs. Therefore, in the present study, situs evaluation was performed through the subcostal view of the echocardiogram, a widely available and non- invasive technique that is reasonably accurate (Figure 2).
We were not able to determine the actual moment of echocardiographic diagnosis of situs, since not all patients had a previous echocardiography record in the institution's database. The median age of patients with situs abnormalities was significantly lower than that of the total sample (5 months compared to 3.08 years). These data suggest that all situs abnormalities were detected early on and were not seen in patients over one year of age at the time of echocardiography. An explanation for these findings is the poor prognosis of patients with polysplenia syndrome. Other authors have reported a lower survival rate with native liver among patients with BA and LI when compared to those with no LI. 14,15 A clear predominance of BA was observed in female patients, reflecting similar findings in the literature. [1][2] The general prevalence of situs abnormalities in the sample (8.6%) was similar to other records (between 7% and 12%), 4,5 with LI being the most common finding (80% of cases); this is in accordance with data in the literature [4][5] . One patient had situs inversus. Other cases of total or abdominal situs inversus in patients with BA have been described. [16][17][18] The prevalence of congenital heart disease was 80% among patients with situs abnormalities and 100% among those with left atrial isomerism (no structural heart disease was detected in the patient with situs inversus). The only structural heart disease found in patients with situs abnormalities, except for the absence of the suprahepatic segment of the IVC (present in almost 100% of patients with LI), was PVS. The prevalence of structural heart disease reported in the literature for patients with left atrial isomerism ranges from 91% to 97%. [19][20][21] Among the cardiac abnormalities of LI found in the literature, the main ones are dextrocardia/mesocardia (24%-42% and 2% of all cases, respectively), anomalous pulmonary venous return (37%-56%), persistent left superior vena cava (33%-59%), unroofed coronary sinus (26%-42%), absence of the suprahepatic segment of the IVC (86%-100%), hepatic venous drainage into the left atrium (32%-41%), common atrium or interatrial communication (80%), total or partial atrioventricular septal defect (49%-80%), univentricular atrioventricular connection (20%-40%), ventricular septal defect (11%), transposition of the great arteries (5%-21%), double-outlet right ventricle (17%-37%), PVS or pulmonary atresia (28%-61%), and aortic stenosis/ atresia or coarctation of the aorta (7%-45%). 19,[22][23][24][25][26][27][28] The literature on the prevalence of heart disease among patients with BA and LI is scarce. Using the MEDLINE, LILACS, and SciELO databases and associating the main descriptors in different manners, our search did not retrieve more than a few case series. Nevertheless, a Canadian study reported 25 cases of polysplenia among 328 patients (7.6%) using a national database of children with biliary atresia and including a 17-year period. Congenital cardiac defects were found in 26 patients. These included pulmonary stenosis (n = 12), ventricular septal defect (n = 10), atrial septal defect (n = 7), patent ductus arteriosus (n = 3), total anomalous pulmonary venous return (n = 3), double-outlet right ventricle (n = 3), bicuspid aortic valve (n = 3), dextrocardia (n = 2), atrioventricular canal defect (n = 2), tetralogy of Fallot (n = 2), partial anomalous pulmonary venous return (n = 1), hypoplastic aortic arch (n = 1), aortic stenosis (n = 1), and mitral stenosis (n = 1). 29 Another report from the United Kingdom, a 28-year single-center retrospective study, described 43 cases of polysplenia among 548 patients (7.8%). Cardiac abnormalities were found in 25 patients, including tetralogy of Fallot (n = 1), aortic arch abnormality and severe left pulmonary hypoplasia (n = 1), aortic coarctation (n = 1), hypoplastic left heart (n = 1), ventricular septal defect (n = 5), atrial septal defect (n = 7), patent ductus arteriosus (n = 7), and pulmonary artery stenosis (n = 1). Nine infants required cardiac surgery. 30 Considering the list of cardiac conditions described in patients affected by LI with or without heterotaxy, it would be reasonable to assume that other conditions might have been observed in this sample. The small absolute number of patients with situs abnormalities in this study prevents any conclusion regarding the real prevalence of cardiac disorders among these patients. Further studies, with larger sample sizes, are required to elucidate this matter in Brazil.
Other conditions of clinical interest associated with LI are sinus node dysfunction (escape rhythm with chronotropic incompetence), supraventricular arrhythmias -tachyarrhythmias and atrioventricular block (up to 71% of all cases, with 7%-22% of atrioventricular block), and the presence of dysfunctional spleens (increasing the risk of infections and sepsis). 7,8,25,26 The overall prevalence of rhythm disorders among patients with situs abnormalities in this sample, at electrocardiographic assessment, was 80% (4/5). The most common one was low atrial rhythm (3/4 or 75%). The incidence of arrhythmias in patients with LI may reach 87% in 3 years of follow-up, with sinus node dysfunction (escape rhythm) being the most common one (around 60% of the cases), followed by total atrioventricular block (20% of all cases). The risk of sinus node dysfunction with an indication for pacemaker implantation reaches 19% in some case series, reinforcing the need for a long-term cardiology follow-up of such patients. 21,31 The abnormalities seen in patients in this study do not seem to lead to increased morbidity at the time of echocardiography. However, the escape rhythm in patients with LI may cause chronotropic incompetence during situations of high cardiac demand.
Despite the low complexity of heart diseases observed in this study and considering the cardiac conditions described in patients affected by BA and LI in larger series, it is possible to infer that the associated cardiac abnormalities may be prognostically important and life-threatening even before biliary malformation can be treated.
The main limitation of this study, apart from the ones inherent to our methodology, are the possible losses of patients related to their death or transfer to other healthcare facilities. At all events, the sample included all patients who were being followed-up by the Pediatric Hepatology team at HC/UFMG during the study period.
The identification of LI is important for the medical specialties that manage most of the care of patients with BA. Even though the diagnosis of isomerism does not represent, at the moment of data collection, a high cardiovascular risk for the patients in our sample, it is important to acknowledge the possibility of late illness and the need for regular cardiology follow-ups.

Conclusion
The prevalence of situs abnormalities in our sample was similar to that described in the literature. As expected, LI was the most common finding.
The prevalence of congenital heart disease in patients with situs abnormalities was high, as anticipated by the analysis of patients with LI only. We found an exclusive prevalence of PVS and a high prevalence of rhythm changes (escape rhythm) in patients with left atrial isomerism. However, due to the small number of affected patients, it is not possible to confirm whether PVS occurs in a preferential association with BA and LI.
Although the diagnosis of isomerism does not increase cardiovascular risk in patients in our sample at first, possible late deterioration should be considered, requiring continued monitoring.