Clinical significance of persistent left superior vena cava diagnosed in fetal life

Transcription

1 Ultrasound Obstet Gynecol 2007; 30: Published online 6 July 2007 in Wiley InterScience ( DOI: /uog.4045 Clinical significance of persistent left superior vena cava diagnosed in fetal life A. GALINDO*, F. GUTIÉRREZ-LARRAYA, D. ESCRIBANO*, J. ARBUES* and J. M. VELASCO *Ultrasound and Fetal Physiopathology Unit, Department of Obstetrics and Gynaecology and Pediatric Heart Institute, Department of Pediatrics, Hospital Universitario 12 de Octubre, Madrid, Spain KEYWORDS: cardiac defect; heterotaxy; nuchal translucency; outcome; persistent left superior vena cava; prenatal diagnosis ABSTRACT Objectives To determine the prevalence and clinical significance of persistent left superior vena cava (PLSVC) in fetuses with and without cardiac and extracardiac anomalies. Methods Charts and recorded images were reviewed from high-risk patients who underwent fetal echocardiography between January 2000 and December This retrospective study included 54 fetuses with confirmed PLSVC who were diagnosed based on the presence of an additional vessel identified to the left of the pulmonary artery in the three-vessel view of the heart. Associated congenital heart defects (CHDs), extracardiac abnormalities, including first trimester nuchal translucency (NT) thickness, and fetal/postnatal outcome were analyzed. Results Of 5737 referrals, 5233 had a normal heart, and PLSVC was observed in 10 of these fetuses (0.2%; Group 1). CHDs were present in 504 and PLSVC was observed in 44 of these cases (9%). In the latter group, 18/44 (41%) fetuses had heterotaxy syndrome (Group 2) where the most common structural heart defects were atrioventricular septal defect and double-outlet right ventricle. Fetuses without heterotaxy syndrome (Group 3) accounted for 26/44 (59%) cases of CHDs associated with PLSVC. In this group of fetuses the most common CHDs were left outflow tract obstructive defects and conotruncal anomalies. Increased NT was observed in 29%, without differences among the three groups. The survival rates among fetuses in Groups 1, 2 and 3 were 100%, 44% and 50%, respectively. After excluding patients who underwent pregnancy termination, there were no significant differences in the survival rates among the groups, probably due to the small size of the samples. Conclusions PLSVC is associated with CHDs. The identification of PLSVC should prompt a thorough examination of the fetus to identify additional cardiac and extracardiac anomalies. The prognosis of affected fetuses largely depends on whether or not the PLSVC is associated with a CHD. Copyright 2007 ISUOG. Published by John Wiley & Sons, Ltd. INTRODUCTION Persistence of the left superior vena cava (PLSVC) is a relatively infrequent vascular anomaly. However, it is the most common variant of the thoracic venous system. Its prevalence is estimated to be approximately % in individuals with a normal heart 1 3, but its true incidence is unknown, as most often PLSVC has no clinical impact because the systemic venous blood continues to return to the right atrium via the coronary sinus, and usually there is a normal right superior vena cava (RSVC) 4,5. Moreover, this condition is difficult to detect postnatally in routine echocardiographic studies. Unfortunately, the prenatal discovery of a normal RSVC often leads to the belief that the superior systemic venous return is normal, so a more detailed examination is not performed, although the PLSVC can be easily and accurately diagnosed prenatally. In recent years, the addition of the three-vessel view at the upper mediastinum to fetal cardiac examination has facilitated the direct diagnosis of PLSVC, showing a supernumerary vessel to the left of the pulmonary trunk and arterial duct (Figure 1) 6. The importance of a PLSVC lies in a greater prevalence of associated congenital heart defects (CHD). PLSVC occurs in up to 3 8% of patients with CHD, making PLSVC the most common venous anomaly associated Correspondence to: Dr A. Galindo, Department of Obstetrics and Gynaecology, Hospital Universitario 12 de Octubre, Carretera de Andalucía km 5400, Madrid 28041, Spain ( agalindo.hdoc@salud.madrid.org) Accepted: 22 February 2007 Copyright 2007 ISUOG. Published by John Wiley & Sons, Ltd. ORIGINAL PAPER

2 Persistent left superior vena cava 153 PA AAo LSVC AD RSVC Figure 1 Abnormal three-vessel view at 20 weeks showing a supernumerary vessel left of the pulmonary trunk and arterial duct (AD). AAo, ascending aorta; LSVC, left superior vena cava; PA, pulmonary artery; RSVC, right superior vena cava; T, trachea. with CHD 7 10.Therefore,therecognitionofaPLSVC in a screening ultrasound examination should alert the examiner and these patients must be referred for complete fetal echocardiography. This strategy could improve the effectiveness of ultrasound screening for detecting CHD. The aim of this study was to analyze the prenatal characteristics of PLSVC, its association with cardiac and extracardiac anomalies, including increased nuchal translucency (NT), and the outcome of affected patients. METHODS This was a retrospective study performed at a tertiary referral center for prenatal diagnosis and management of fetal and neonatal pathology. The study period was January 2000 December During the study period, 5737 first-, second- and third-trimester echocardiographic examinations were performed in high-risk patients. The majority of the women (n = 3901, 68%) were attending our hospital primarily for standard prenatal care as well as for routine scans and were referred to our unit due to high-risk factors (mainly suspected fetal anomalies at routine scans). The remaining women were referred from their local hospitals under the same criteria, for a detailed anomaly scan, counseling and perinatal management if necessary. Fetal ultrasound examinations included a detailed extracardiac structural survey and a complete echocardiographic examination, performed following standardized guidelines 11. The cardiac scans were carried out in a segmental approach, combining two-dimensional mode and color/pulsed Doppler flow imaging, and included observation of the situs, the position of the heart, the four-chamber-view, the outflow tracts, the three-vessel view, the aortic and ductal arches and both systemic and pulmonary venous returns. The three-vessel view was obtained as originally described 6. Early examinations T were performed in a similar fashion by using both transabdominal and transvaginal ultrasound. Over the study period, the presence of PLSVC was systematically searched for in all fetal echocardiograms. The prenatal diagnosis of PLSVC was made on the basis of an abnormal three-vessel view at the upper mediastinum, showing a supernumerary vessel left of the pulmonary trunk and arterial duct (Figure 1). The diagnosis was then confirmed in the long-axis view, demonstrating direct or indirect drainage via the coronary sinus into the left or right atrium (Figure 2). The coronary sinus was defined arbitrarily by the experienced sonographer as being enlarged when it appeared unusually prominent behind the left atrium in the four-chamber view of the heart, but this was not routinely quantified during the study period. With these criteria, 54 cases of PLSVC were retrieved from our database and analyzed. The following variables were analyzed: reason for referral; gestational age at diagnosis; and associated cardiac, extracardiac and chromosomal abnormalities, including review of the NT thickness in the first trimester. The conditions associated regularly with heterotaxy (e.g. malposition of the abdominal viscera, absence of the gallbladder) were not taken into account when we calculated the rate of associated extracardiac anomalies in fetuses with PLSVC. We also evaluated the presence of fetal growth restriction and/or amniotic fluid volume disorders and fetal/postnatal outcome. All this information was entered prospectively in the appropriate computer database at the time of scanning. If required, additional information was retrieved from recorded videotapes and stored images. Postnatal follow-up for at least 6 months was available for all the survivors of our series. All cases were analyzed together with a pediatric cardiologist. All parents received detailed counseling regarding the diagnosis and therapeutic options after the prenatal ultrasound scan. Follow-up scans every 4 6 weeks were offered in ongoing pregnancies and LV PA CS LSVC LA RSVC DAo Figure 2 Longitudinal view of the persistent left superior vena cava (LSVC) showing drainage into the coronary sinus. CS, coronary sinus; DAo, descending aorta; LA, left atrium; LV, left ventricle; PA, pulmonary artery; RSVC, right superior vena cava.

3 154 Galindo et al. perinatal management in an appropriate tertiary center was strongly recommended to parents, mainly when the fetus had a CHD. The vast majority of the postnatal results (48/54, 89%) were obtained from our own hospital and the remaining results by contacting the parents by telephone or mail. Left isomerism was diagnosed prenatally in the presence of a combination of at least two of the following: azygos continuation of an interrupted inferior vena cava; structural heart disease, with or without fetal heart block; and viscerocardiac heterotaxy. A diagnosis of right isomerism was made when at least two of the following features were observed: structural heart disease; viscerocardiac heterotaxy; juxtaposition of inferior vena cava and aorta on the same side of the spine 12. Viscerocardiac heterotaxy or situs ambiguous was defined as any situs other than situs solitus or situs inversus. The status of the spleen was ascertained postnatally (by radioisotope scan, ultrasound or the presence of Howell Jolly bodies) or at necropsy. The patency of the innominate vein was not specifically addressed postnatally. The thickness of the fetal NT in the first trimester of pregnancy was available in 42/54 cases (78%). This measurement was obtained in a sagittal plane following standard guidelines 13. Observed values were compared with normal reference ranges for gestational age 14 and those 95 th centile of the normal range were considered as abnormal. Fetal karyotyping was performed in 35 cases (65%). In the rest, the karyotype was studied postnatally according to clinical findings. Fetal invasive testing was recommended following standard indications (maternal age > 35 years; family history of aneuploidy; risk on biochemical screening for Down syndrome > 1/270; or ultrasound anomalies, including malformations or NT > 95 th centile). Fluorescence in situ hybridization (FISH) studies to test for 22q11 deletion was also offered whenever a conotruncal anomaly was diagnosed. Ultrasound examinations were performed using highquality equipment (Siemens-Antares, Siemens Medical Solutions, Mountain View, CA, USA; Logic 500, GE Medical Systems, Milwaukee, WI, USA). The odds ratio for CHD in the presence of PLSVC and the odds ratio for the more common types of associated CHD were estimated and presented with 95% confidence intervals (CI). Statistical analysis was performed using the χ 2 and Fisher s exact test. P < 0.05 was considered significant. RESULTS There were 54 cases of fetal PLSVC detected in our center during the observation period. This represents 0.9% of all echocardiographic examinations performed in that period (54/5737). Of 5737 referrals, 5233 fetuses had no structural heart defects and PLSVC was observed in 10 patients (0.2%). By contrast, in almost 9% of all CHD diagnosed prenatally in that period there was associated PLSVC (44/504). Thus, taking into account the whole group of fetuses included in our series, the odds ratio of a fetus having a cardiac defect if a PLSVC was detected was 49.9 (95% CI ). Confirmation of the prenatal diagnosis through necropsy reports or at postnatal echocardiography/surgery was obtained in all cases. There were neither false-positive nor incorrect diagnoses of PLSVC. Forty-four fetuses (81%) had associated CHD, while the remaining 10 fetuses (19%) did not. For an easier and better understanding of the clinical significance of PLSVC, we divided the full study group into three subgroups. The first (Group 1) included those cases without associated structural heart disease and comprised 10 fetuses (19%). The second subgroup (Group 2) included fetuses with PLSVC and heterotaxy or cardiosplenic syndrome and comprised 18 fetuses (33%), 12 with right isomerism and six with left isomerism. Finally, the third subgroup (Group 3) included fetuses with PLSVC and structural heart disease but not in the context of heterotaxy. This was the largest group and comprised 26 fetuses (48%). The case details of the three subgroups, including associated conditions and outcome, are shown in Tables 1 3. The main reason for referral in the three subgroups was suspected CHD. In Group 1, where the fetuses had no structural heart abnormalities, the suspected CHD in almost all cases was atrial septal defect. Table 4 summarizes the most important characteristics related to the diagnosis, associated anomalies and outcome of fetuses with PLSVC for the group as a whole, as well as for each subgroup. All but one fetus in Group 2 had structural heart disease (94%). One fetus with left isomerism showed only dextrocardia and azygos continuation of an interrupted inferior vena cava (Case 21). In 15/17 fetuses (88%), the CHD was atrioventricular septal defect (AVSD). All fetuses with right isomerism had unbalanced AVSD with double-outlet right ventricle; in 8/12 cases (67%) the defect was further complicated by a severe obstructive disease of the pulmonary valve (six fetuses) or of the aortic valve (two fetuses). Almost all fetuses with left isomerism showed azygos continuation of an interrupted inferior vena cava (5/6, 83%) and the most common CHD was also AVSD (50%) (Table 2). When only fetuses with CHD were taken into account, the odds ratio to have a heterotaxy syndrome if a PLSVC is also present was 18.0 (95% CI ). Among the six fetuses with left isomerism, four had polysplenia and two had a normal left-sided spleen. All fetuses with right isomerism had asplenia. A wide spectrum of CHD was observed in fetuses of Group 3, although they could essentially be classified into four groups (Table 3). The first and largest group included 10 fetuses with left outflow tract obstruction (Cases 29, 30, 35, 39, 42, 47, 48, 51, 52 and 54), with the most common CHD being coarctation of aorta (six fetuses). The second group included nine fetuses with conotruncal anomalies (Cases 32, 33, 37, 40, 41, 44, 45, 49 and 53), the most common being pulmonary atresia with ventricular septal defect (four cases), two of them with right aortic arch. The third group included four fetuses

4 Persistent left superior vena cava 155 Table 1 Details of 10 fetuses with persistent left superior vena cava and without associated congenital heart defect (CHD) or heterotaxy syndromes (Group 1) Case MA (years) GA (weeks) Reason for referral NT (mm) Extracardiac anomalies Outcome Increased NT 8.0 No Well CHD suspected (ASD) 4.3 Enlarged cisterna magna Well CHD suspected 1.4 Complete persistence of left umbilical Well vein, agenesis of the ductus venosus Pleural effusion 4.6 Pleural effusion, polyhydramnios Pleuroamniotic shunt, well SUA 1.8 SUA, dorsal hemivertebra Well CHD suspected (ASD) 1.2 SUA Well CHD suspected (ASD) 1.9 No Well CHD suspected (ASD) 1.7 SUA, duodenal atresia, cleft palate Surgery, well Omphalocele 2.0 SUA, omphalocele, Surgery, well Beckwith Wiedemann syndrome CHD suspected (ASD) 1.2 No Well ASD, atrial septal defect; GA, gestational age; MA, maternal age; NT, nuchal translucency; SUA, single umbilical artery. with septal defects (Cases 31, 34, 36 and 46). Finally, the fourth group included three fetuses with complex abnormalities (Cases 38, 43 and 50). PLSVC was observed in 10/136 fetuses with left outflow tract obstruction and in 9/120 fetuses with conotruncal anomalies. Taking into account the whole group of fetuses included in our series, the odds ratio of a fetus having either a cardiac defect of the spectrum of obstructive disease of the left outflow tract or a conotruncal anomaly if PLSVC was detected was 10.0 (95% CI ) and 10.0 (95% CI ), respectively. Finally, for the 89 fetuses with AVSD, of which 18 had PLSVC and 22 a heterotaxy syndrome, the odds ratio of a fetus having a cardiosplenic syndrome if PLSVC was present was 45.7 (95% CI ). A wide spectrum of major or minor extracardiac abnormalities was observed in 26/54 fetuses (48%) (Tables 1 4). Three fetuses (6%) had PLSVC as an isolated finding. The predominant extracardiac anomaly was single umbilical artery (n = 16, 30%) and this was also the most common extracardiac finding in the three subgroups, although it was more common in Group 1 and Group 3 (40% and 38%, respectively) than in Group 2 (11%). Abnormalities of the umbilical vein system, including two fetuses with agenesis of the ductus venosus, were diagnosed in four fetuses and they were distributed equally in the three subgroups (Cases 3, 26, 43 and 54). In 11 of the 54 cases the parents opted for termination of pregnancy (TOP). The overall survival rate at 6 months was 57.4% (31/54) and this rate rose to 72% if TOP was excluded. There were eight neonatal deaths and four infant deaths. The results for each group are summarized in Table 4. There were no differences in the survival rates of Group 2 and Group 3, even after exclusion of TOP. The survival rate in Group 1 was significantly higher than in Group 2 and Group 3, but if TOP were excluded the differences did not reach statistical significance, probably due to the small size of the samples. The survival rate in Group 2 was 44% (62% if TOP was excluded). There were five TOP (28%) and five postnatal deaths (28%). One of these patients (Case 13) died in the neonatal period after the parents refused active management of the newborn. There were two other neonatal deaths: one after Norwood operation (Case 12) and the other after performing a fistula between the innominate artery and right pulmonary artery (Case 26). There were also two infant deaths (Cases 11 and 17), both after performing superior cavopulmonary anastomosis operation in single-ventricle palliation. Corrective or palliative surgical procedures have been performed in the remaining eight babies and these are doing well at the time of writing. Only one patient in this group has not needed any surgical intervention (Case 21). The survival rate in Group 3 was 50% (65% after exclusion of TOP cases). There were six TOP (23%) and seven postnatal deaths (27%). Three newborns died in the neonatal period without any surgical attempt; in two of these patients (Cases 33 and 54) the parents opted for only compassionate, comfort care of the newborn, while the remaining neonate died as a consequence of extreme prematurity (Case 53). There were two neonatal deaths: one after Norwood operation (Case 30) and the other after switch operation and ventricular septal defect closure (Case 40). There were also two infant deaths: one patient died intraoperatively at 7 months during the repair of coarctation of the aorta (Case 29), while the other (Case 41) died at 5 months due to infectious and respiratory complications related to surgical approaches for both esophageal atresia and truncus. Corrective or palliative surgical procedures were performed in all the remaining babies of this subgroup and all these were doing well at the time of writing. DISCUSSION The left superior vena cava is a persistent remnant of a vessel that normally disappears. During the fourth week of gestation, the venous blood of the upper half of the body is drained by way of the bilateral symmetrically

5 156 Galindo et al. Table 2 Details of 18 fetuses with persistent left superior vena cava associated with heterotaxy syndromes (Group 2) Case MA (years) GA (weeks) Reason for referral NT (mm) Heterotaxy Congenital heart defect Extracardiac anomalies Outcome CHD suspected NM RI, asplenia Unbal. AVSD, DORV, PS No Surgery, infant death SUA Surgery, NND CHD suspected 1.6 LI, normal spleen Unbal. AVSD, DORV, AA, Int. IVC, AV block, dextrocardia CHD suspected NM RI, asplenia Unbal. AVSD, DORV-transposition, AA No NND before surgery CHD suspected NM RI, asplenia Unbal. AVSD, DORV, PA No Surgery, well CHD suspected 1.2 RI, asplenia Unbal. AVSD, DORV, PA SUA TOP CHD suspected 2.1 RI, asplenia Unbal. AVSD, DORV-transposition, PA No Surgery, well CHD suspected 1.7 RI, asplenia Unbal. AVSD, DORV, supradiaphragmatic No Surgery, infant death TAPVR, dextrocardia Pregestational DM 3.2 LI, polysplenia ASD, int. IVC No Surgery, well CHD suspected 1.7 RI, asplenia Unbal. AVSD, DORV, dextrocardia No TOP CHD suspected 2.2 RI, asplenia Unbal. AVSD, DORV-transposition No TOP CHD suspected 1.4 LI, normal spleen Dextrocardia, int. IVC No Well CHD suspected 4.0 RI, asplenia Unbal. AVSD, DORV, IAA Type B No Surgery, well CHD suspected NM RI, asplenia Unbal. AVSD, DORV-transposition No Surgery, well CHD suspected 4.2 LI, polysplenia Unbal. AVSD, Int. IVC, AV block, absent RSVC No TOP Previous child with trisomy 21 NM LI, polysplenia AVSD, Int. IVC Jejunal atresia, dorsal hemivertebra CHD suspected 1.8 RI, asplenia Unbal. AVSD, DORV, PA, supradiaphragmatic PAPVR CHD suspected 4.3 RI, asplenia Unbal. AVSD, DORV-transposition, PA, double aortic arch Agenesis of the ductus venosus No TOP Surgery, well Surgery, NND CHD suspected 2.1 LI, polysplenia CoA No Surgery, well AA, aortic atresia; ASD, atrial septal defect; AV block, atrioventricular block; AVSD, atrioventricular septal defect; CHD, congenital heart defect; CoA, coarctation of aorta; DM, diabetes mellitus; DORV, double-outlet right ventricle, normally related great arteries; DORV-transposition, double-outlet right ventricle, transposed great arteries; GA, gestational age; IAA, interrupted aortic arch; int. IVC, interrupted inferior vena cava; LI, left isomerism; MA, maternal age; NM, not measured; NND, neonatal death; NT, nuchal translucency; PA, pulmonary atresia; PAPVR, partial anomalous pulmonary venous return; PS, pulmonary stenosis; RI, right isomerism; RSVC, right superior vena cava; SUA, single umbilical artery; TAPVR, total anomalous pulmonary venous return; TOP, termination of pregnancy; Unbal., unbalanced.

6 Persistent left superior vena cava 157 Table 3 Details of 26 fetuses with persistent left superior vena cava associated with congenital heart defect (CHD) not in the context of heterotaxy syndromes (Group 3) Case MA (years) GA (weeks) Reason for referral Chromosomal defect NT (mm) Congenital heart defect Extracardiac anomalies IUGR Outcome IUGR No NM Sub-Ao VSD, CoA No Yes Surgery, infant death CHD suspected No NM HLHS No No Surgery, NND CHD suspected Trisomy AVSD Mild hydronephrosis No TOP CHD suspected No NM VSD, PA No No Surgery, well CHD suspected No NM VSD, PA Caudal regression syndrome Yes NND CHD suspected Trisomy 21 NM AVSD No No Surgery, well CHD suspected No NM CoA No No Surgery, well Increased NT No 7.0 VSD, LV diverticulum Pentalogy of Cantrell No Surgery, well CHD suspected 22q11 microdeletion 1.7 VSD, PA, RAA No No Surgery, well CHD suspected No 1.1 Tricuspid atresia, TGA SUA No TOP CHD suspected No 1.2 CoA SUA No Surgery, well CHD suspected No 1.8 TGA, VSD No No Surgery, NND CHD suspected No 1.1 Truncus arteriosus Esophageal atresia Type III, PH No Surgery, infant death CHD suspected No 1.8 HLHS, IAA Type B Right kidney agenesis, SUA No Surgery, well CHD suspected No 1.5 Double-outlet single ventricle, TGA, PS Persistence of right umbilical vein No Surgery, well CHD suspected 22q11 microdeletion 2.0 VSD, PA, RAA No No Surgery, well CHD suspected No 2.3 Truncus arteriosus SUA, PH No TOP Increased NT No 7.0 VSD SUA No Surgery, well CHD suspected No NM CoA No No Surgery, well CHD suspected No 2.0 Sub-Ao. VSD, CoA SUA No Surgery, well Increased NT No 4.1 Sub-pulm. VSD, DORV No Yes Surgery, well CHD suspected No 5.0 AVSD, TOF, APVS, RAA SUA, PH No TOP CHD suspected No 1.9 IAA Type B SUA No TOP CHD suspected No 3.8 HLHS No No TOP SUA No 1.1 TOF Mild hydronephrosis, SUA No PTD at 25 weeks, NND CHD suspected Partial trisomy CoA SUA, enlarged cisterna magna, Yes NND agenesis of the ductus venosus APVS; absent pulmonary valve syndrome; AVSD, atrioventricular septal defect; CoA, coarctation of aorta; DORV, double-outlet right ventricle; GA, gestational age; HLHS, hypoplastic left heart syndrome; IAA, interrupted aortic arch; IUGR, intrauterine growth restriction; LV, left ventricle; MA, maternal age; NND, neonatal death; NT, nuchal translucency; NM, not measured; PA, pulmonary atresia; PH, polyhydramnios; PS, pulmonary stenosis; PTD, preterm delivery; RAA, right aortic arch; SUA, single umbilical artery; Sub-Ao, sub-aortic; Sub-pulm., sub-pulmonary; TGA, transposition of great arteries; TOF, tetralogy of Fallot; TOP, termination of pregnancy; VSD, ventricular septal defect.

7 158 Galindo et al. Table 4 Main diagnostic features, associated conditions and outcome for fetuses with persistent left superior vena cava Parameter Total (n = 54) Group 1 (n = 10) Group 2 (n = 18) Group 3 (n = 26) Bilateral superior vena cava (n) Prominent coronary sinus (n (%)) 39 (72) 10 (100) 3 (17) 26 (100) Mean GA at diagnosis (weeks (range)) 26 (18 38) 27 (20 35) 25 (18 38) 27 (19 38) Diagnosis 22 weeks (%)* Increased nuchal translucency (n (%)) 12/42 (29) 3/10 (30) 4/13 (31) 5/19 (26) Chromosomal defects (%) Extracardiac abnormalities (%) Polyhydramnios (maximum vertical pocket > 8 cm) (%) Intrauterine growth restriction (%) TOP (%) TOP (cases diagnosed 22 weeks (%)) Survival rate at 6 months (%) Survival rate at 6 months, excluding TOP (%) *Group 2 vs. Group 1 and Group 2 vs. Group 3, P > Group 1 vs. Group 2, P = 0.01; Group 1 vs. Group 3, P = Group 1 vs. Group 2, P = 0.08; Group 1 vs. Group 3, P = 0.9. GA, gestational age; TOP, termination of pregnancy. arranged anterior cardinal veins. By the eighth week they become connected by an oblique anastomosis (left brachiocephalic or innominate vein) and then the part of the left anterior cardinal vein that is situated below this anastomosis regresses 3,15. While the right anterior and common cardinal veins form the RSVC, the presence of a PLSVC can be attributed to the persistence of the proximal part of the left anterior cardinal vein. The PLSVC runs between the left atrial appendage and the left pulmonary veins, and almost always runs down the back of the left atrium and enters the right atrium through the orifice of an enlarged coronary sinus 16. In 65% the innominate vein is absent or small 17. In approximately 8% of patients the PLSVC drains directly into the left atrium, between the left atrial appendage and pulmonary veins 18. This anomaly is termed complete unroofing of the coronary sinus or coronary sinus atrial septal defect 19. PLSVC represents the most common variation of systemic venous return but its true incidence is not well known. It is estimated that PLSVC occurs in % of the general population 1 3 and our results agree with this estimation: 0.2% of our fetuses without CHD had PLSVC. Although we do not know the real rate of false-negative cases, since postnatal echocardiography was not routinely performed to neonates without prenatal diagnosis of CHD, it seems logical to assume that this rate must be very low, as PLSVC may be easily and accurately detected in fetal echocardiographic examination, with no false-positive cases detected in our series. Postnatal series have reported that the prevalence of PLSVC is greater among patients with CHD, with figures 10 times higher (3 8%) than that observed in the normal healthy population 7 10, making PLSVC the most common venous anomaly associated with CHD. Our results show that this association is even greater in fetal life, with a prevalence that is twice the postnatal rate, and is in full agreement with the 9% prevalence reported from pathological series 20. These differences may be explained by the high neonatal mortality rate of babies born with complex and severe CHD, many of them with associated extracardiac defects, as well as by the impact of TOP. Another factor that may contribute to these disparities between prenatal and postnatal series concerns the diagnosis itself. Postnatally, PLSVC is most often asymptomatic and its detection is difficult during routine transthoracic echocardiographic studies, with many cases being incidental findings However, PLSVC can be easily and accurately recognized prenatally with ultrasound examination. Classically, the prenatal diagnosis of PLSVC has been based upon indirect findings, such as the visualization of an unusually large coronary sinus on two-dimensional transverse views of the fetal thorax, in the four-chamber view or in a more caudal plane, appearing as a dilated space posteriorly in the left atrioventricular groove. Although an enlarged coronary sinus is indeed highly suggestive of PLSVC 24 27, this indirect diagnosis has both false-negative and falsepositive results In recent years, the addition of the three-vessel view at the upper mediastinum to fetal cardiac examinations has facilitated the direct diagnosis of PLSVC, showing four vessels instead of the normal three, with a supernumerary vessel to the left of the pulmonary trunk and arterial duct. In the rare instances where there is absence of the RSVC, there will again be three vessels but abnormally arranged; from right to left, aorta, pulmonary artery and PLSVC. Given the ease of prenatal diagnosis of PLSVC and the close relationship between this venous variation and CHD, we believe that PLSVC may be considered a marker of CHD. As shown by our own results, PLSVC is associated in fetal life with a 50-fold increase in CHD. However, it is noteworthy that none of the cases in our series was referred because of suspected PLSVC in ultrasound screening. Even many of the fetuses with PLSVC and without CHD, all of them with a dilated coronary sinus, were referred because of suspected atrial septal defect and, indeed, it is well known that an enlarged coronary sinus can be mistaken for an atrial septal defect or even an AVSD in fetal life 30.This reflects the fact that many cardiac screening programs are

8 Persistent left superior vena cava 159 still limited to the four-chamber view, with or without visualization of the outflow tracts, but they still have not incorporated the three-vessel view into the routine cardiac views. The addition of the three-vessel view to fetal echocardiographic examinations may also have contributed to the spectrum of CHD observed in our series and others. While previous prenatal series have reported an excess of conotruncal anomalies, septal defects and leftheart obstructive disease in fetuses with PLSVC detected mainly because of an enlarged coronary sinus 2,22,24 27, our results also show that heterotaxy syndromes are overrepresented in fetuses with PLSVC. An analogy with the significance of the double vessel sign in the thorax and upper abdomen or the malposition of the inferior vena cava could then be done 31. PLSVC is a common finding in heterotaxy syndromes, and in almost all cases the coronary sinus is unroofed, and therefore is not enlarged, in spite of the presence of PLSVC 32,33. Although in our study the size of the coronary sinus was assessed subjectively, its enlargement was observed in only 3/18 fetuses with heterotaxy. Structural cardiac defects in the context of heterotaxy syndromes represent a significant subset of the CHD observed in fetuses with PLSVC (41%) and contribute largely to the different spectrum of CHD in our series. As shown by our own results, PLSVC is associated in fetal life with an 18-fold increase in the probability that a cardiac defect is included in heterotaxy syndromes. In these fetuses, the most common structural heart disease was AVSD, mostly with unbalanced ventricular volumes and in many cases with double-outlet right ventricle. These results match well with previously published results 12,29,34,35. In our experience, the combination of AVSD and PLSVC is associated with an almost 50-fold increase in heterotaxy syndromes. Previous papers addressing the significance of PLSVC in fetal life have reported that the most common CHD in fetuses without heterotaxy syndromes was coarctation of aorta 29,36. This correlates well with our own results, although our series, with a larger number of fetuses included, showed that not only coarctation but the wide spectrum of left-heart obstructive diseases are overrepresented in fetuses with PLSVC. Moreover, conotruncal anomalies, the most common being pulmonary atresia with ventricular septal defect, were also equally associated with PLSVC. Both groups of CHD are increased 10-fold in fetuses with PLSVC. This wide spectrum of CHD also correlates well with that observed in postnatal series 37,38. More than 90% of individuals with PLSVC also possess a normal RSVC 15. The combination of PLSVC and absent RSVC is extremely rare 22 and has been reported in 0.05% of autopsied children who died prematurely 39. In all these cases there were associated CHDs, mainly in the context of heterotaxy. Moreover, up to 11% of patients with right isomerism had PLSVC and absent RSVC 40. Our data are in partial agreement with these results, with only one fetus, also having heterotaxy syndrome, showing this anatomical combination, but this represents only 6% of all fetuses with right isomerism seen in the study period. It is now commonly accepted that an increased NT, irrespective of the fetal karyotype, should constitute an indication for fetal echocardiography, as in this group of fetuses the prevalence of major CHD is higher than in the general population 41. Our interesting observation that up to 29% of the fetuses with PLSVC had an increased NT in the first trimester provides further evidence of the strong association between increased NT and cardiovascular defects. In accordance with the still poorly understood mechanism underlying the relationship between increased NT and CHD, we did not observe any differences in the rate of increased NT among the three subgroups (Group 1, 30%; Group 2, 31%; Group 3, 26%). However, since the mean gestational age at diagnosis of CHD in our series was relatively late, with only 19/44 (43%) of the cases being detected 22 weeks, it is important to re-emphasize that the detection of increased NT in the first trimester should prompt referral for fetal echocardiography, as this strategy may facilitate the early diagnosis of CHD in a substantial proportion of fetuses with CHD 42. Nevertheless, as shown by our results, increased NT does not seem to discriminate between the presence or absence of CHD in fetuses with PLSVC. The association between PLSVC and chromosomal defects has already been reported, and some authors have advocated that fetal karyotype should be routinely offered whenever PLSVC is detected prenatally 25,27. However, most of the aneuploid fetuses had associated cardiac and/or extracardiac defects. In our series, all the chromosomal defects were also detected in fetuses with CHD not in the context of heterotaxy, suggesting that the cardiac defects themselves are the true conditions associated with chromosomal disorders and not the PLSVC. Therefore, in our experience, PLSVC itself should not constitute an indication for fetal karyotype. Our results agree with the concept that, among fetuses with CHD, abnormal visceral situs is strongly predictive of a normal karyotype, and no chromosomal defect was observed in fetuses with heterotaxy syndrome 12,35,43. Especially noteworthy is the high rate and wide spectrum of associated extracardiac defects observed in the three subgroups. These observations are similar to those reported by others who have considered PLSVC as a marker, not only of CHD but also of fetal congenital defects 25,29. Although the overall rate of PLSVC in our series correlates well with the prevalence observed in the general population, it may be interpreted that our results may be biased by the intrinsic high-risk characteristics of our study population. Many of the fetuses in our series and most of those with PLSVC (even in Group 1), were referred because of suspected cardiac or extracardiac anomalies on routine scans; therefore, it is highly probable that the rate of associated defects with PLSVC seen in fetal life is greater than that in the healthy population. It is likely that the addition of the three-vessel view to the routine cardiac screening examination may lead in

9 160 Galindo et al. the future to the detection of more fetuses with isolated PLSVC. To conclude, our study provides further information concerning PLSVC and confirms that this venous variation can be accurately diagnosed by fetal echocardiography. Our series, the largest one reported to date from a single center, demonstrates that PLSVC is a powerful marker of fetal CHD and, therefore, its recognition in ultrasound screening should prompt the referral of these patients for fetal echocardiography as well as a thorough fetal anatomical survey. However, the real efficiency of this marker in screening protocols must be established in further studies performed in unselected populations. If such efficiency is confirmed, this marker might improve the effectiveness of ultrasound screening for CHD. Leftheart obstructive diseases, conotruncal anomalies and AVSD in the context of heterotaxy syndromes are the most common defects in fetuses with PLSVC. Although the differences in survival rates among the three subgroups were not statistically significant, probably due to the small size of the samples, the outlook for fetuses with PLSVC mainly depends on the presence or absence of an associated CHD and its severity. If isolated, PLSVC is a benign vascular anomaly. REFERENCES 1. Steinberg I, Dubilier W, Lukas D. Persistence of left superior vena cava. Dis Chest 1953; 24: Hairston P. Left superior vena cava to left atrial drainage associated with double outlet right ventricle. Arch Surg 1969; 98: Cherian SB, Ramesh BR, Madhyastha S. Persistent left superior vena cava. Clin Anat 2006; 19: Pahwa R, Kumar A. Persistent left superior vena cava: an intensivist s experience and review of the literature. South Med J 2003; 96: Sarodia BD, Stoller JK. Persistent left superior vena cava: case report and literature review. Respir Care 2000; 45: Yoo SJ, Lee YH, Kim ES, Ryu HM, Kim MY, Choi HK, Cho KS, Kim A. Three-vessel view of the fetal upper mediastinum: an easy means of detecting abnormalities of the ventricular outflow tracts and great arteries during obstetric screening. Ultrasound Obstet Gynecol 1997; 9: Mantini E, Grondin CM, Lillehei CW, Edwards JE. Congenital anomalies involving the coronary sinus. Circulation 1966; 33: Cha EM, Khoury GH. Persistent left superior vena cava. Radiology 1972; 103: Buirski G, Jordan SC, Joffe HS, Wilde P. Superior vena cava abnormalities: their occurrence rate, associated cardiac abnormalities and angiographic classification in a paediatric population with congenital heart disease. Clin Radiol 1986; 37: Cochrane AD, Marath A, Mee RB. Can a dilated coronary sinus produce left ventricular inflow obstruction? An unrecognized entity. Ann Thorac Surg 1994; 58: Allan LD. The normal fetal heart. In Fetal Cardiology, Allan LD, Hornberger L, Sharland G (eds). Greenwich Medical: London, 2000; Berg C, Geipel A, Smrcek J, Krapp M, Germer U, Kohl T, Gembruch U, Baschat AA. Prenatal diagnosis of cardiosplenic syndromes: a 10-year experience. Ultrasound Obstet Gynecol 2003; 22: Nicolaides KH, Azar G, Byrne D, Mansur C, Marks K. Fetal nuchal translucency: ultrasound screening for chromosomal defects in first trimester of pregnancy. Br Med J 1992; 304: Pandya PP, Snijders RJM, Johnson SP, Brizot ML, Nicolaides KH. Screening for fetal trisomies by maternal age and fetal nuchal translucency thickness at weeks of gestation. Br J Obstet Gynaecol 1995; 102: Peltier J, Destrieux C, Desme J, Renard C, Remond A, Velut S. The persistent left superior vena cava: anatomical study, pathogenesis and clinical considerations. Surg Radiol Anat 2006; 28: Mantini E, Grondin CM, Lillehei CW, Edwards JE. Congenital anomalies involving the coronary sinus. Circulation 1966; 33: Moore KL. The circulatory system. In The Developing Human. Clinically Oriented Embryology, Moore KL (ed.). WB Saunders: Philadelphia, PA, 1982; Fraser RS, Dvorkin J, Rossall RE, Eidem R. Left superior vena cava: a review of associated congenital heart lesions, catheterization data and roentgenologic findings. Am J Med 1961; 31: Vydt T, Cools F, Rademakers FE. Absent right and persistent left superior vena cava. Acta Cardiol 2003; 58: Nsah EN, Moore GW, Hutchins GM. Pathogenesis of persistent left superior vena cava with a coronary sinus connection. Pediatr Pathol 1991; 11: Menendez B, Garcia del Valle S, Marcos RC, Azofra J, Gomez- Arnau J. Left superior vena cava: a vascular abnormality discovered following pulmonary artery catheterization. Can J Anaesth 1996; 43: Pasquini L, Belmar C, Seale A, Gardiner HM. Prenatal diagnosis of absent right and persistent left superior vena cava. Prenat Diagn 2006; 26: Larsen AI, Nilsen DW. Persistent left superior vena cava. Use of an innominate vein between left and right superior caval veins for the placement of a right ventricular lead during ICD/CRT implantation. Eur Heart J 2005; 26: Rein AJ, Nir A, Nadjari M. The coronary sinus in the fetus. Ultrasound Obstet Gynecol 2000; 15: Machevin-Surugue E, David N, Verspyck E, Labadie G, Blaysat G, Durand I, Ickowicz V, Marpeau L. Dilated coronary sinus in prenatal echocardiography; identification, associations and outcome. Prenat Diagn 2002; 22: Chaoui R, Heling KS, Kalache KD. Caliber of the coronary sinus in fetuses with cardiac defects with and without left persistent superior vena cava and in growth-restricted fetuses with heart-sparing effect. Prenat Diagn 2003; 23: Kalache KD, Romero R, Conoscenti G, Qureshi F, Jacques SM, Chaiworapongsa T, Treadwell M, Johnson A. Prenatal diagnosis of dilated coronary sinus with persistent left superior vena cava in a fetus with trisomy 18. Prenat Diagn 2003; 23: Papa M, Camesasca C, Santoro F, Zoia E, Fragasso G, Giannico S, Chierchia SL. Fetal echocardiography in detecting anomalous pulmonary venous connection: four false positive cases. Br Heart J 1995; 73: Berg C, Knuppel M, Geipel A, Kohl T, Krapp M, Knopfle G, Germer U, Hansmann M, Gembruch U. Prenatal diagnosis of persistent left superior vena cava and its associated congenital anomalies. Ultrasound Obstet Gynecol 2006; 27: Park JK, Taylor DK, Skeels M, Towner DR. Dilated coronary sinus in the fetus: misinterpretation as an atrioventricular canal defect. Ultrasound Obstet Gynecol 1997; 10: Sheley RC, Nyberg DA, Kapur R. Azygous continuation of the interrupted inferior vena cava: a clue to prenatal diagnosis of the cardiosplenic syndromes. J Ultrasound Med 1995; 14: Van Praagh S, Kreutzer J, Alday L, Van Praagh R. Systemic and pulmonary venous connections in visceral heterotaxy with emphasis on the diagnosis of the atrial situs: a study of 109 postmortem cases. In Developmental Cardiology:

10 Persistent left superior vena cava 161 Morphogenesis and Function, Clark EB, Takas A (eds). Futura: New York, 1990; Rubino M, Van Praagh S, Kadoba K, Pessotto R, Van Praagh R. Systemic and pulmonary venous connections in visceral heterotaxy with asplenia. Diagnostic and surgical considerations based on seventy-two autopsied cases. J Thorac Cardiovasc Surg 1995; 110: Lin JH, Chang CI, Wang JK, Wu MH, Shyu MK, Lee CN, Lue HC, Hsieh FC. Intrauterine diagnosis of heterotaxy syndrome. Am Heart J 2002; 143: Huggon IC, Cook AC, Smeeton NC, Magee AG, Sharland GK. Atrioventricular septal defects diagnosed in fetal life: associated cardiac and extra-cardiac abnormalities and outcome. JAm Coll Cardiol 2000; 36: Pasquini L, Fichera A, Tan T, Ho SY, Gardiner H. Left superior caval vein: a powerful indicator of fetal coarctation. Heart 2005; 91: Byerregaard P, Laursen HB. Persistent left superior vena cava. Acta Paediatr Scand 1980; 69: Peoples WM, Moller JH, Edwards JE. Polysplenia: a review of 146 cases. Pediatr Cardiol 1983; 4: Lenox CC, Zuberbuhler JR, Park SC, Neches WH, Mathews RA, Fricker FJ, Bahnson HT, Siewers RD. Absent right superior vena cava with persistent left superior vena cava: implications and management. Am J Cardiol 1980; 45: Phoon CK, Neill CA. Asplenia syndrome: insight into embryology through an analysis of cardiac and extracardiac anomalies. Am J Cardiol 1994; 73: Galindo A, Comas C, Martinez JM, Gutierrez-Larraya F, Carrera JM, Puerto B, Borrell A, Mortera C, De la Fuente P. Congenital heart defects in chromosomally normal fetuses with increased nuchal translucency at weeks of gestation. J Matern Fetal Neonat Med 2003; 13: Makrydimas G, Sotiriadis A, Ioannidis JP. Screening performance of first-trimester nuchal translucency for major cardiac defects: a meta-analysis. Am J Obstet Gynecol 2003; 189: Brown DL, Emerson DS, Shulman LP, Doubilet PM, Felker RE, Van Praagh S. Predicting aneuploidy in fetuses with cardiac anomalies: significance of visceral situs and noncardiac anomalies. J Ultrasound Med 1993; 12: