Effect of detailed fetal echocardiography as part of routine prenatal ultrasonographic screening on detection of congenital heart disease

THE LANCET

Effect of detailed fetal echocardiography as part of routine prenatal ultrasonographic screening on detection of congenital heart disease Ingrid Stümpflen, Andreas Stümpflen, Maria Wimmer, Gerhard Bernaschek

Summary Background Cardiac abnormalities are frequently not detected by routine ultrasound screening ex aminations. Although detailed fetal echocardiography is more sensitive in detection of congenital heart disease, it is used only for high-risk cases. The main aim of this study was to assess the prenatal detection of congenital heart disease by detailed fetal echocardiography in an unselected, consecutive group of pregnant women. Methods Between Jan 1, 1993, and Sept 30, 1994, all w omen w ho attended our antenatal-care unit w ere routinely offered a detailed fetal echocardiography ex amination at 18–28 weeks’ gestation. 3085 consecutive women were screened: 2181 were screening cases with no known risk factor for congenital heart disease; 540 had maternal risk factors for congenital heart disease, such as a family history or coex isting maternal disease; 364 had sonographically detected abnormalities. The ex amination included the four-chamber view, outflow-tract scan, and colour-flow mapping; doppler and M-mode investigations were also done when appropriate. Findings 46 cases of congenital heart disease were detected prenatally by echocardiography—15 in the group with no risk factors, three in the group with maternal risk factors, and 28 in the group with sonographic abnormalities. Postnatal assessments found six further cases of congenital heart disease that had not been detected prenatally, but these were all minor cases. There w ere no false-positive diagnoses ( sensitivity 85·5%, specificity 100%). The incidence of congenital heart disease in screening cases with no risk factors and in those with maternal risk factors was low (6·9% per 1000, 5·6 per 1000) and similar to the ex pected overall incidence of 8·0 per 1000 livebirths in the general population. In the group with sonographic abnormalities congenital heart disease was found significantly more often (79·9 per 1000). Interpretation Inclusion of detailed fetal echocardiography as a screening ex amination has a substantial effect on detection of congenital heart disease since a major proportion of prenatally detectable cases occur in a low-risk population.

Lancet 1996; 348: 854–57 See Commentary page 836

Departments of Prenatal Diagnosis and Therapy (I Stümpflen MT, Prof G Bernaschek MD), Internal Medicine II (A Stümpflen MD), and Paediatric Cardiology (Prof M Wimmer MD), University of Vienna Medical School, A-1097 Vienna, Austria Correspondence to: Prof Gerhard Bernaschek

854

Introduction Congenital heart disease, with a prevalence of 8·0 per 1000 livebirths is the commonest of the severe congenital abnormalities.1,2 Specific factors that increase a mother’s risk of carrying a fetus with congenital heart disease have been identified: family history of congenital heart disease; maternal age older than 35 years; coexisting maternal disease (eg, diabetes mellitus, collagen vascular disease, or phenylketonuria); exposure to teratogen; and rubella infection. In such cases, detailed fetal echocardiography is commonly done as part of the sonographic screening at between 18 and 22 (or 24) weeks’ gestation.3,4 In addition, the risk of a fetal cardiac abnormality is substantially higher when routine prenatal ultrasound screening shows an abnormal four-chamber view or certain extracardiac abnormalities (organ malformations, intrauterine growth retardation, amniotic fluid excess or deficiency, fetal arrhythmias).1,5,6 A detailed fetal echocardiographic examination is, therefore, also indicated in such cases.6 Routine diagnosis or exclusion of a congenital cardiac malformation by transabdominal sonography is possible as early as 18 weeks’ gestation. 7,8 However, fetal echocardiography is a time-consuming procedure that requires many two-dimensional cross-sectional views of the heart3,4,9 and additional doppler10 and colour-flow investigations;11,12,13,14 skilled investigators with special experience in fetal cardiology are also needed. Thus, detailed fetal echocardiography is not part of routine prenatal screening programmes but is reserved for cases at high risk of congenital heart disease. Consequently, cardiac abnormalities are among the major malformations that are most frequently missed in prenatal ultrasound examinations,15 which is a cause for concern because undetected congenital heart disease increases the risk of early neonatal mortality. Therefore, we carried out a prospective study of routine detailed fetal echocardiography in unselected, consecutive pregnant women at our antenatal-care centre. The main aim was to assess the prenatal detection rate of congenital heart disease by detailed fetal echocardiography as a screening examination. Furthermore, indication-dependent detection rates for three different groups were assessed and compared: consecutive women without recalled risk factors or sonographically detected abnormalities (screening examinations); women with recalled risk factors (screening indicated for material risk factors); and women whose fetuses showed abnormalities on sonography during pregnancy (screening indicated for fetal abnormalities).

Methods All pregnant women who attended our antenatal unit between Jan 1, 1993, and Sept 30, 1994, were routinely offered a detailed

Vol 348 • September 28, 1996

THE LANCET

fetal echocardiographic examination once during pregnancy to detect cases of fetal congenital heart disease. All women were eligible including those with no known risk factors for congenital heart disease. The detailed fetal echocardiography examination3,4 was done between 18 and 28 weeks’ gestation; when women first attended later than 28 weeks, the examination was done at the first visit. All examinations were done by one investigator (IS) who used a two-dimensional ultrasound system with a 3·5 MHz transducer and doppler and colour-flow option (Toshiba 270-SSA, Toshiba Medical Systems Inc, Japan). The procedure3,4 included the four-chamber view, outflow-tract scans, and colour-flow mapping; additional doppler and M-mode investigations were done when appropriate. In cases of unfavourable fetal position or maternal obesity the examination was repeated the same day or up to 1 week later. The fetal heart could be viewed adequately in all cases. The mean duration of the heart examination for normal cases with good visibility was about 4 min. Of the 3142 consecutive pregnancies included in the study, 57 (1·8%) had normal fetal heart scans but were lost for followup; these cases were excluded from the subsequent analyses. Of the remaining 3085 pregnancies, 2181 (70·7%) were consecutive cases screened without any known risk factors for congenital heart disease (screening examinations). 540 (17·5%) cases were classified as indicated examinations, because of a family history of congenital heart disease or coexisting maternal disease. 364 (11·8%) cases were classified as indicated examinations because of sonographically detected fetal abnormalities. When fetal echocardiography revealed congenital heart disease more detailed diagnostic investigations were done after delivery; necropsy was done after all terminations of pregnancy. All newborn infants with normal fetal heart scans were examined by a neonatologist at the Department of Paediatric Cardiology if congenital heart disease was suspected; they remain under paediatric follow-up until the age of 4 years (according to the Austrian federal screening programme). The oldest children in the study group are now aged 3 years.

Results Congenital heart disease was detected by detailed echocardiography in 46 of the 3085 fetuses (14·9% per 1000). Of the 46, five had isolated ventricular septal defect; six ventricular and atrial septal defects; seven tetralogy of Fallot; five an atrioventricular septal defect; four hypoplastic left-heart syndrome; three double-outlet right ventricles; two transposition of the great arteries; two pulmonary or aortic stenosis; two truncus arteriosus; one an intracardiac tumour; and nine complex or combined cardiac malformations. The table shows the distribution of cases of congenital heart disease diagnosed in each group. There were 15 cardiac malformations (33% of all diagnosed cases) in the group with no risk factors. Only three of the 46 heart defects were found among the fetuses screened because of maternal risk factors. There were 28 cardiac abnormalities (61%) in the group examined because of fetal abnormalities. This high rate is not unexpected; six of the latter 28 cases had already been referred to our unit for detailed diagnosis because of an abnormal four-chamber view. The overall incidence of congenital heart disease was 14·9% per 1000 compared with a mean of 8·0 per 1000 in the general population. The incidence was below this mean in the groups with no risk factors or maternal risk factors, but was much higher in the group with fetal abnormalities (table). The mean time of diagnosis of congenital heart disease was 26·9 (SD 5·8) weeks’ gestation for the total study population. A diagnosis of congenital heart disease was made earlier in the groups with no risk factors (23·1 [4·2]

Vol 348 • September 28, 1996

Group

Number of fetuses investigated (n=3085)

Number of cases diagnosed (n=46)

Screening examinations (no risk factors)

2181 (70·7%)

15

6·9

Screening examinations (maternal risk factors) Previous child with congenital heart disease Maternal age older than 35 years Diabetes mellitus Teratogen exposure Infection

540 (17·5%)

3

5·6

Indicated examinations (fetal abnormalities) Abnormal four-chamber view Arrhythmia Extracardiac organ abnormality Amniotic fluid abnormality IUGR IUGR and amniotic fluid abnormality

122

..

257 78 49 34

1 2 .. ..

364 (11·8%)

28

25 26 167 70 62 14

Incidence (cases per 1000)

76·9

6 3 9 1 6 3

IUGR=intrauterine growth retardation.

Table: Cases of congenital heart disease by group

weeks) or maternal risk factors (25·3 [5·5] weeks) than in the group with fetal abnormalities detected during routine ultrasound screening, which led to referral for detailed fetal echocardiography (29·1 [5·6]). Fetal karyotyping was done prenatally in 45 of the 46 fetuses with cardiac defects by cordocentesis or chorion biopsy; in one case karyotyping was done postnatally in accordance with the parents’ wishes. In 17 (37%) a concomitant chromosomal abnormality was found— trisomy 18 in seven cases, trisomy 21 in four, trisomy 13 in three, 45X0 (Turner’s syndrome) in one, 46XY, del(4p) (Wolf’s syndrome) in one, and an unbalanced translocation 46XX, der(7)t(7;15) in one. In 13 of these 17 cases fetal heart disease was present in conjunction with extracardiac abnormalities (fetal abnormalities group). The other four cases of cardiac malformations with abnormal karotypes were found among the group with no risk factors—two cases of trisomy 21 and two of trisomy 18. In 22 of the 46 cases with prenatally detected cardiac malformations the pregnancies were terminated. The indication for termination was concomitant chromosomal abnormality in 13: nine were judged non-viable (trisomy 18 in six cases, trisomy 13 in three), and the parents requested termination before viability in the other cases (trisomy 21 in three cases, unbalanced translocation in one case). Indications for termination of pregnancy before 24 weeks’ gestation in the remaining nine cases were: cardiac abnormality, not judged viable (hypoplastic left-heart syndrome); severe cardiac abnormality and parents’ request for termination after discussions between neonatologist, paediatric cardiologist, and psychologist (three cases); and cardiac abnormality combined with severe extracardiac malformations (five cases). Postnatal assessment detected six cases of congenital heart disease not detected by echocardiography (false negative rate 0·19%); however, all were minor cardiac abnormalities (three atrioseptal defect, three ventricularseptal defect—perimembranous in one, muscular in one, supra-apical in one). The false-negative rate was 0·09% (2/2181) among the group with no risk factors and 1·09% (4/364) in the group examined because of fetal abnormalities. All six babies did well after birth and none needed surgical intervention. No major cardiac abnormalities were missed by echocardiography and 855

THE LANCET

diagnosed postnatally. Thus, the sensitivity of detailed fetal echocardiography in the detection of fetal heart disease in the series of 3085 was 88·5% (90% for the screening cases [no risk factors and maternal risk factors] and 87·5% for the group with fetal abnormalities). Since there was no false-positive cases (prenatal diagnosis of a cardiac abnormality not confirmed postnatally or at necropsy), specificity and positive predictive value in the series were 100%.

Discussion Prenatal diagnosis of major malformations by ultrasonography may significantly lower perinatal mortality.16 Previous studies have shown the importance of a routine fetal screening for congenital heart disease.15,17,18 Transabdominal fetal echocardiography was introduced for high-risk cases.19,20,21 Allan et al22 found that prenatal methods detected more severe congenital heart disease than postnatal examination. The introduction of transvaginal sonography allowed assessment of the fetal heart earlier in gestation with good results,23,24 and prompted debate about the possibility of a general screening before the 16th gestational week.25 All series reported have shown definite limitations of early transvaginal fetal echocardiography, therefore, a complementary transabdominal examination at about week 20 is still mandatory.21,26 The introduction of doppler and colour-flow studies completed detailed morphological and functional diagnostic capabilities27 and improved prognostic possibilities. Detailed fetal echocardiography is a costly procedure and is not offered as part of routine obstetric screening. The procedure is used only in cases with a suspected higher incidence of CHD because of recalled risk factors or associated abnormalities detected by routine prenatal ultrasonography screening. Of the 52 cases of congenital heart disease in our large series, 46 were diagnosed prenatally (sensitivity 88·5%, specificity 100%). Even in the group with no risk factors, the detection rate was satisfactory: 15 (88%) of 17 cases were correctly diagnosed. The estimated mean incidence of congenital heart disease in the general population is about 8·0 per 1000 livebirths.1,2 In our consecutive series, the prenatally detected incidence of congenital heart disease was 14·9 per 1000 livebirths—46 cardiac abnormalities in 3085 fetal heart scans. This incidence is higher than expected because of the inclusion of a preselected highrisk group of patients with sonographic characteristics linked with congenital heart disease. By contrast, the incidence rate of the group with maternal risk factors was lower. This incidence is even lower than the expected 8·0 per 1000 livebirths in the general population. Although the number of fetuses in this group (540) was small and may not necessarily be representative, the risk of congenital heart disease in indicated screening examinations might have been overestimated in the past. Further studies are needed to investigate this issue. We did not expect the incidence rate to be higher in the group with no risk factors than in the group with maternal risk factors. Retrospective review of videotape material showed that only seven of the 15 cases were detectable in the four-chamber view alone. Thus, without detailed fetal echocardiography the remaining eight cases would not 856

have been detected prenatally, or would have been diagnosed only later in gestation after the occurence of secondary signs, such as growth retardation or cardiac failure. We believe that the use of the standard fourchamber view alone in routine fetal screening is insufficient.28,29 Indeed, Ott30 showed that the additional visualisation of only left-ventricular outflow-tract views in low-risk fetuses led to poor diagnostic accuracy.30 Other studies have shown that the inclusion of the great vessels in routine ultrasound screening increases the rate of detection of critical congenital heart defects.31,32 Our data show that a high proportion of prenatally detectable cases of congenital heart disease (15 of 46 cases) occur in a low-risk population without any risk factors or extracardiac abnormalities; such cases can be diagnosed only by full-scale detailed fetal echocrdiography as part of the routine prenatal screening investigation offered to all pregnant women. In this study, the importance of fetal echocardiography was underlined by the association between cardiac defects and otherwise undiagnosable chromosomal aberrations. In four cases the diagnosis of an isolated cardiac malformation without any extracardiac abnormality by subsequent chromosomal anlaysis led to the detection of an abnormal karyotype. These cases would not have been detected prenatally without detailed fetal heart examination. We believe that the most important reason for the implementation of a routinely offered detailed prenatal heart screening is early diagnosis. Such screening for critical cardiac defects allows earlier diagnosis because secondary symptoms, such as retardation or amniotic fluid abnormalities, which occur only later in gestation, are not needed as extracardiac indicators for the investigation of the heart itself. In addition, since the detection of congenital heart disease by routine detailed fetal echocardiography screening is possible before 24 weeks’ gestation, the parents can be informed about diagnosis, severity, and prognosis before viability is reached. Thus, the parents are able to make informed decisions about the further course of pregnancy. This approach can also improve the chances of survival of a baby with a critical congenital defect by ensuring that the necessary prenatal and postnatal care is provided.

References 1

2

3 4 5

6

7

8

Hoffman JIE, Christianson R. Congenital heart disease in a cohort of 19 502 births with long-term follow-up. Am J Cardiol 1978; 42: 641–47. Mitchell SC, Korones SB, Berendes HW. Congenital heart disease in 56 109 births: incidence and natural history. Circulation 1971; 43: 323–32. Allan LD. Manual of fetal echocardiography. Lancaster: MTP Press, 1986. Reed KL, Anderson CF, Shenker L. Fetal echocardiography—an atlas. New York: Allan R Liss, 1988. Greenwood RD, Rosenthal A, Parisi L, Fyler DC, Nadas AS. Extracardiac abnormalities in infants with congenital heart disease. Pediatrics 1975; 55: 485–92. Ferencz C, Rubin JD, McCarter RJ, et al. Cardiac and non-cardiac malformations: observations in a population-based study. Teratology 1987; 35: 367–78. Sahn DJ, Lange LW, Allen HD, et al. Quantitative real-time crosssectional echocardiography in the developing normal human fetus and newborn. Circulation 1980; 62: 588–97. Kleinman CS, Hobbins JC, Jaffe CC, Lynch DC, Talner NS. Echocardiographic studies of the human fetus: prenatal diagnosis of congenital heart disease and cardiac dysrhythmias. Pediatrics 1980; 65: 1059–67.

Vol 348 • September 28, 1996

THE LANCET 9

10

11

12

13

14

15 16

17

18 19

20

21

Allan LD, Tynan MJ, Campbell S, Wilkinson JL, Anderson RH. Echocardiographic and anatomical correlates in the fetus. Br Heart J 1980; 44: 444–51. Reed KL, Mejboom EJ, Sahn DJ, Scagnelli SA, Valdes-Cruz LM, Shenker L. Cardiac Doppler flow velocities in human fetuses. Circulation 1986; 73: 41–46. Copel JA, Morotti R, Hobbins JC, Kleinman CS. The antenatal diagnosis of congenital heart disease using fetal echocardiography: is color flow mapping necessary? Obstet Gynecol 1991; 78: 1–8. Devore GR. Color Doppler examination of the outflow tracts of the fetal heart: a technique for identification of cardiovascular malformations. Ultrasound Obstet Gynecol 1994; 6: 463–71. Gembruch U, Hansmann M, Redel DA, Bald R. Fetal twodimensional Doppler echocardiography (color flow mapping) and its place in prenatal diagnosis. Prenat Diagn 1989; 9: 535–47. Gembruch U, Chatterjee MS, Bald R, Redel DA, Hansmann M. Color Doppler flow mapping of fetal heart. J Perinat Med 1991; 19: 27–32. D’Alton ME, DeCherney AH. Prenatal diagnosis. N Engl J Med 1993; 328: 114–20. Saari-Kemppainen A, Karjalainen O, Ylostalo P, Heinonen OP. Ultrasound screening and perinatal mortality: controlled trial of systematic one-stage screening in pregnancy. Lancet 1990; 336: 387–91. Copel JA, Pilu G, Green J, Hobbins JC, Kleinman CS. Fetal echocardiographic screening for congenital heart disease: the importance of the four-chamber view. Am J Obstet Gynecol 1987; 157: 648–55. Allan LD, Crawford DC, Chita SK, Tynen MJ. Prenatal screening for congenital heart disease. BMJ 1986; 292: 1717–19. Sandor GG, Farquarson D, Wittmann B, Chow TC, Lau AE. Fetal echocardiography: results in high-risk patients. Obstet Gynecol 1986; 67: 358–64. Allan LD, Crawford DC, Chita SK, Anderson RH, Tynan MJ. Famial recurrence of congenital heart disease in a prospective series of mothers referred for fetal echocardiography. Am J Cardiol 1986; 58: 334–37. Allan LD. Fetal Cardiology. Ultrasound Obstet Gynecol 1994; 4: 441–44.

Vol 348 • September 28, 1996

22 Allan LD, Crawford DC, Anderson RH, Tynan M. Spectrum of congenital heart disease detected echocardiographically in prenatal life. Br Heart J 1985; 54: 523–26. 23 Rottem S, Bronshtein M, Thaler I, Brandes JM. First trimester transvaginal sonographic diagnosis of fetal anomalies. Lancet 1989; i: 444–45. 24 Gembruch U, Knöpfle G, Chatterjee M, Bald R, Hansmann M. First-trimester diagnosis of fetal congenital heart disease by transvaginal two-dimensional and Doppler echocardiography. Obstet Gynecol 1990; 75: 496–98. 25 Achiron R, Weissman A, Rotstein Z, Lipitz S, Mashiach S, Hegesh J. Transvaginal echocardiographic examination of the fetal heart between 13 and 15 weeks’ gestation in a low-risk population. J Ultrasound Med 1994; 13: 783–89. 26 Todros T, Presbitero P, Gaglioti P. Errors and pitfalls in prenatal diagnosis of congenital heart disease. J Maternal Fetal Invest 1992; 2: 157–61. 27 Devore GR, Horenstein J, Siassi B, Platt LD. Fetal echocardiography VII: Doppler color flow mapping: a new technique for the diagnosis of congenital heart disease. Am J Obstet Gynecol 1987; 156: 1054–64. 28 Bromley B, Estroff JA, Sanders SP, et al. Fetal echocardiography: accuracy and limitations in a population at high and low risk for heart defects. Am J Obstet Gynecol 1992; 166: 1473–81. 29 Wigton TR, Sabbagha RE, Tamura RK, Cohen L, Minogue JP, Strasburger JF. Sonographic diagnosis of congenital heart disease: comparison between the four-chamber view and multiple cardiac views. Obstet Gynecol 1993; 82: 219–24. 30 Ott WJ. The accuracy of antenatal fetal echocardiography screening in high- and low-risk patients. Am J Obstet Gynecol 1995; 172: 1741–47. 31 Achiron R, Glaser J, Gelernter I, Hegesh J, Yagel S. Extended fetal echocardiographic examination for detecting cardiac malformations in low-risk populations. BMJ 1992; 304: 671–74. 32 Tegnander E, Eik-Nes SH, Johansen OJ, Linker DT. Prenatal detection of heart defects at the routine fetal examination at 18 weeks in a non-selected population. Ultrasound Obstet Gynecol 1995; 5: 372–80.

857