echocardiographic examinationfor detecting cardiac ... · of anacsthesia using midazolam ini cardiac surgical patictits. Br 7 Anuaslih 1982;54:1053-7. 31 Merin RG. Is anesthesia beneficial

Chisholin (GD. Benign prostatic hyperplasia: the best treatment. 1BMI71989;299:2 15-6.

2 Neal DE. Irostatectom\-an open or shut case. Br7 Urol 1990;66:449-54.3 Wrennburg J E, Roos N, Sola L, Schori A, Jalfe R. Use of claims data s\stcms to

cvaluatc he:alth care ouitcomes. Mortality and re-operation aftcr prostatec-tormr. J.1IA 1987;257:933-6.

4 Holtgrcwe HL, Valk W`L. Factiirs influcincing the mortality and morbidityof transurethral prostatectomy: a study ot' 2015 cases. 7 Urol 1962;87:450-9.

5 Melchior J, V'alk VI'_, Foret JD, Mlebntst W'K. Transurethral prostatectomrv;computcrised analysis of'2223 consecutive cases. J Urol 1974;112:634-42.

6 Blandv JP. Benign prostatic enlargemeiti. BMij7 1971;i:31-5.7 Sach R, Marshall VR. Prostatectomv: its safety in ani Australian teaching

hospital. BM317 1977;64:210-4.8 Allen PIM, Zagcr M, Goldman M. Electivc repair of groin hcrnias in the

elderly. Br.7 St*rg 1987;74:987.9 Tingwald GR, Cooperman M. Inguinal and femoral hernia repair in geriatric

patients. Surgq Gynecol Obstet 1982;154:704-6.10 Officc ot l'optulation Censuses and Surveys. Hospital inpatient enqat-ires 1983.

London: HMSO, 1985.11 Roos N, Ramsav EW. A population based study of prostatectomy: imutcomes

associated with differing surgical approaches. Urn! 1987;137:1 184-8.12 Roos NP', Wentnburg JE, Malenka DJ, Fisher ES, McPherson K, Andersen

TF, et at. Mortality and re-operation after open and transurethral resectionof the prostate for benign prostatic hyperplasia. N Englj7 Med 1989;320:1120-4.

13 *ru or not TU [Editorial]. Lancet 1989;i: 1361-2.14 Coppinger SWV, Hudd C. Risk factor for mvocardial infarctioti in trans-

urethral resection of the prostate? Lancet 1989;ii:859.15 Miebust WK, Brady TW, Valk WL. Observations on cardiac output, blood

volume, central senous pressure and electrolyte changes in patientsundergoing transurethral prostatectomy. J U-rol 1970;103:632-6.

16 De Angelis L, Chang P, Kaplan JH, Kudish H, Sacks S, Wender R,et al. Hemodynamic changes during prostatectomy in cardiac patients.Crtu Care Med 1982;10:38-40.

17 Evans JWH, Singer M, Chapple CR, Macartnev N, Coppinger SWV, MilroyEJG. Haemodynatmic evidence for per-operative cardiac stress duringtransurethral prostatectomv: a preliminary communication. Br J Urol1991;67:376-80.

18 Singer MN, Clarke I, Bennett ED. Continuous hemodynamic monitoring byesophageal lDoppler. Crit Care Med 1989;17:447-52.

19 Singer M. Ciontinuous haemodynamic moniitoring by oesophageal Doppler(MD thcsis). London: University of London, 1990.

20 Kunin SA, Limbert DJ. Central venous pressure monitoring during trans-turethral prostatectomy. Utrol 1969;102:469-72.

21 Cork RC, Vaughan RW, Humphrev L.S. Plrecision and accuracy of intra-operatiVe temperature monitoring. .Anesth Analg 1983;62:211-4.

22 Imrie MM, Hall GMi. Body temperature and anaesthesia. Br J Anaesth1990;64:346-54.

23 Ghanem AN, Ward JP. Osmotic and metabolic sequelae of volumctricoverload in relation to the TUR syndrome. Brj L'rol 1990;66:71-8.

24 Armitage P'. Siatistical mcthids in medical research. Oxford: Blackwell Scientific1971:185.

25 Matthews JNS, Altman DG, Campbell NIJ, Royston P. AnalNsis otf serialmcasturcmcnt in medical rescarch. HBl] 1990300:230-5.

26 Perkins JB, Miller HC. Blood loss diuring transurethral prostatctotny. LUrol1969;101:93-7.

27 Flechnlcr SM, Williams RD. Continuotis flow anid cotiventiontal resectoscopemcthods in tranisuretliral prostatectomy; comparative study. ] UIrl 1982;127:257-9.

28 Heathcotc PS, D)ver PMNI. 'Thc cffcit of warm irrigationi oit blood loss dtiringtranisurcthral prostatectomr under spitial ataaestliesia. Brj7 I rol 1986;58:669-71.

29 Mebtust WK, Holtgrew HL, Cocket ATK, I'etcrs l'C. 'I'ransturethral prosta-tcctom\: immediate and postoperative complications. A cooperative studyof 13 participating institutionis evaluLating 3,855 patictis. 7 I'rol 1989;141:243-7.

30 Scliultc-Scltasse IT, Hess \', Tartiow J. Hactniodynarnic responisc to itidtlctionof anacsthesia using midazolam ini cardiac surgical patictits. Br 7 Anuaslih1982;54: 1053-7.

31 Merin RG. Is anesthesia beneficial for the ischemic heart? Anesthesiology1980;53:439-40.

32 \errilli RA, lThlman RC, Vick NF, Hunisicker WC Jr. The h potensivc effectof a prostatic extract.J lU'rol 1962;87:184-6.

33 Braunwald E. Control of mvocardial oxygen consumption. Physiologic andclinical considerations. Am]j Cirdtiol 197 1;27:416-32.

34 Berne RM, Rubio R. Coronary circulation. In: Bernc RMI, Sperckalis N,Gciger SR. eds. Handbook (if phlvsologv. Section 2: the cardiovtsiculr sYstenI.Baltimore: Williams and Wilkinis, 1979:873-952.

35 Rabke HB, Jenicek JA, Khouri E Jr. Hypothermia associatcd with trails-uircthral rescction of the prostatc. J trol 1962;87:447-9.

36 Allen TD. Body tcmperature changes dtiring prostatic resection as related tothe temperaturc otf' the irrigation solition.,7 Urol 1973;110:433-5.

37 Kulatilake AL, Roberts PN, Evans D)F, WYright J. The use of cooled irrigatingsolution during transurethral resection of'the prostate. Br] Urol 1981;53:261-2.

38 Rawstroti RE, Walton JK. Bod\s temperature changes ditritig tranisurethralprostatettomy. Anlnaesth mitens Care, 198 1;9:43-6.

39 Carpenter AA. Hypothermia duritig tratasurcthral resection of' prostatc.U'rologv 1984;23: 122-4.

40 Carli P. Mletabolic disturbanccs of hypotthermia. In: Ballihure'sri lilicalanaesthesiologyv 1989;3:405-21.

41 Rculer JB. Hypothermia: pathophysiology, clinical settings and managcment.Ann InternMeled 1978;89:519-27.

42 Keatinge W'R, Coleshaw SRK, Cottcr F, Mattock M, Murphv MN, Chelliah R.Increase in platelct and red cell couiits, blood viscosity, and arterial pressurcduring mild surface cooling: factors in mortality from coronary and cerebralthrombosis in winter. BMA7 1984;289:1405-8.

43 Williams G, Jager R, McLoughlin J, El Din A, Machan L, Gill K, et al. Use ofstents for treating obstruction of urinary outflow in patiecits unfit forsurgery. BM.V] 1989;298:1429.

44 Chapple CR, Milrov EJG, Rickards D). A permanently implanted uirethralstent f'or prostate obstruction in the unfit patient; preliminary report.Br]7Irol 1990;66:58-65.

(Accepted 20 D)eember 1991,!

Extended fetal echocardiographic examination for detecting cardiacmalformations in low risk pregnancies

Reuwen Achiron, Joram Glaser, Ilana Gelernter, Julius Hegesh, Simcha Yagel

AbstractObjective-To improve the rate of prenatal

detection of cardiac malformations in a low riskpopulation.Design-Comparison of extended fetal echo-

cardiography with the standard four chamber view indetecting abnormalities. Extended echocardio-graphy comprised the four chamber view and visual-isation of the left ventricular outflow tract, the rightventricular outflow tract, and the main pulmonaryartery and its branches. In cases with abnormalresults complete echocardiographic studies wereperformed by a paediatric cardiologist using Mmode, Doppler, and colourflow mapping techniques.

Setting-Obstetric ultrasonographic unit atShaare-Zedek Medical Centre, Jerusalem.Subjects-5400 fetuses in low risk pregnancies

between 18 and 24 weeks' gestation (mean 21 weeks);53 were lost to follow up.Main outcome measures-Detection of abnor-

mality before and after birth.Results-During the study 23 infants (0.4%) were

born with cardiac abnormalities, 21 of whom hadmajor structural and functional heart disease. 18fetuses had heart disease diagnosed prenatally, 11 bythe four chamber view alone (sensitivity 48%) anda further seven by extended echocardiography(sensitivity 78%). Five fetal cardiac defects were

missed prenatally (false negative rate 22%). Theseincluded coarctation of aorta, persistent truncusarteriosus, tetralogy of Fallot, ventricular septaldefect, and pulmonic stenosis. Only one falsepositive diagnosis (coarctation of aorta) was made(specificity 99-9%, false positive rate 0-1%). Theabnormality was correctly identified in 17 out of 18cases.Conclusions-The extended fetal heart exami-

nation detected 86% (18/21) of major abnormalitiesin a low risk population. The examination should beincorporated into routine prenatal ultrasonographicinvestigations.

IntroductionThe incidence of congenital heart disease is 8/1000

live births. Half of the cases are considered minor andare easily corrected by surgery. The remainder aremore serious and account for about 50% of deaths fromlethal malformations in childhood.' 2Congenital heartdiseases are 6 5 times more common than chromosomalabnormalities and four times more common thanneural tube defects.' Unlike chromosomal and neuraltube defects, for which there has been extensiveprenatal screening, congenital heart diseases are oftennot identified until the infant is born. The importanceof early diagnosis of congenital heart disease is further

Department of Obstetricsand Gynaecology, ChaimSheba Medical Centre,52621 Tel Hashomer, IsraelReuwen Achiron, MD,director ofdiagnosticultrasound unitJulius Hegesh, MD, directorofpaediatric cardiology unit

Shaare Zedek MedicalCentre, JerusalemJoram Glaser, MD, director ofpaediatric cardiology unit

Statistical Laboratory,School of MathematicalSciences, Tel AvivUniversity, Tel AvivIlana Gelernter, MBA

Department of Obstetricsand Gynaecology, MountScopus Hospital,JerusalemSimcha Yagel, MD, director ofdiagnostic ultrasound unit

Correspondence to: DrAchiron.

B&V 1992;304:671-4

BMJ VOLUME 304 14 MARCH 1992 671

on 2 July 2019 by guest. Protected by copyright.

http://ww

w.bm

j.com/

BM

J: first published as 10.1136/bmj.304.6828.671 on 14 M

arch 1992. Dow

nloaded from

emphasised by the fact that 20% to 32% of perinataldeaths are due to congenital heart disease.4 Althoughrecent technology has allowed fetal heart evaluationand detection of anomalies from the 17-18th week ofgestation or even earlier with transvaginal ultra-sonography, the prenatal diagnosis of congenital heartdefects is still a challenge for the modern obstetricultrasonographer.The usefulness of fetal echocardiography as a clinical

technique for the prenatal diagnosis of congenital heartdisease was shown in the early 1980s.'6 Allan et aldetected 60% of severe cardiac anomalies using thefour chamber view for screening congenital heartdiseases.7 To increase the detection rate of fetal echo-cardiography, women at high risk of having childrenwith congenital heart disease have been selected fordetailed fetal echocardiographic examination inspecialised referral centres.8 However, since mostneonates with congenital heart disease are born towomen without any previous known risk factors this

prenatal selection seems ineffective. We conducted astudy to evaluate whether extended echocardiographyin the low risk pregnant population can increase thedetection rate of congenital heart disease.

Subjects and methodsScreening for congenital heart disease is offered to all

pregnant women routinely attending the obstetricultrasonographic unit at Shaare-Zedek Medical Centre,Jerusalem. Women with the following risk factors wereexcluded from our study: history of congenital heartdisease, maternal diabetes or connective tissue diseases,exposure to alcohol or lithium, phenylketonuria, fetussmall for gestational age, and viral infection during thefirst trimester. Only women below 35 years old, withwell documented medical follow up and who had bornelive infants were included in the study.A detailed biometric and structural evaluation of all

fetuses was performed with sector scanners of 5 and

Fetal echocardiography showing (a) four chamber view; (b) long axis view through inlet and outlet ofright atrium showing inferior and superiorvena cava and (c) pulmonary vein entering left atrium-arrow indicates opening offoramen ovale; (d) long axis view showing left ventricular outflow tract obtained by slight cranial angulation of transducer from four chamber view and slight clockwise rotation of transducer on maternalabdomen in case ofposterior fetal spine; (e) short axis view showing right ventricular outflow tract; (f) main pulmonary artery and its bifurcationobtained by further cranial angulation and counter clockwise rotation of transducer. LA = left atrium, RA = right atrium, LV = left ventricle,RA = right ventricle, IVC = inferior vena cava, SVC = superior vena cava, PV = pulmonary vein, FO = foramen ovale, RPA = rightpulmonary artery, AAO = root of ascending aorta, MV = mitral valve, TV = tricuspid valve, AO = aorta in transverse section, PV =pulmonary valve, MPA = main pulmonary artery, andSPA = left pulmonary artery

BMJ VOLUME 304 14 MARCH 1992

:1

672

on 2 July 2019 by guest. Protected by copyright.

http://ww

w.bm

j.com/

BM

J: first published as 10.1136/bmj.304.6828.671 on 14 M

arch 1992. Dow

nloaded from

TABLE I-Rate (per 1000 live births) ofthe types ofcongenital heart disease compared with rate derivedfromthree large studies"-'3

Abnormality No of cases Observed rate Mean (range) calculated rate p Value

TetralogyofFallot 5 0 9 0-6(0 4-0 8) 0 31Hypoplastic right heart 3 0-5 0-2 (0-17-0 24) 0-06Hypoplastic left heart 2 0 3 0-45 (03-0-6) 0 79Coarctation 2 0 3 0-62 (0 45-0-8) 0-41Persistent truncus arteriosus 2 0 3 0-12 (0-08-0-16) 0-08Atrio ventricular septal defect 2 0 3 0-36 (0 32-0-4) 0-86Double outlet in right ventricle 1 0-18 0-12 (0-016-0-24) 0-82Ventricular septal defect 1 0-18 1-8 (1-2-2-4) 0-005Pulmonic stenosis 1 0-18 0-4 (0-2-0-6) 0-43Miscellaneous 4 0-7 1-13 (0-9-1-36) 0-42

TABLE II-Congenital heart diseases detected by four chamber view and by extended echocardiographyexamination

Anomaly No of cases Ultrasonographic findings

Abnormal on four chamber view (n= 1 1):Hypoplastic heart syndrome 5 Single ventricle and atriaAtrioventricular septal defect 2 Abnormal cruxHeart failure 2 Pericardial effusion, cardiomegalyEpstein anomaly 1 Giant right atrium, abnormal tricuspidEctopic Cordis 1 Extrathoracic heartNormal on four chamber view, abnormalon extended examination (n=7):

Tetralogy of Fallot 4 Septal defect in ventricular outlet, overriding aortaDouble outlet in right ventricle 1 Septal defect in ventricular outlet, two outlets, from

right ventricleCoarctation 1 Narrow aortic root, relatively small left ventriclePersistent truncus arteriosus 1 Ventricular septal defect, single outlet vessel

TABLE IlI-Diagnostic accuracy ofabnormal results on four chamber view and extended echocardiography

Positive predictive Negative predictiveSensitivity Specificity value value

Abnormal four chamber view(n= 11) 11/23 (48%) 5323/5324 (99 9%) 11/12 (92%) 5323/5335 (99-8%)95% Confidence interval (38% to 58%) (88% to 96%)

Abnormal extendedexamination (n= 18) 18/23 (78%)* 5323/5324 (99 9%) 18/19 (95%) 5323/5328 (99-9%)95% Confidence interval (71% to 85%) (93% to 97%)

*p=0O0078 Compared with fourchamber view by correlated xI test. 3-5 MHz mounted with pulsed wave Doppler (ESI

2000, Elscint, Haifa, Israel) and 3MHz (ATL MarkIV, Squibb Medical Systems). Colour flow mappingwas performed with a Hewlett-Packard (Ultra-system77020A). During cardiac evaluation the standard fourchamber view was visualised (figure (a)); once it wasobtained simple rotation of the transducer along its axisby about 90° brought the major great veins (b and c)and major great arteries into view, allowing propervisualisation of the left ventricle outlet (d), theshort axis of the right ventricle outlet with the mainpulmonary artery (e and f), and sometimes the aorticarch and ductus.9 The average duration of routineechocardiographic examination was 15 minutes, givinga total time for fetal imaging of about 30 minutes.When a cardiac abnormality was detected furtherevaluation was performed in the paediatric cardiologyunit by using Doppler colour flow and M modetechniques and the scans were recorded on videotapefor later review. Personal data on each woman enteringthe study were recorded on a computerised medicalfile. The progress and the outcome of each pregnancywas followed up. Postmortem examinations andneonatal echocardiography were performed on affectedinfants to confirm the prenatal findings. The detectionrate of the procedure (sensitivity) was the proportion ofthe fetuses with congenital heart disease who had apositive result on screening. This criterion could beascertained only within the limitation of the studydesign-that is, diagnosis was based on neonatalphysical examinations, and infants with asymptomaticheart disease that might develop later were notascertained. Neonatal registration and personalcommunications with the only three paediatric cardio-logist centres in our community were used to determinethe true negative, false negative, and false positivecases. Three fifths of the fetuses studied were delivered

at Shaare-Zedek Medical Center; the remainder weredelivered in three other university hospitals.

Statistical analysis was performed with the x2 orcorrelated X2 test."' The level of significance was set atp<o.0.

ResultsDuring August 1988 to April 1990, 5400 fetuses,

including 50 pairs of twins, were screened at 18-24weeks' gestation. Of the fetuses screened, 53 were lostto follow up and therefore omitted from the study,leaving 5347 fetuses. The average age of the womenwas 24-6 (range 18-45) years. The mean gestational ageat the first routine screening was 21 (18-24) weeks; afour chamber view was obtained in 5240 (98%) of thefetuses; the extended fetal heart examination wascompleted in 4812 (90%). Twenty three fetuses wereborn (0-4%) with congenital cardiac anomalies. Table Ilists the cardiac malformations in the 23 cases. Theobserved rate ofeach malformation has been comparedwith the mean of the rates obtained in three largestudies."'3 There was broad agreement between theobserved rate and the calculated rates for most of theanomalies. However, the rate ofright hypoplastic heartsyndrome was increased in our study, while the rate ofventricular septal defect was lower.

Prenatal echocardiography with the four chamberview detected 11 out of the 23 fetuses with cardiacmalformations (table II). These included five fetuseswith the hypoplastic heart syndrome (three right andtwo left); two with atrioventricular septal defect, twowith congestive heart failure-due to paroxysmal atrialtachycardia in one fetus and twin to twin transfusion inthe second; one with Epstein anomaly; and one withectopic cardia. Extended echocardiographic evaluationof the ventriculoarterial connections detected cardiacanomalies in an additional seven fetuses. Outlet ven-tricular septal defect associated with large overridingaortic root was pathognomonic of tetralogy of Fallot infour of these seven cases; the other three fetuses haddouble outlet in the right ventricle, persistent truncusarteriosus, and coarctation of aorta. Five fetuses withcardiac abnormalities (22%) did not have the abnor-malities detected by echocardiography. Of these five,two had minor abnormalities (ventricular septaldefect and pulmonic stenosis) and three had majorabnormalities (persistent truncus arteriosus, tetralogyof Fallot, and coarctation of aorta). Only in one normalfetus was a false positive diagnosis of coarctation ofaorta made by screening (specificity of 99 9%).The malformation was misdiagnosed in only one

case, in which tetralogy of Fallot was erroneouslyidentified as tricuspid atresia with ventral septal defect.In all other fetuses postnatal examination by a paediatriccardiologist or pathologist confirmed the prenataldiagnosis.

Table III summarises and compares the diagnosticaccuracy of four chamber view echocardiographyand extended examination in terms of sensitivity,specificity, and positive and negative predictive values.Sensitivity of screening increased from 48% to 78%with the extended examination, and the positivepredictive value increased from 92% to 95%. Thespecificity and negative predictive value were notsignificantly different for the four chamber view andextended examination.Chromosomal analysis was performed prenatally in

four cases. Two (9%) abnormal karyotypes weredetected: trisomy 21 and trisomy 18. Table IV givesthe outcomes of the pregnancies. Five pregnancieswere terminated and the mortality of fetuses incompleted pregnancies was 28% (5/18). Overall, 10 of23 fetuses (43%) with cardiac anomalies died. Two ofthe 13 surviving fetuses, had severe genetic syndromes.

BMJ VOLUME 304 14 MARCH 1992 673

on 2 July 2019 by guest. Protected by copyright.

http://ww

w.bm

j.com/

BM

J: first published as 10.1136/bmj.304.6828.671 on 14 M

arch 1992. Dow

nloaded from

TABLE IV-Outcome ofpregnancies with fetuses with congenital heart defects

Termination ofAbnormality No of cases pregnancy Neonatal death Survived

Tetralogy of Fallot 5 1 4*Atrioventricular septal defect 2 1 1 *Right hypoplastic heart syndrome 3 3Left hypoplastic heart syndrome 2 1 1Persistent truncus arteriosus 2 1 1Ectopic cordis 1 1Double outlet in right ventricle 1 IMiscellaneous 7 7

Total 23 5 (22%) 5 (22%) 13 (56%)

* One case had a severe genetic syndrome.

DiscussionAlthough much has been published on the ability of

ultrasonography to detect heart defects prenatally,'3routine screening of the general population and itsfeasibility has been scarcely reported. Our results arenoteworthy because they emphasise the importance ofroutine ultrasonography in detecting congenital heartmalformations and show that most congenital heartdiseases may be detected by an obstetric ultrasono-grapher with an interest in fetal heart imaging. The0-4% incidence of congenital heart defects observed inour study agrees with the reports ofthe Royal Bromptonhospital" and the New England regional infant careprogramme,'2 which cited incidences of 0 3%.

Furthermore the observed rate ofmost of the varioustypes of heart defects did not differ significantly fromthe rate calculated from three large studies (table I),suggesting that all infants with major congenital heartdiseases were detected by our screening programme.The increased rate of right hypoplastic heart syndromein our study may be explained by the fact that the threestudies gave rates for surviving infants not for prenataldiagnosis. If we omit from the analysis the fetus thatdied in utero of right hypoplastic heart syndrome theobserved rate becomes 0-3/1000 live births, which doesnot differ significantly from the calculated rate.The calculated rate of ventricular septal defect wassignificantly higher than the observed rate; this isbecause of the inability of ultrasonography to detectthe smallest of these defects.

Regarding the reproducibility of fetal cardiacimaging at 21 weeks' gestation the four chamber viewcould be imaged in 98% of cases and the extendedexaminations could be completed in 90% of cases; only10% of patients required repeat ultrasonography at 24weeks. Recently DeVore and Platt concluded in anextensive review that the ability to obtain the fourchamber view increased from 9% at 14 weeks to 20% at20 weeks. The authors examined 400 low risk fetusesbetween 15 and 23 weeks of gestation. They were ableto obtain the four chamber view in 90% of the cases.Allan et al, were the first to report a sensitivity of 66%for diagnosing fetal congenital heart disease withthe four chamber view.7 Copel et al reported 92%sensitivity with the four chamber view. '4 However, thedata in their study were drawn from a biased group,based on patients with a high risk for congenital heartdisease who had been referred for fetal echocardio-graphy. Therefore the high sensitivity obtained bythese authors reflects the prevalence of the diseasesrather than the accuracy of the four chamber view.Benacerraf et al examined a mixed group of high andlow risk women for fetal heart defects and foundsimilar results to those in other studies on high riskpregnancies, in which severe forms of heart diseasewere reliably detected."

In our study the four chamber view detected 48% offetal cardiac anomalies; this rate is comparable with thefindings of Allan et al in the general population.7 The

addition of two relatively simple echocardiographicinvestigations at the level of the left and right ven-tricular outflow resulted in a significant increase in thedetection rate, particularly of abnormalities in theventriculoarterial connection. This simple extendedfetal heart evaluation achieved a sensitivity of 78%,with only 22% of congenital anomalies undetectedduring the prenatal life. Of the five fetuses withfalse negative results, two had severe malformations(tetralogy of Fallot and truncus arteriosus). Since thesefetuses were among the first screened in our pro-gramme, these results probably reflect our lack ofexperience. The other three cases had minor defects(ventricular septal defect, pulmonic stenosis, andcoarctation), and we believe this represents the truelimitation of the method. The false positive rate of0 1% encourages us to believe that only a few parentswould be worried unnecessarily as a result of thescreening programme.As most of the undetected anomalies were minor-

that is, small septal defects and valvular lesions-theircontribution to perinatal mortality and morbidity wasnegligible. Detection of the major congenital heartdiseases in utero has a considerable impact on theprenatal counselling and perinatal management.After counselling by a paediatric cardiologist and aperinatologist, four patients had genetic evaluation. Intwo cases, the finding of trisomic fetuses persuaded theparents to terminate the pregnancy. Termination ofpregnancy was requested in five cases (22%), thusavoiding further neonatal mortality and morbidity.The high early neonatal death rate (28%) among theaffected fetuses surviving to term confirms that some ofthe cardiac anomalies detected were associated withpoor chances of survival. Early prenatal diagnosis ofthe type of the cardiac anomaly and cooperation withthe paediatric cardiologist permit accurate informeddecisions regarding obstetrical management-forexample, avoidance of caesarean section for lethalanomaly or the continuation ofpregnancy with deliveryin a tertiary centre.Widespread teaching of the technique will result in

increased prenatal diagnosis of congenital heart diseaseand should reduce perinatal mortality and morbidity.We have shown prenatal screening ofcardiac anomaliesto be accurate, and it therefore should be incorporatedinto standard fetal surveillance.

1 Hoffman JI, Christian R. Congenital heart disease in a cohort of 19 502 birthswith long term follow up. Am3r Cardiol 1978;42:641-7.

2 Keith JD, Rowe RD, Vlad P. Heart disease in infancy and childhood. 3rd ed.New York: Macmillan, 1978:9.

3 Lian ZH, Zack MM, Erickson JD. Paternal age and the occurrence of birthdefects. AmJ7 Hum Genet 1986;39:648-60.

4 Department of Health and Human Resources. Vital statistics of the UnitedStates. Vol II. Mortality. Washington, DC: DHHR, 1985:5.

5 Kleinman CS, Hobbins JC, Jaffe CC, Lynch DC, Talner NS. Echocardio-graphic studies of the human fetus: prenatal diagnosis of congenital heartdisease and cardiac dysrhythmias. Pediatrics 1980;65:1059-67.

6 Allan LD, Tynan MJ, Campbell S, Wilkinson JL, Anderson RH. Echocardio-graphic and anatomical correlates in the fetus. Br HeartJ 1980;44:445-5 1.

7 Allan LD, Crawford DC, Chita SK, Tynan MJ. Prenatal screening forcongenital heart disease. BMJ 1986;292:1717-9.

8 Allan LD, Crawford DC, Anderson RH, Tynan MJ. Echocardiographic andanatomical correlations in fetal congenital heart disease. Br Heart J1984;52:542-8.

9 Gembruch UK Hansman M, Redel DA, Bald R. Fetal two-dimensionalDoppler echocardiography (colour flow mapping) and its place in prenataldiagnosis. Prenatal Diagn 1989;9:535-47.

10 Brownlee KA. Statistical theory and methodology in science and engineering.Malabar Florida: Robert E Krieger 1984:154-7.

11 Scott DJ, Rigby ML, Aliller GAH, Shinebourne EA. The presentation ofsymptomatic heart disease in infancy based on 10 years' experience (1973-82). Implications for the provision of service. BrHeartj 1984;52:248-57.

12 Fyler DC, Buckley LP, Hellenbrand WE, Cohn HE. Report of the NewEngland regional infant cardiac program. Pediatrics 1980;65:375-461.

13 DeVore GR, Platt LD. Prenatal diagnosis of congenital heart disease-anupdate. Ultrasound Quarterly 1988;6:299-30.

14 Copel JA, Gianluigi P, Green J, Hobbins JC, Kleinman CS. Fetal echocardio-graphic screening for congenital heart disease: the importance of the fourchamber view. Am.7 Obstet Gvnecol 1987;157:648-55.

15 Benacerraf BR, Pober BR, Sanders SP. Accuracy of fetal echocardiography.Radiology 1987;165:847-9.

(Accepted 20 December 1991)

674 BMJ VOLUME 304 14 MARCH 1992

on 2 July 2019 by guest. Protected by copyright.

http://ww

w.bm

j.com/

BM

J: first published as 10.1136/bmj.304.6828.671 on 14 M

arch 1992. Dow

nloaded from