jurnal diagnosis radiologi
Embed Size (px)
Archives of Disease in Childhood, 1989, 64, 964-968
Diagnosis of fetal cardiac abnormalitiesL D ALLAN
Guy's Hospital, London
The feasibility of cross sectional imaging of thestructure of the fetal heart was first described in1980.1-3 Since then fetal echocardiography has beencarried out in many pregnancies that are at in-creased risk of congenital heart disease, and morerecently the study of one section of the fetal hearthas been included in routine obstetric ultrasoundexaminations.4 These studies have yielded surpris-ing results about the incidence, range, and naturalhistory of congenital heart disease that may be seenprenatally. Also the pattern of recurrence of con-genital heart disease within family groups does notseem to conform to previously accepted genetictheories.The mothers who are at increased risk of fetal
heart disease include: those with a family history ofcongenital heart disease; those with diabetes; orthose in whom fetal hydrops, arrhythmia, or anextracardiac anomaly has been found. Our mostimportant source of detection of cardiac anomaliesin recent years comes from apparently normalpregnancies in which the obstetrician has noticed anabnormal appearance of the heart on a routine scan.The idea of prenatal cardiac screening came from
a French study first reported in 1985 (Fermont L,De Geeter B, Aubry MC, Kachaner J, Sidi D. Aclose collaboration between obstetricians and car-diologists allows antenatal detection of severe car-diac malformations by 2D echo. Abstract, SecondWorld Congress of Pediatric Cardiology, New York,1985). A small group of obstetricians was taught tovisualise and understand the normal appearance ofthe four chamber view of the heart and incorporatethis into their routine scan (fig 1). When difficultywas experienced in seeing the normal four chamberview the case was referred to a paediatric cardiolog-ist for further evaluation. During a two year period,20 000 pregnancies were screened, and 39 seriousanomalies were detected. This is in accord with theexpected figure of 2/1000 that might be detected bythis view alone from analysis of data from infants.5Screening for congenital heart disease is thereforepossible, and it is now being increasingly practised in
the United Kingdom. Our success has been lessdramatic than the French, because most routinescans in this country are done at 16 weeks' gestation,whereas in France they are done at 20-24 weekswhen the fetal heart can be seen more easily. Alsosome obstetric units in the United Kingdom arehandicapped by poor ultrasound equipment orlimited time available for adding extra work to theschedule. These problems are slowly being resolved,and we currently have a research programme that istrying to improve the quality and reliability ofscreening. Although there are still a substantialnumber of serious anomalies missed by this pro-gramme the referral pattern over the years haschanged enormously as a result of teaching. Fig 2shows the increasing number of anomalies detectedsince 1980, and the proportion of these diagnosedfrom routine scans.Although the overall frequency of 2 anomalies/
1000 detected by the screening programme isconsistent with the expected rate from postnatal
Fig 1 Normalfour chamber view ofthe fetal heartshowingtwo equally sized atria and ventricles. The heart shouldoccupy about a third of the thorax and the atrial andventricular septa should meet the atrioventricular valves in across at the centre ofthe heart.
Fetal cardiac abnormalities 965
80 E Patients referred with suspected disease and the high rate of intrauterine death1 congenital heart disease suggest that the incidence of congenital heartTotal No disease in early fetal life may be significantly higher
than that in postnatal life.60 Some abnormalities change or progress during
2 fl~~~~~~~~~~~~~~2E Guy'saJ Hospital&9140 Brompton
1980 1981 1982 1983 1984 1985 1986 1987 Z 510 ......
Year......Fig 2 Total number ofdetected abnormalities continues torise, but an increasingproportion is suspected beforereferral.
studies, the range of congenital heart disease is &w°s lc0 #6*' '
different.6 7 Fig 3 shows a comparison of the O 3 ':tdiagnostic categories of cases seen prenatally and ° othose reported postnatally. Severe forms of heart *l j ,disease such as hypoplastic left heart syndrome vpredominate in prenatal life, and forms of heartdisease that are unusual postnatally commonly 20present in fetal life. Tricuspid dysplasia and therelated Ebstein's malformation (fig 4) make up 10%of the anomalies seen in our series. In this conditionthere is gross tricuspid regurgitation and right atrialdilatation, and there may be associated anatomicalor functional pulmonary atresia. The prenatal car- |diomegaly restricts lung development resulting in a
severe pulmonary hypoplasia. Such a neonate isimpossible to resuscitate, and dies soon after deliv- oery. Early death accounts for the fact that thiscondition is rarely encountered in paediatric car-1diological practice.The natural history of some forms of fetal heart
disease is unexpected. Many conditions, includinghypoplastic left heart syndrome, atrioventricular 0septal defects, and critical aortic stenosis may opresent with intrauterine heart failure (fetal hy- kcY C.,C F
drops). Spontaneous intrauterine death occurred in & 01' .*14% of our total series of cases detected prenatally, .p , .Z * &and this does not include those cases where termin- Ak 0g 4*.'ation of pregnancy took place in anticipation of A:C)1intrauterine death. There is a particularly high rate A0of spontaneous fetal loss in cases of heart disease Fig 3 The range ofdisease seen in prenatal life in the Guy'sassociated with chromosomal anomalies; the overall unit compared with the scatter seen in infancy reportedfromincidence in our series of cardiac malformations is the Brompton Hospital, and the New England Regional20%. The finding of unusual forms of prenatal heart Infant Cardiac Care Program.
Fig 4 The heart is enlarged relative to the thorax becausetricuspid valve dysplasia and regurgitation causes atrialdilatation, mainly on the right.
prenatal life. For example, a patient with coarcta-tion of the aorta may develop severe hypoplasia ofthe aortic arch; the pulmonary arteries in a patientwith tetralogy of Fallot may become more hypoplas-tic; and tumours may enlarge as pregnancy pro-gresses. In several cases the pulmonary valve andright ventricle have looked normal in the early fetalstudy but serious pulmonary stenosis has developedby full term.The recurrence of congenital heart disease within
a family group seems to depend on the lesion foundin the index case, but recurrences do not occur inpatterns consistent with generally accepted genetictheories.8 The overall rate of recurrence is close tothe predicted 1/50 but some conditions recur moreoften than others. The hypoplastic left heart syn-drome, coarctation of the aorta, common arterialtrunk, and conditions associated with atrial isomer-ism, are examples of malformations that may recurat a rate of as high as 1/10 when a previous child inthe family has been affected. The recurrence ratewhen a parent is affected is still not clear, but is ofthe order of 1/10.
The advantage of detection of fetal congenital heartdisease
There are several advantages to the prenatal detec-tion of fetal congenital heart disease. The first is togive the parents the option of termination ofpregnancy when severe heart disease with a poorlong term outlook is found early in gestation.Despite advances in paediatric cardiac surgery thereis still a high mortality for many conditions inchildhood, before, during, and after operative treat-
ment. The information provided about the heartshould be as accurate and objective as possible. Theechocardiographer must be aware of the strongassociation of some forms of heart disease withextracardiac and chromosomal anomalies and takesteps to ensure that the parents make decisions withall the possible information available to them. Fig 5shows an atrioventricular septal defect, which led tothe diagnosis of Down's syndrome in the fetus of ayoung mother. Most parents when faced with thefacts of the individual condition found in their fetuschoose termination. In our series 75% have takenthis step when the diagnosis has been made before28 weeks' gestation. If screening becomes morewidespread there are enormous potential implica-tions for the future practice of paediatric cardiology.Most parents have thought of the possibility of fetalabnormality and how they might react if it happenedto them. They already have established views on themorality of abortion in the context of their ownlives. Those who have previous experience ofcongenital heart disease in the family, especially ifthat child died, are usually particularly anxious toavoid repeating the experience. My attitude is thatthis decision belongs to the parents alone and thatthey should be supported whatever decision theymake.
Theoretically, when a prenatal diagnosis of heartdisease is made and antenatal care is transferred to aunit with combined facilities for obstetrics andpaediatric cardiology, the infant should have theoptimum chance of survival. Our results, however,do not bear this out, mainly because of thecomplexity and severity of abnormality selected forpresentation for fetal echocardiography, and the
Fig 5 A large defect in the atrial and ventricular septa at thecrux ofthe heart. There is a common atrioventricular valve,which is open in this diastolicframe.
Fetal cardiac abnormalities 967
incidence of multiple associated extracardiac ano-malies. In our series, of 109 babies born alive only46 survived the first month of life. Of the totalseries, there are only 22 normal children still alivewith either corrected or readily correctable lesions.
Psychological assessment of mothers seen in ourdepartment suggests that a mother is emotionallybetter able to cope with the infant when she has hadtime to prepare herself prenatally for the problemsto be encountered in neonatal life than the motherwho learns of her newborn baby's heart disease inthe postnatal ward. Many doctors feel that if amother finds out about a fetal abnormality duringpregnancy it will interfere with the experience ofpregnancy, but they underestimate most mothers'subconscious and ever present fears of fetal ano-maly. Most mothers would elect to have everypossible test for the detection of fetal abnormalitiesif there was no risk attached to the test. The demandfor fetal echocardiography has risen from 35 highrisk pregnancies in 1980 to over 1000 patients in1987. The few mothers who do not wish prenataldiagnosis of any kind are able to opt out if they wish.A further advantage of fetal echocardiography is
our increased knowledge of the prevalence andnatural history of congenital heart disease in fetallife. This raises the possibility of intervention duringprenatal life. It is possible to insert a needle safelyinto the umbilical cord and even directly into theheart.9 10 Will it ever be possible to dilate an arterialvalve or burn it with a laser at the end of a catheterprenatally? Would this change the outcome inconditions with a poor prognosis, such as criticalaortic stenosis or pulmonary atresia with an intactventricular septum? Some forms of hypoplastic leftheart syndrome have an intact foramen ovale.Would opening this change the course of leftventricular development? These are futuristic con-cepts but perhaps not so impractical as they firstseem. In the United States the time betweenprenatal diagnosis of the hypoplastic left heartsyndrome and delivery is already being used toadvantage by seeking a suitable donor heart fortransplantation. If this proves to be successful in thelong term it may be an avenue to explore in thiscountry. Although I feel that it is fraught withmedical, social, and financial problems, someparents are willing to submit their child to anyprocedure that may give a possible chance ofsurvival. The maximum possible freedom of choice,based on clear, accurate, and unbiased informationshould be available.
Cardiac function in prenatal life
Most heart disease in infancy and childhood is
structural rather than functional. Similarly, dis-turbed anatomy of the heart is the most commondisorder to loQk for prenatally. Functional anoma-lies, however, do occur. As in postnatal life, cardiacfunction can be assessed by looking for peripheralsigns of cardiac failure or, more directly, by echocar-diography. Cardiac failure in utero is manifest notonly by ascites and cardiomegaly as measured by thecardiothoracic ratio, but also by pericardial andpleural effusions and skin oedema. These signs canalso be caused by non-cardiac lesions.
Cardiac function can also be assessed by Dopplerexamination of the intracardiac valves. The passivefilling phase of the atrioventricular valve tracing willgive some indication of ventricular compliance.Atrioventricular valve regurgitation is commonlyassociated with fetal hydrops. Using a product of themean velocity of valvar flow and the area of thevalve orifice, the volume of blood flow expelled byeach side of the heart can be calculated. 11-13Although this estimation is sometimes inaccurate, itis reasonably reproducible. The output from theright heart is always slightly greater than the left,with a ratio of about 1-3:1. Cardiac output is wellmaintained until late in the course of fetalcompromise14 but a fall in output indicates a poorprognosis.
Treatment of the fetus
Some possible future invasive interventional techni-ques have already been mentioned. At present,treatment is confined to giving drugs to the motherto improve the condition of the fetus, but this ismainly for fetal tachycardias.t5 The type of arrhyth-mia must be documented by M mode echocardio-graphy before treatment is started. An atrial tachy-cardia without evidence of cardiac failure willusually be converted to sinus rhythm with digoxinalone. For cardiac failure, digoxin and verapamil orflecainide have been successfully used. Prematuredelivery of a hydropic fetus should be avoided asthese infants have a particularly high incidence ofnecrotising enterocolitis. Once the tachycardia isunder control the hydrops can resolve and deliverycan take place nearer full term.
In summary, ultrasound examination of the fetalheart can accurately diagnose structural and func-tional heart disease. It has provided extensiveinformation on the natural history of many forms ofcongenital heart disease and on normal haemodyna-mics. There are many possibilities for future de-velopments and applications of the technique.
Lange LW, Sahn DJ, Allen HD, Goldberg SJ, Anderson C,Giles H. Qualitative real-time cross-sectional echocardiographic
imaging of the human fetus during the second half of pregnancy.Circulation 1980;62:799-806.
2 Kleinman CS, Hobbins JC, Jaffe CC, Lynch DC, Talner NS.Echocardiographic studies of the human fetus: prenatal diag-nosis of congenital heart disease and cardiac dysrhythmias.Pediatrics 1980;65:1059.
3 Allan LD, Tynan MJ, Campbell S, Wilkinson J, Anderson RH.Echocardiographic and anatomical correlates in the fetus. BrHeart J 1980;44:444-51.
4 Allan LD, Crawford DC, Chita SK, Tynan MJ. Prenatalscreening for congenital heart disease. Br Med J 1986;292:1717-19.
5 Flyer DC, Buckley LP, Hellenbrand WE, Cohn HE. Report ofthe New England regional infant cardiac care programme.Pediatrics 1980;65:375-461.
6 Scott DJ, Rigby ML, Miller GAH, Shinebome EA. Thepresentation of symptomatic heart disease in infancy based on10 years experience (1973-82). Implications for the provision ofservices. Br Heart J 1984;52:248-57.Allan LD, Crawford DC, Anderson RH, Tynan MJ. Spectrumof congenital heart disease detected echocardiographically inprenatal life. Br Heart J 1985;54:523-6.
8 Allan LD, Crawford DC, Chita SK, Anderson RH, Tynan MJ.The familial recurrence of congenital heart disease in aprospective series of mothers referred for fetal echocardio-graphy. Am J Cardiol 1986;58:334-7.
9 Daffos F, Forestier F, Pavlovsky MC. Fetal blood samplingduring the third trimester of pregnancy. Br J Obstet Gynaecol1984;91:118-21.
10 Westgren M, Selbing A, Stangenberg M. Fetal intracardiactransfusions in patients with severe rhesus isoimmunisation. BrMed J 1988;296:885-6.
" De Smedt MCH, Visser GHA, Meijboom EJ. Fetal cardiacoutput estimated by Doppler echocardiography during mid- andlate gestation. Am J Cardiol 1987;60:338-42.
12 Kenny JF, Plappert T, Doubilet P, et al. Changes in intracardiacblood flow velocities and right and left ventricular strokevolumes with gestational age in the normal human fetus: aprospective Doppler echocardiographic study. Circulation1986;74:1208-16.
13 Allan LD, Chita SK, Al-Ghazali WH, Crawford DC, Tynan MJ.Doppler echocardiographic evaluation of the normal humanfetal heart. Br Heart J 1987;57:528-33.
14 Al-Ghazali WH, Chapman MG, Allan LD. Doppler assessmentof the cardiac and utero-placental circulations in normal andabnormal fetuses. Br J Obstet Gynaecol 1988;95:575-80.
15 Maxwell DJ, Crawford DC, Curry PVM, Tynan MJ, Allan LD.Obstetric significance, diagnosis and management of fetaltachycardias. Br Med J 1988;297:107-10.
Correspondence to Dr LD Allan, Guy's Hospital, London SE19RT.