Pediatr Cardiol 13:1-4, 1992
Pediatric Cardiology 9 Springer-Veflag New York Inc. 1992
Original Articles P r o g n o s i s in Fetal T e t r a l o g y o f Fallot Lindsey D. Allan and Gurleen K. Sharland Department of Perinatal Cardiology, Guy's Hospital, London, UK
SUMMARY. The outcome in a series of 23 cases of tetralogy of Fallot diagnosed prenatally was compared to published figures for this condition when (a) identified postnatally or (b) treated surgically. There was a marked difference in the survival between these groups, with the highest mortality (75%) occurring when the diagnosis had been made prenatally. The high incidence of chromosomal and extracardiac anomalies in this group (60%) largely accounted for the discrepancy. This, and the potential for progressive changes in the anatomy of the defect during pregnancy, must be taken into account by the pediatric cardiologist offering prognosis in early pregnancy. KEY WORDS: Tetralogy of Fallot h Fetus
The combination of a malalignment ventricular septal defect, right ventricular hypertrophy, pulmonary outflow tract obstruction, and aortic override was first described by Steno in 1673 [9], but the condition received its eponym after Fallot, in 1888, reported the findings in a postmortem specimen of a young adult [5]. Tetralogy of Fallot (TF) is found in about 3.5% of infants with congenital heart disease (CHD) [10], but it has been recognized as the most common cause of cyanotic CHD to be associated with survival into the second and third decades [1]. Untreated, 30% of children will survive to 10 years of age, whereas other forms of cyanotic CHD are fatal before this time. It was first treated surgically by Blalock and Taussig in 1945 by the creation of an aorticopulmonary shunt [3], and total correction was first described in 1954 by Lillehei et al. [8]. During those early years it became evident that the mortality for total correction was high in infants; therefore, a two-stage repair evolved of a palliative shunt procedure in infancy, followed by definitive correction at 2-4 years of age [2]. However, the surgical results have continued to improve as techniques have advanced, some surgeons now reporting no infant deaths even at early total correction [4]. Thus, this condition has become widely recognized as a common and benign form of CHD. A recent report from the Washington-Baltimore InAddress offprint requests to: Dr. L. Allan, 15th Floor, Guy's Tower, Guy's Hospital London, SE1 9RT, UK.
fant Study, which collected the data from all liveborns with this diagnosis, disclosed a different picture [6]. They found that there was a 20% mortality associated with this condition. It became apparent to us that there appeared to be an even higher mortality when the diagnosis of TF was made prenatally. We, therefore, examined our case material in detail to try to account for this.
Patients and Methods Over 7000 high-risk pregnancies have been examined in our unit by fetal echocardiography since early 1980. Pregnancies were selected as being at high risk on the basis of maternal factors, such as a family history of CHD, maternal diabetes, or exposure to known teratogens, or because of abnormal findings on fetal ultrasound examination, such as arrhythmias, nonimmune hydrops, extracardiac, or suspected cardiac anomalies. The mothers were examined with an Advanced Technical Laboratories Mark 3 sector scanner initially, and later with an Advanced Technical Laboratories Mark 4 sector scanner or a HewlettPackard 77020A phased-array scanner. Doppler and color flow mapping have been available in the last 5 years of the study time. In general, a 5-MHz transducer was used, but occasionally 3 MHz was necessary for adequate penetration in later pregnancy. Of over 500 cases of CHD recognized in prenatal life, 23 had the features of TF, that is a malalignment ventricular septal defect (VSD) with evidence of pulmonary outflow tract obstruction. In fetal life, pulmonary outflow obstruction was diagnosed when the main pulmonary artery was smaller than the aorta. Infundibular narrowing was not always present prenatally and Doppler velocities could be normal or only mildly elevated.
2
Pediatric Cardiology Vol. 13, No. I, !992
it
[] FAMILYHIST [] F,ABN
[] ?Crib [] [] [] []
[] []
EXTRACARDIAC ANOMALY NO EXTRACARDIAC ANOMALY
?CHD+F.HIST ?CHD+F.ABN F.HYDROPS F.ARR
g-----i
"6
Fig. 1. Reasons for referral of 23 cases subsequently found to have tetralogy of Fallot. F.ABN, fetal abnormality; ?CHD, suspected congenital heart disease; F.HIST, family history; F.HYDROPS, fetal hydrops; F.ARR, fetal arrhythmia.
E z
0 TOP
IUD
NND
i il INFD
ALIVE
OUTCOME Right ventricular hypertrophy is not a recognizable feature of TF in the fetus. Fetuses with a typical VSD but with pulmonary atresia were excluded from the study. Data on the outcome of each pregnancy with TF were collected and studied. The reason for referral, gestational age at presentation, and the presence of extracardiac anomalies were noted.
Results
The reason for referral (Fig. 1) was the presence of an extracardiac anomaly in 13 fetuses. In two of these 13, the heart was also suspected to be abnormal by the referring obstetrician. In five cases, the heart alone was thought to be abnormal when examined at a routine scan; one of these mothers also had a family history of CHD. There were three referrals because of a family history of CHD, and one because of fetal bradycardia. The latter fetus had complete heart block in association with the TF. One fetus presented because of hydrops. The gestational age at presentation varied between 18 and 38 weeks, 52% of cases being recognized after 24 weeks. Fourteen fetuses proved to have extracardiac anomalies, six of whom had chromosomal defects. There were four cases of trisomy 18, and one each of trisomy 21 and triploidy. Of the nonchromosomal anomalies, there were five cases of exomphalos, and one each of hydrocephalus, VATER syndrome, and diaphragmatic hernia. In three of the continuing pregnancies the pulmonary artery failed to keep pace with normal growth and, therefore, became relatively more hypoplastic as pregnancy advanced. In a further one patient, an additional VSD became evident later in gestation. The outcome in these 23 cases is summarized in Fig. 2. Seven sets of parents chose termination of pregnancy after counselling; four of these seven fetuses had chromosomal defects and a further one
Fig. 2. Outcome in 23 cases tetralogy of Fallot. TOP, termination of pregnancy; IUD, intrauterine death; NND, neonatal death; INFD, infant death.
had the absent pulmonary valve syndrome in association with TF. There were four spontaneous intrauterine deaths; three had extracardiac anomalies and the remaining fetus had the absent pulmonary valve syndrome. There were five neonatal deaths, all associated with multiple congenital abnormalities. There were three infant deaths, one of whom had had an exomphalos repaired as a neonate. There are four survivors, two of whom have had corrective surgery and a further one has had a shunt inserted. One of these fetuses also had exomphalos. Thus, for the continuing 16 pregnancies there was a 75% mortality. Or, of five infants born alive with no other anomalies, there are three survivors, representing a 40% mortality. One of the two babies who died had fits, thought to be associated with anoxic brain damage, before and after a surgical shunt procedure. The other died in the postoperative period at 8 months of age after total correction. His condition was complicated by congenital complete heart block requiring pacing since birth.
Discussion
When the prenatal and postnatal series are compared it is clear that the outcome is much worse when the diagnosis is made prenatally than when it is found postnatally. There was a 75% mortality in the prenatal series compared to 20% in the infants. However, the incidence of associated abnormalities in the prenatal series is higher. There were additional anomalies in 60% of our total group, compared to 28% in the unselected series of infants.
Allan and Shariand: Fetal Tetralogy of Fallot
There is no doubt that the method of selection of patients for fetal echocardiography yields a spectrum of TF which is strongly associated with other anomalies--in fact over half the fetuses were referred because of a fetal anomaly. The majority of uncomplicated cases of TF will not be recognized by four-chamber view screening and will thus not be detected in utero except by specialized scanning, such as was electively performed in three of our cases because of a family history of CHD. This small group of cases might be expected to have a better prognosis, but two of the three sets of parents chose interruption of pregnancy, despite being given an optimistic prognosis. In one case this was because a previous child had died of CHD. In the other case, where termination was chosen, the father had had TF corrected as a child. It is unfortunate, in view of the frequency of associated anomalies and the eventual poor outcome, that over half were not detected till after 24 weeks gestation. This, however, may be expected to change as the expertise in detecting extracardiac and cardiac abnormalities continues to improve in local hospitals. In four of the 16 continuing pregnancies the echocardiographic appearance of the anatomical defect changed as pregnancy advanced. In three cases, pulmonary outflow tract obstruction progressed and the pulmonary arteries failed to grow normally, resulting in increasing pulmonary artery hypoplasia. An additional VSD became evident in a further case with advancing gestation. Each continuing pregnancy, therefore, must be followed throughout gestation to observe the developing picture, as these changes may influence the prognosis for corrective surgery. However, some of the factors which can influence the results of surgery may not be detectable prenatally, for example, multiple small VSDs or pulmonary artery branch stenosis [71. The surgical mortality, variously reported as 0, 6, or 11 5 % , is considerably different from the 20% overall mortality in the group of 96 infants born during a 5-year period in the Baltimore-Washington area [4, 11]. Selection of cases for surgery takes place in such a way that those with multiple anomalies either die in utero, are allowed to die as neonates or die related to extracardiac surgery, or are not submitted for surgery. It is well recognized that additional cardiac defects, such as multiple VSDs, the absent pulmonary valve syndrome, peripheral pulmonary stenosis, or heart block occurring in association with TF, will deleteriously affect the results of surgery. It is possible that, in the series with a very low surgical mortality, such examples of the spectrum of TF are underrepresented, presumably by chance. It is also important to recognize
3
that surgeons are inclined to report their best results. The mortality related to cardiac surgery in the unselected postnatal series was 11.5%. It is intriguing to speculate, in embryological terms, why this form of CHD is so frequently associated with extracardiac anomalies, in contrast to some conditions, such as transposition of the great arteries, which are very rarely found with other defects. In our series, 60% of cases had additional anomalies. It is very likely that at conception there are even more cases of TF which pass undiagnosed because of spontaneous early fetal loss. The condition can be induced with regularity in the animal model by administering a bisdiamine to pregnant rats [12]. The timing of the drug damage correlates with neural crest migration. It is this abnormal migration which is thought to produce the incorrect conotruncal formation. At this stage in development the gut is still exteriorized, and the normal process of resolution of this exomphalos may be damaged at the same time, accounting for the frequent association of exomphalos with TF. In trisomies it is possible that the function of the neural crest is fundamentally impaired or, alternatively, more susceptible to an external insult. In summary, it is important to be aware that TF, when detected prenatally, is not the benign condition suggested by the publicized surgical results or even by the outcome in those cases identified as infants. When the diagnosis is made in the fetus, the first step is to exclude chromosomal anomalies by fetal karyotype. The next step is to examine the fetus thoroughly by ultrasound to exclude nonchromosomal syndromes or associated anomalies. Third, the anatomy of the lesion itself must be observed. The absent pulmonary valve syndrome can be readily diagnosed and will adversely influence prognosis. Similarly, severe pulmonary artery hypoplasia or atresia in early pregnancy will not have such a favorable outlook [7]. Only when all this information is assimilated can a prognosis be offered, but it may still require modification as pregnancy advances. Thus, the pediatric cardiologist becoming involved in this new speciality must be aware of the pitfalls in prognosis in this form of CHD.
Acknowledgment. This work was supported by the British Heart Foundation.
References 1. Anderson RH, Macartney FJ, Shinebourne EA, Tyrian M (eds) (1987) In: Paediatric Cardiology. Churchill Livingstone, Edinburgh, p 765
4
Pediatric Cardiology Vol. 13, No. l, 1992
2. Barratt-Boyes BG, Neutze JM (1973) Primary repair of tetralogy of Fatlot in infancy using profound hypothermia with circulatory arrest and limited cardiopulmonary bypass. A comparison with conventional two-state management. Ann Surg 178:406-411 3. Blalock A, Taussig HB (1945) The surgical treatment of malformation of the heart in which there is pulmonary stenosis or pulmonary atresia. JAMA 128:189-192 4. Castenada AF, Freed MD, Williams RG, Norwood W (1977) Repair of tetralogy of Fallot in infancy. Early and late results. J Thorac Cardiovasc Surg 74:372-381 5. Fallot A (1888) Contribution a l'anatomie pathologique de la maladie bleue (cyanose cardiaque). Marseille Medicine 25:77-403 6. Karr SS, Brenner J, Neill C, Ferenze C, Perry L (1988) Tetralogy of Fallot: Identification and outcome. A regional experience, 1981-1985 [abstract]. 21st Annual Scientific Sessions of the Southeastern Pediatric Cardiology Society (unpublished) 7. Kirklin JW, Blackstone EH, Pacifico AD, Kirklin JK,
8.
9. 10.
11.
12.
Bargeron LM (1984) Risk factors of early and late failure after repair of tetralogy of Fallot and their neutraIization. Thoracic Cardiovasc Surg 32:208-214 Lillehei CW, Cohen W, Warden HE, Read RC, Aust JB, De Wall RA, Varco RL (1955) Direct vision intracardiac surgical correction of tetralogy of Fallot, pentalogy of Fallot and pulmonary atresia defects. Report of 10 cases. Ann Surg 142:418-443 Marquis RM (1956) Longevity and the early history of the tetralogy of Fallot. Br Med J 1:819-822 Mitchell SC, Korones SB, Berendes HW (1971) Congenital heart disease in 56,109 births: Incidence and natural history. Circulation 43:323-332 Pacifico AD, Sand ME, Bargeron LM, Colvin EC (1987) Transatrial-transpulmonary repair of tetralogy of Fallot. J Thoracic Cardiovasc Surg 93:919-924 Taleporos P, Salgo MP, Oster G (1978) Teratogenic action of a bis(dichloroacetyl)diamine on rats: Patterns of malformation produced in high incidence at time limited periods of development. Teratology 18:5-16
Pediatric Cardiology 9 Springer-Veflag New York Inc. 1992
Original Articles P r o g n o s i s in Fetal T e t r a l o g y o f Fallot Lindsey D. Allan and Gurleen K. Sharland Department of Perinatal Cardiology, Guy's Hospital, London, UK
SUMMARY. The outcome in a series of 23 cases of tetralogy of Fallot diagnosed prenatally was compared to published figures for this condition when (a) identified postnatally or (b) treated surgically. There was a marked difference in the survival between these groups, with the highest mortality (75%) occurring when the diagnosis had been made prenatally. The high incidence of chromosomal and extracardiac anomalies in this group (60%) largely accounted for the discrepancy. This, and the potential for progressive changes in the anatomy of the defect during pregnancy, must be taken into account by the pediatric cardiologist offering prognosis in early pregnancy. KEY WORDS: Tetralogy of Fallot h Fetus
The combination of a malalignment ventricular septal defect, right ventricular hypertrophy, pulmonary outflow tract obstruction, and aortic override was first described by Steno in 1673 [9], but the condition received its eponym after Fallot, in 1888, reported the findings in a postmortem specimen of a young adult [5]. Tetralogy of Fallot (TF) is found in about 3.5% of infants with congenital heart disease (CHD) [10], but it has been recognized as the most common cause of cyanotic CHD to be associated with survival into the second and third decades [1]. Untreated, 30% of children will survive to 10 years of age, whereas other forms of cyanotic CHD are fatal before this time. It was first treated surgically by Blalock and Taussig in 1945 by the creation of an aorticopulmonary shunt [3], and total correction was first described in 1954 by Lillehei et al. [8]. During those early years it became evident that the mortality for total correction was high in infants; therefore, a two-stage repair evolved of a palliative shunt procedure in infancy, followed by definitive correction at 2-4 years of age [2]. However, the surgical results have continued to improve as techniques have advanced, some surgeons now reporting no infant deaths even at early total correction [4]. Thus, this condition has become widely recognized as a common and benign form of CHD. A recent report from the Washington-Baltimore InAddress offprint requests to: Dr. L. Allan, 15th Floor, Guy's Tower, Guy's Hospital London, SE1 9RT, UK.
fant Study, which collected the data from all liveborns with this diagnosis, disclosed a different picture [6]. They found that there was a 20% mortality associated with this condition. It became apparent to us that there appeared to be an even higher mortality when the diagnosis of TF was made prenatally. We, therefore, examined our case material in detail to try to account for this.
Patients and Methods Over 7000 high-risk pregnancies have been examined in our unit by fetal echocardiography since early 1980. Pregnancies were selected as being at high risk on the basis of maternal factors, such as a family history of CHD, maternal diabetes, or exposure to known teratogens, or because of abnormal findings on fetal ultrasound examination, such as arrhythmias, nonimmune hydrops, extracardiac, or suspected cardiac anomalies. The mothers were examined with an Advanced Technical Laboratories Mark 3 sector scanner initially, and later with an Advanced Technical Laboratories Mark 4 sector scanner or a HewlettPackard 77020A phased-array scanner. Doppler and color flow mapping have been available in the last 5 years of the study time. In general, a 5-MHz transducer was used, but occasionally 3 MHz was necessary for adequate penetration in later pregnancy. Of over 500 cases of CHD recognized in prenatal life, 23 had the features of TF, that is a malalignment ventricular septal defect (VSD) with evidence of pulmonary outflow tract obstruction. In fetal life, pulmonary outflow obstruction was diagnosed when the main pulmonary artery was smaller than the aorta. Infundibular narrowing was not always present prenatally and Doppler velocities could be normal or only mildly elevated.
2
Pediatric Cardiology Vol. 13, No. I, !992
it
[] FAMILYHIST [] F,ABN
[] ?Crib [] [] [] []
[] []
EXTRACARDIAC ANOMALY NO EXTRACARDIAC ANOMALY
?CHD+F.HIST ?CHD+F.ABN F.HYDROPS F.ARR
g-----i
"6
Fig. 1. Reasons for referral of 23 cases subsequently found to have tetralogy of Fallot. F.ABN, fetal abnormality; ?CHD, suspected congenital heart disease; F.HIST, family history; F.HYDROPS, fetal hydrops; F.ARR, fetal arrhythmia.
E z
0 TOP
IUD
NND
i il INFD
ALIVE
OUTCOME Right ventricular hypertrophy is not a recognizable feature of TF in the fetus. Fetuses with a typical VSD but with pulmonary atresia were excluded from the study. Data on the outcome of each pregnancy with TF were collected and studied. The reason for referral, gestational age at presentation, and the presence of extracardiac anomalies were noted.
Results
The reason for referral (Fig. 1) was the presence of an extracardiac anomaly in 13 fetuses. In two of these 13, the heart was also suspected to be abnormal by the referring obstetrician. In five cases, the heart alone was thought to be abnormal when examined at a routine scan; one of these mothers also had a family history of CHD. There were three referrals because of a family history of CHD, and one because of fetal bradycardia. The latter fetus had complete heart block in association with the TF. One fetus presented because of hydrops. The gestational age at presentation varied between 18 and 38 weeks, 52% of cases being recognized after 24 weeks. Fourteen fetuses proved to have extracardiac anomalies, six of whom had chromosomal defects. There were four cases of trisomy 18, and one each of trisomy 21 and triploidy. Of the nonchromosomal anomalies, there were five cases of exomphalos, and one each of hydrocephalus, VATER syndrome, and diaphragmatic hernia. In three of the continuing pregnancies the pulmonary artery failed to keep pace with normal growth and, therefore, became relatively more hypoplastic as pregnancy advanced. In a further one patient, an additional VSD became evident later in gestation. The outcome in these 23 cases is summarized in Fig. 2. Seven sets of parents chose termination of pregnancy after counselling; four of these seven fetuses had chromosomal defects and a further one
Fig. 2. Outcome in 23 cases tetralogy of Fallot. TOP, termination of pregnancy; IUD, intrauterine death; NND, neonatal death; INFD, infant death.
had the absent pulmonary valve syndrome in association with TF. There were four spontaneous intrauterine deaths; three had extracardiac anomalies and the remaining fetus had the absent pulmonary valve syndrome. There were five neonatal deaths, all associated with multiple congenital abnormalities. There were three infant deaths, one of whom had had an exomphalos repaired as a neonate. There are four survivors, two of whom have had corrective surgery and a further one has had a shunt inserted. One of these fetuses also had exomphalos. Thus, for the continuing 16 pregnancies there was a 75% mortality. Or, of five infants born alive with no other anomalies, there are three survivors, representing a 40% mortality. One of the two babies who died had fits, thought to be associated with anoxic brain damage, before and after a surgical shunt procedure. The other died in the postoperative period at 8 months of age after total correction. His condition was complicated by congenital complete heart block requiring pacing since birth.
Discussion
When the prenatal and postnatal series are compared it is clear that the outcome is much worse when the diagnosis is made prenatally than when it is found postnatally. There was a 75% mortality in the prenatal series compared to 20% in the infants. However, the incidence of associated abnormalities in the prenatal series is higher. There were additional anomalies in 60% of our total group, compared to 28% in the unselected series of infants.
Allan and Shariand: Fetal Tetralogy of Fallot
There is no doubt that the method of selection of patients for fetal echocardiography yields a spectrum of TF which is strongly associated with other anomalies--in fact over half the fetuses were referred because of a fetal anomaly. The majority of uncomplicated cases of TF will not be recognized by four-chamber view screening and will thus not be detected in utero except by specialized scanning, such as was electively performed in three of our cases because of a family history of CHD. This small group of cases might be expected to have a better prognosis, but two of the three sets of parents chose interruption of pregnancy, despite being given an optimistic prognosis. In one case this was because a previous child had died of CHD. In the other case, where termination was chosen, the father had had TF corrected as a child. It is unfortunate, in view of the frequency of associated anomalies and the eventual poor outcome, that over half were not detected till after 24 weeks gestation. This, however, may be expected to change as the expertise in detecting extracardiac and cardiac abnormalities continues to improve in local hospitals. In four of the 16 continuing pregnancies the echocardiographic appearance of the anatomical defect changed as pregnancy advanced. In three cases, pulmonary outflow tract obstruction progressed and the pulmonary arteries failed to grow normally, resulting in increasing pulmonary artery hypoplasia. An additional VSD became evident in a further case with advancing gestation. Each continuing pregnancy, therefore, must be followed throughout gestation to observe the developing picture, as these changes may influence the prognosis for corrective surgery. However, some of the factors which can influence the results of surgery may not be detectable prenatally, for example, multiple small VSDs or pulmonary artery branch stenosis [71. The surgical mortality, variously reported as 0, 6, or 11 5 % , is considerably different from the 20% overall mortality in the group of 96 infants born during a 5-year period in the Baltimore-Washington area [4, 11]. Selection of cases for surgery takes place in such a way that those with multiple anomalies either die in utero, are allowed to die as neonates or die related to extracardiac surgery, or are not submitted for surgery. It is well recognized that additional cardiac defects, such as multiple VSDs, the absent pulmonary valve syndrome, peripheral pulmonary stenosis, or heart block occurring in association with TF, will deleteriously affect the results of surgery. It is possible that, in the series with a very low surgical mortality, such examples of the spectrum of TF are underrepresented, presumably by chance. It is also important to recognize
3
that surgeons are inclined to report their best results. The mortality related to cardiac surgery in the unselected postnatal series was 11.5%. It is intriguing to speculate, in embryological terms, why this form of CHD is so frequently associated with extracardiac anomalies, in contrast to some conditions, such as transposition of the great arteries, which are very rarely found with other defects. In our series, 60% of cases had additional anomalies. It is very likely that at conception there are even more cases of TF which pass undiagnosed because of spontaneous early fetal loss. The condition can be induced with regularity in the animal model by administering a bisdiamine to pregnant rats [12]. The timing of the drug damage correlates with neural crest migration. It is this abnormal migration which is thought to produce the incorrect conotruncal formation. At this stage in development the gut is still exteriorized, and the normal process of resolution of this exomphalos may be damaged at the same time, accounting for the frequent association of exomphalos with TF. In trisomies it is possible that the function of the neural crest is fundamentally impaired or, alternatively, more susceptible to an external insult. In summary, it is important to be aware that TF, when detected prenatally, is not the benign condition suggested by the publicized surgical results or even by the outcome in those cases identified as infants. When the diagnosis is made in the fetus, the first step is to exclude chromosomal anomalies by fetal karyotype. The next step is to examine the fetus thoroughly by ultrasound to exclude nonchromosomal syndromes or associated anomalies. Third, the anatomy of the lesion itself must be observed. The absent pulmonary valve syndrome can be readily diagnosed and will adversely influence prognosis. Similarly, severe pulmonary artery hypoplasia or atresia in early pregnancy will not have such a favorable outlook [7]. Only when all this information is assimilated can a prognosis be offered, but it may still require modification as pregnancy advances. Thus, the pediatric cardiologist becoming involved in this new speciality must be aware of the pitfalls in prognosis in this form of CHD.
Acknowledgment. This work was supported by the British Heart Foundation.
References 1. Anderson RH, Macartney FJ, Shinebourne EA, Tyrian M (eds) (1987) In: Paediatric Cardiology. Churchill Livingstone, Edinburgh, p 765
4
Pediatric Cardiology Vol. 13, No. l, 1992
2. Barratt-Boyes BG, Neutze JM (1973) Primary repair of tetralogy of Fatlot in infancy using profound hypothermia with circulatory arrest and limited cardiopulmonary bypass. A comparison with conventional two-state management. Ann Surg 178:406-411 3. Blalock A, Taussig HB (1945) The surgical treatment of malformation of the heart in which there is pulmonary stenosis or pulmonary atresia. JAMA 128:189-192 4. Castenada AF, Freed MD, Williams RG, Norwood W (1977) Repair of tetralogy of Fallot in infancy. Early and late results. J Thorac Cardiovasc Surg 74:372-381 5. Fallot A (1888) Contribution a l'anatomie pathologique de la maladie bleue (cyanose cardiaque). Marseille Medicine 25:77-403 6. Karr SS, Brenner J, Neill C, Ferenze C, Perry L (1988) Tetralogy of Fallot: Identification and outcome. A regional experience, 1981-1985 [abstract]. 21st Annual Scientific Sessions of the Southeastern Pediatric Cardiology Society (unpublished) 7. Kirklin JW, Blackstone EH, Pacifico AD, Kirklin JK,
8.
9. 10.
11.
12.
Bargeron LM (1984) Risk factors of early and late failure after repair of tetralogy of Fallot and their neutraIization. Thoracic Cardiovasc Surg 32:208-214 Lillehei CW, Cohen W, Warden HE, Read RC, Aust JB, De Wall RA, Varco RL (1955) Direct vision intracardiac surgical correction of tetralogy of Fallot, pentalogy of Fallot and pulmonary atresia defects. Report of 10 cases. Ann Surg 142:418-443 Marquis RM (1956) Longevity and the early history of the tetralogy of Fallot. Br Med J 1:819-822 Mitchell SC, Korones SB, Berendes HW (1971) Congenital heart disease in 56,109 births: Incidence and natural history. Circulation 43:323-332 Pacifico AD, Sand ME, Bargeron LM, Colvin EC (1987) Transatrial-transpulmonary repair of tetralogy of Fallot. J Thoracic Cardiovasc Surg 93:919-924 Taleporos P, Salgo MP, Oster G (1978) Teratogenic action of a bis(dichloroacetyl)diamine on rats: Patterns of malformation produced in high incidence at time limited periods of development. Teratology 18:5-16
