The activities of the Hungarian Centre for Congenital Anomaly Control / Andrew Czeizel

- 219 - THE ACTIVITIES OF THE HUNGARIAN CENTRE FOR CONGENITAL ANOMALY CONTROL Andrew Czeizel a In Hungary today congenital anomalies (CAs) are the eighth leading cause of death and the second cause of infant mortality, accounting for 23% of all infant deaths. Altogether, CAs are responsible for 5 OOO years of lost life and about 4 500 years of actually impaired life within the population each year (1). In the early 1960s the grow- ing public health importance of CAs and in particular the thalidomide tragedy prompted Hungarian paediatricians to suggest the establishment of a CA registry. In 1962 the notification of CAs was made compulsory and the Hungarian Congenital Malformation Registry (HCMR) was established. In 1970 the Department of Human Genetics and Teratology of the National Institute of Hygiene was charged with the direction of the HCMR according to the recommendations of WHO and other expert groups. Other systems were later added to the HCMR and the Hungarian Centre for Congenital Anomaly Control (HCCAC) was established in 1973. The Hungar- ian experience might be useful as a possible model for other countries interested in this field of epidemiology and public health. Hungary had a reasonably stable popu- lation size of 10.3-10. 7 million between 1970 and 1986, with an average of about 150 OOO births per year. The HCCAC has three major activities: maintaining a register, conducting surveillance programmes, and carrying out special diagnostic, preventive and rehabili- tation tasks. The Centre is staffed by a team of seven: a physician, a mathematician, a biologist and four techni- cal assistants who work full-time in close collaboration with a group of computer experts and other units, par- ticularly the Epidemiology Unit of the Department of Human Genetics and Teratology of the National Institute of Hygiene. The Registry A genetic register can be defined as a system for the collection, storage and application of data on diseases and/or cases where mutation is the cause of harmful or potentially disadaptive manifestations.b Some uses of registers were outlined by Miller in 1976 (2). The Centre's register function is fulfilled by the HCMR which collects and maintains a computerized permanent record of medical and personal information about malformed newborns and infants born in Hungary since 1970 (3). The study population comprises all births. The purpose of the HCMR is to determine as reliably as possible base- line figures of different types of recorded specific CA entities in order to help plan medical and social services for those affected; to estimate the public health impor- tance of different groups of CAs (Tables 1 & 2) so that • Department of Human Genetics and Teratology, WHO Collaborating Centre for the Community Control of Hereditary Diseases, National Insti- tute of Hygiene, Budapest, Hungary. b World Health Organization. The methodology for binh defects mon- itoring. Report on a WHO consultation, Rome, 3-4 June 1985. (Document HMG/Cons/85.6). c The term "notification" is intentionally used instead of ··report" because the input of data from medical services to the HCMR is called notification while the output of data from the HCMR to medical services is named report. Wld hlth statist. quan .• 41 ( 1988) resources can be allocated properly; to detect temporal and/or spatial changes in birth prevalences; to give adequate information in connection with public concerns (for example, the Chernobyl nuclear plant incident); to organize population-based prevention programmes and to check the efficacy of prenatal and neonatal screen- ings; to help the social rehabilitation of handicapped children through special programmes; and to provide material for research projects. Following are some special features of the HCMR. • The neonatal period, mainly the first days of life, is the usual time for diagnosis and notificationc of CAs; how- ever. the study period covers the interval from birth to the age of 1 year. Recently, prenatally diagnosed and selectively aborted malformed fetuses have also been registered. • Notification of CAs is compulsory as is the case for several infectious diseases. (At present, the number of infant deaths due to CAs is 26 times higher than the total of deaths due to infectious diseases in Hungary.) • Notification is exclusively the task of physicians, mainly obstetricians (in Hungary nearly all deliveries take place in hospital) and paediatricians (who are working in obstetric inpatient clinics and various inpatient and out- patient paediatric clinics). They notify the Congenital Malformation Registry directly using a standard form. In case of death (stillbirths and infant deaths due to con- genital anomalies) the pathologist sends a copy of the detailed autopsy record to the Registry. Autopsy is obli- gatory for all infant deaths. • In addition to the multiple sources of notification out- lined above, notification is made each time an affected child is admitted to any paediatric institution. Although this results in a considerable overlap, duplicate registra- tion on the same child can be eliminated with the help of personal data. At the same time this overlap increases the efficiency of ascertainment and enhances the validity of diagnosis. Furthermore, it provides an opportunity to cross-check the notification. Failure to notify could be detected on the basis of recorded data (paediatric noti- fication forms and autopsy records provide the address of the institution where delivery took place; autopsy records include the address of the paediatric inpatient clinic where death occurred). Additionally, the observed annual total CA rate is compared with the rate expected in each obstetrical and paediatric inpatient clinic. • Submitted data (mainly diagnoses) are critically evalu- ated upon receipt, according to the following proce- dure: (i) Forms with missing or unreliable data are sent back for completing or checking. (ii) CAs are divided into two categories: isolated (sin- ~le, complex or monotopic field defect, polytopic field defect, sequence) and multiple (CA-syndrome, CA-association, unidentified multiple CAs). Distin- guishing between isolated and multiple CAs is - 220 - TABLE 1. TRUE AND REGISTERED BIRTH PREVALENCE OF 10 COMMON CONGENITAL ANOMALIES, COMPLETENESS OF NOTIFICATION, ANNUAL TOTAL YEARS LOST AND ACTUALLY IMPAIRED LIFE, HUNGARY, 1973-1982 TABLEAU 1. PREVALENCE REELLE ET ENREGISTREE A LA NAISSANCE DE 10 ANOMALIES CONGENITALES COURANTES SELON LE POURCENTAGE DES NOTIFICATIONS COMPLETES, LE NOMBRE TOTAL D'ANNEES PERDUES ET LE NOMBRE D'ANNEES DE VIE AVEC UN HANDICAP, HONGRIE, 1973-1982 ICD code Code CIM 740.0- 741.0 745.3-4 749.1-2 750.5 752.5 752.6 754.3 754.5 550.0 758.0 Common congenital anomalies Anomalies cong6nitales courantes Isolated - Anomalies isolees Neural tube defect/anencephalus and/or spina bifida cystica - Anomalie du tube neural/anencephalie et/ou spina bifida cystica Ventricular septa! defect - Communication interventriculaire Cleft lip ± cleft palate - Bee de lievre ± fissure du palais Congenital hypertrophic pyloric stenosis - Stenose congeni- tale hypertrophique du pylore Undescended testis after third month - Ectopie testiculaire apres le troisieme mois Hypospadias - Hypospadias Liability for congenital dislocation of hip - Predisposition a la luxation congenitale de la hanche Congenital structural talipes equinovarus - Pied bot varus equin congenital Congenital inguinal hernia - Hernie inguinale congenitale Multiple - Anomalies multiples Down's syndrome - Syndrome de Down True binh Recorded binh prevalence per prevalence 1 OOO total per 1 OOO binhs total binhs Prevalence Prevalence reelle a la enregistree a naissance la naissance pour 1 OOO pour 1 OOO naissances naissances 1.7 1.5 1.0 1.5 3.6 2.2 28.0 1.3 11.4 1.2 1.6 1.2 1.0 1.3 1.1 1.7 8.6 1.9 3.3 0.9 Completeness of notification % Notifications completes % 94 80 100 87 31 77 31 146 29 75 Total years lost Nombre total d'annees perdues 621 92 22 0 0 0 0 0 4 283 Total years of actually impaired life Nombre total d'annees de vie avec un handicap 189 0 141 0 980 308 180 101 0 536 TABLE 2. TRUE AND REGISTERED BIRTH PREVALENCE OF ISOLATED MODERATELY FREQUENT CONGENITAL ANOMALIES, COMPLETENESS OF NOTIFICATION, ANNUAL TOTAL YEARS LOST AND YEARS OF ACTUALLY IMPAIRED LIFE, HUNGARY, 1973-1982 TABLEAU 2. PREVALENCE REELLE ET ENREGISTREE A LA NAISSANCE D'ANOMALIES CONGENITALES ISOLEES MODEREMENT FREQUENTES, SELON LE POURCENTAGE DES NOTIFICATIONS COMPLETES, LE NOMBRE TOTAL D'ANNEES PERDUES ET LE NOMBRE D'ANNEES AVEC UN HANDICAP, HONGRIE, 1973-1982 ICD code Code CIM 742.1 742.3 745.0 745.1 745.2 745.3 746.3-4 747.0 747.1 747.3 749.0 750.3 751.1 751.2 753.0 753.1 753.2,6 755.0 755.1 755.2-4 756.0 756.7 Common congenital anomalies Anomalies cong6nitales courantes Microcephaly - Microcephalie Congenital hydrocephaly - Hydrocephalie congenitale Common truncus - Trone arterial commun Transposition of great vessels - Transposition des gros vais- seaux Teratology of Fallot - Tetralogie de Fallot Atrial septal defect, type II - Communication interauriculaire. type II Aortic stenosis - Stenose aortique Patent ductus arteriosus - Persistance du canal arterial Coarctation of aorta - Coarctation de l'isthme aortique Pulmonary stenosis - Stenose de l'artere pulmonaire Cleft palate/without Robin - Fissure du palais/sans syndrome de Robin Oesophageal atresia/stenosis ± TE fistula - Atresie de l'c.esophage/stenose ± fistula tracheo-c.esophagienne Atresia/stenosis of small intestine - Atresie/stenose de l'in- testin grele Atresia/stenosis of rectum and anal canal - Atresie/stenose du rectum et du canal anal Renal agenesis/dysgenesis - Agenesie/dysgenesie du rein Cystic kidney disease. type I - Maladie polykystique du rein, type I Obstructive defect of urinary system - Anomalies obstructi- ves de 1 ·uretere Polydactyly - Polydactylie Syndactyly - Syndactylie Reduction anomalies of limbs - Raccourcissement des mem- bres Anomalies of diaphragm - Anomalies du diaphragme Exomphalos/gastroschisis - Exomphale/gastroschisis True binh Recorded binh prevalence per prevalence 1 OOO total per 1 OOO binhs total binhs Prevalence Prevalence reelle a la enregistree a naissance la naissance pour 1 OOO pour 1 OOO naissances naissances 2.0 7.6 1.7 2.9 3.6 7.4 5.0 7.4 3.2 2.9 4.2 1.8 1.6 1.8 2.3 1.8 1.1 3.0 2.6 3.8 1.6 2.0 1.3 6.0 0.4 1.2 1.2 2.9 1.3 2.2 0.6 1.1 3.6 1.8 1.5 2.0 0.8 0.9 0.8 3.8 3.2 3.8 2.2 2.2 Completeness of notification % Notifications completes % 65 79 24 41 33 39 11 30 19 38 86 100 94 111 35 50 73 100 138 110 Total years lost Nombre total d'annees perdues 75 362 60 62 34 43 143 48 85 23 12 52 69 49 71 56 28 0 0 0 36 18 Total years of actually impaired life Nombre total d'annees de vie avec un handicap 49 9 45 141 131 0 82 0 0 76 83 22 0 0 21 13 34 0 0 78 0 22 Rapp. trimest. statist. sanit. mond .• 41 (1988) - 221 - extremely important from the pathogenetic point of view (5) because, in general, isolated CAs (e.g. cleft lip) have a more or less homogeneous origin and probably differ from anomalies that may be part of multiple entities of very heterogeneous origin (e.g. the case of cleft lip as part of trisomy 13 and several other anomalies of monogenic or environmental etiology, such as EEC syndrome (Electrodactyly - Ectodermal dyplasia - Clefting) and fetal hydantoin syndromes (6). (iii) Minor anomalies, such as morphogenetic informa- tive variants (7), and functional anomalies, such as mental retardation, must be excluded because of incomplete information and selective notification. (iv) The registry-diagnosis method is used in the case of 22 well-known congenital anomaly entities on the basis of the notified component (6). It means that some congenital anomalies and CA-association syn- dromes are identified on the basis of predetermined criteria in the HCMR. For example, congenital cataract and congenital cardiovascular malforma- tions, mainly patent ductus arteriosus with or with- out microcephaly but without other major CAs, are diagnosed as congenital rubella syndrome, and cases with three or more VACTERL-typed CAs with- out other major CAs are registered as VACTERL- association. (v) Data are centrally coded according to the Interna- tional Classification of Diseases (ICD) with certain modifications. Since ICD codes refer to isolated or single anomalies, multiple ones are evaluated sepa- rately under a modified version of ICD rubric 759. • The unit of the HCMR is the affected individual. An index patient with two or more CAs is considered a multimalformed person. Component CAs are coded and recorded. Thus, double registration (due to the differ- ence between the number of anomalies and of mal- formed persons) can be excluded. • The deadline for receipt of data for a calendar year is 31 December of the following year. The annual report, pub- lished in May of each year, contains the following infor- mation : the birth prevalence of each anomaly, classified according to ICD-9 four-digit codes; the sex ratio; the outcome (still- and live births, and infant death rates); the proportion of singleton and multiple births; maternal age; birthweight and gestational age; the monthly and geographical distribution (20 administrative units: 19 counties and Budapest, the capital) ; as well as the com- pleteness of notifications of 45 CA groups. However, a preliminary evaluation of data is made by 31 March of the following year and also published in the annual report. The difference between the birth prevalence in these two data sets is about 5-20%. • Other programmes based on the HCMR provide op- portunities to check, revise, complete, and update the data base. The recent annual total birth prevalences of CAs regis- tered in Hungary have exceeded 4 7 per 1 OOO total births. It is estimated that about 60 of all births per 1 OOO exhibit diagnosed congenital anomalies (1). However, the birth prevalences of severe and lethal CAs are about 23 and 5 per 1 OOO total births, respectively, and these cases are nearly all ascertained. The functioning of the HCMR is continuously checked to ensure that the following seven criteria for good regis- tries are fulfilled : d VACTERL - an acronym for Vertebral, Anal, Cardiac, Tracheal, Esophagal, Renal and Limb; used to designate a pattern of congenital anomalies. Wld hlth statist. quart., 41 ( 1988) 1 . Validity of recorded diagnoses. The multiplicity of sources of notifications considerably enhances the accu- racy of notified diagnoses. Additionally, the diagnoses are continuously checked through the Case-Control Sur- veillance System, the nationwide follow-up of multimal- formed babies, the Population Surveillance of Indicator CAs, and ad hocepidemiological studies of common (8) and moderately frequent CAs performed by the Epide- miology Unit. These parallel activities help to limit the well-known dual effect of size which increases statistical power and decreases accuracy. The proportion of mis- diagnosis for different types of CAs ranges from 0% for cleft lip to 22% for ventricular septal defect. 2. Completeness of ascertainment. As a result of ad hoe epidemiological studies and other programmes the ap- proximate true birth prevalences of common (Table 1), moderately frequent (Table 2), and of some rare CAs are known. Furthermore the annual evaluation of data from 20 administrative units offers another opportunity to estimate true birth prevalences, because the major cause of lower rates is incomplete notification. It should be noted that the completeness of notification also depends on the type of anomaly, e.g. it is about100% for cleft lip but only 29% for congenital inguinal hernia. 3. Time factor. There is an inverse correlation between promptness of notification, accuracy of diagnoses and completeness of notification. As a part of the surveil- lance function of the Registry (see below) 45 CA groups are evaluated preliminarily on a quarterly basis but all CA entities are later re-evaluated in detail. Through the Case-Control Surveillance System, several families routinely inform the Centre about changes in the index patients' diagnosis, address, etc. In addition the Central Statistical Office provides the HCMR with infor- mation on all deaths due to CAs. The data from ad hoe epidemiological studies and other programmes are also attached to each patient's file. 4. A pathogenetica/ly-oriented classification. A localiza- tion-oriented anatomic classification was used previous- ly, but an etiologically-oriented nosological classification would be ideal since information on causes is the most useful. However, as the cause is still unknown for several CA entities, a reasonable compromise is a pathogeneti- cally-oriented classification (9). The HCCAC follows this approach (Fig. 1). 5. Continuous evaluation of confounding factors. Any change of pregnancy outcomes other than CAs can con- siderably modify the birth prevalences of some types of CAs. These confounding factors include induced abor- tions (mainly therapeutic), fetal deaths (spontaneous abortion and stillbirths), as well as low and very low birthweight. Thus, it is necessary to continuously evalu- ate pregnancy outcomes. In addition, some demogra- phic variables (e.g. maternal age or differential fertility with respect to socioeconomic status) are also taken into consideration in evaluating CAs. 6. Confidentiality of personal data. The issue of confi- dentiality of records related to reproduction and birth defects is of growing significance and public concern in many parts of the world. In Hungary, data on index patients, without personal identification, are available only to experts. Personal identification numbers, which have been used in this country since 1983, are of great help for record linkage. Additionally, a written informed consent is requested from the parents of index patients and it is obtained in 98% of cases; the names and addresses of the remainder are deleted from the record. - 222 - FIG. 1. EXAMPLES OF CLASSIFICATION SCHEMES FOR CONGENITAL ANOMALIES• EXEMPLES DES SYST~MES DE CLASSIFICATION DES ANOMALIES CONGENITALES• Severity Gravite Lethal CA (e.g., anencephaly) AC mortelle (anencephalie) (8%-5%) Major AC (e.g. omphalocele) AC majeure (omphalocele) (30%-18%) Mild CA (e.g., undescended testicle) AC legere (ectopie testiculaire) (62%-37%1 Minor anomaly or normal variant (e.g .. low-seat ear) Anomalie mineure ou variante de la normalite (oreille implantee bas) Pathogenesis Pathogenie Congenital abnormalities Anomalies congenitales Birth prevalence per 1 OOO births Prevalence ii la na1ssance pour 1 OOO naissances Deformation (pied bot) Dysplasia (e.g., osteogenesis imperfecta) Dysplasie (periostale) Isolated - Anomalies isolees (90%) Tres rares (:$001/5%) Multiple - Anomalies multiples (10%) Single Sequence (e.g. ventricular (e.g., diaphragmatic Complex or mono- topic field defect (e.g., tetralogy Polytopic field defect (e.g., holoprosencephaly) CA-syndrome (e.g., Down) CA-association (e.g., VACTEAL) Associations malformatives (V ACTEAL) 2% Random combination (e.g .. cleft lip and septal defect) defect and Simples its consequences) of Fallot) Anomalies polytopiques (holoprosencephalie) Syndromes malformatifs (Down) 4% anal atres1a) Associations al8atoires (bec-de-lievre et atresie du canal anal) (communication S~quences interventriculaireJ (anomalies 80% du diaphragme et ses cons4quences) 40% Anomalies complexes ou monotopiques 1% des champs morphogenetiques (tetralogie de Fallo!) 5% Manifestation a Mainly congenital abnormalities. i.e. a developmental defect, gross or microscopic, present at birth whether detected at that time or not. Percentage figures indicate the proponion of special groups within the given classification, while binh prevalences are expressed per 1 OOO total binhs. - II s'agit essentiellement d'anomalies cong6nitales, c·est-a-dire d·anomalies du developpement, flagrantes ou invisibles. presentes a la naissance. qu'elles aient alors 6te decelees ou non. Les pourcentages indiquent la proponion de groupes sp6ciaux dans la classification donn6e, tandis que la prevalence a la naissance est exprimM pour 1 OOO naissances. 7. Record linkage. The potential benefits of parallel evaluation of different registries have not yet been exploited. At present, only sib-occurrence is detected by this method on the basis of the maternal data. Multiple births to women in and around Budapest have been recorded by the Budapest Twin Registry since 1 January 1970 (10). Obstetric institutions are required to notify multiple deliveries and the completeness of noti- fications exceeds 90%. Placentas are collected and ex- amined by the pathologist. In like-sexed twins with dichorial placenta, zygosity is determined between ages 3 and 6 years by examining blood and serum protein groups. The zygosity of twins born between 1970 and 1979 was determined in 78.7% of cases. Congenital structural talipes equinovarus, congenital inguinal hernia, hypospadias and undescended testicles are significantly more frequent in twins while congenital hypertrophic pyloric stenosis is significantly less frequent. There is also a voluntary Hungarian Twin Registry involv- ing adult twins, and several scientific studies (e.g. inher- itability of lactose absorption, methane excretion, pep- sinogen, psychosexual development) have been carried out on the basis of the data collected by this institu- tion (11). Surveillance programmes Recently a distinction has been made between the terms surveillance and monitoring (12). Monitoring means to study a population at risk, i.e. exposed to known or suspected environmental factors (mutagens, terstro- gens, carcinogens), while surveillance means to study a population at large to determine the baseline occurrence of diseases and detect changes in it. The HCMR has a national surveillance function. The preliminary quarterly analysis of 45 CA groups based on Rapp. trimest. statist. sanit. mond., 41 (1988) - 223 - notifications received up to the 90th day after birth aims to detect any time and/or space clusters as quickly as possible. In Hungary the two most important findings were a temporary cluster of congenital limb-reduction abnormalities between 1975 and 1978 (13) and a signifi- cant increase in the birth prevalence of hypospadias from 1978 (14). The annual reports contain the main practical conclusions of the surveillance, i.e. changes and clusters in the birth prevalences of different CA groups. Special surveillance systems At present there are three special surveillance systems within the HCCAC based on the HCMR. 1 . The Hungarian Surveillance System of Selected Con- genital Anomalies within the International Clearinghouse for Birth Defects Monitoring Systems. The Clearing- house was established in 1974 (15, 16) with Hungary as one of the 10 founding members. At present the Clear- inghouse comprises 23 systems in 27 countries. Eleven selected CAs and MCA cases are notified within 90 days of birth to the coordinator who collates the data and information which are published quarterly. An annual report with revised data and an evaluation of some CA groups has also been published since 1980. The main purpose of this system is to recognize case-clusters as soon as possible and to identify causes, mainly terato- gens, through case-control epidemiological studies. 2. The Case-Control Surveillance System of Congenital Anomalies attempts to obtain etiological information immediately after the notification of anomalies to the HCMR in order to identify the cause (e.g. teratogens) and assist in prevention. Mild anomalies, such as Ortolani click, inguinal hernia, nevus, and haemangiomas, as well as syndromes of known origin, except for Down's syn- drome, are excluded. This surveillance system was established following a feasibility study in 1979 and has been operating since 1980. Three negative controls (i.e. newborns without anoma- lies) are matched to every index patient according to birth week, sex and district of residence of the parents. The parents of control infants receive the same ques- tionnaire and lists of drugs and diseases, but the text of the explanatory letter is different. The prenatal care log- book and other medical documents are also requested. The response rate is 67%; however, in general, at least one of the three control cases is available for compara- tive evaluation. Index patients, such as those with Down's syndrome, are evaluated by the same method and are considered as positive control cases. In the evaluation, the rates of the total control group are compared with the rates for the 25 CA groups according to the categories of drugs, maternal diseases, and occu- pational exposures; the X2 test is used for statistical evaluation. The rates of each CA group are then com- pared with those of its matched control cases; the McNemar test is used for the statistical evaluation. The final analysis is limited to significant differences and each drug within a category is evaluated separately according to gestational age. So far diazepam, chlordiazepoxide and nitrazepam; allylestrenol ; Bactrim and nitrofuran- toin; reserpin and bromcriptine have been evaluated from the teratogenic point of view. Chemical and other occupational exposures are evaluated separately (17). This case-control surveillance system has some useful supplementary functions as well: (i) to check (and cor- rect, if necessary) ; (ii) to complete the data of the HCMR Wld hlth statist. quan., 41 (1988) (paternal age, birth order, outcome of previous pregnan- cies, marital status, family history); (iii) to get written consent for further registration and action; (iv) to obtain the name and address of the child's paediatrician; (v) to inform parents about possible causes and pathogenesis, time schedule of medical treatment and possible pre- vention of recurrence by means of a special booklet. In the past, parents generally received insufficient explana- tions from medical personnel. Data obtained through the supplementary functions of the Case-Control Surveillance System are also evalu- ated. Thus the sib-occurrence of CAs detected from family history is particularly interesting in multimal- formed babies because it might help to identify the same CA entity. If the nosological diagnosis requires hospital- ization, a special department of the inpatient paediatric clinic is available in Budapest. The sib-occurrence of un- identified MCAs is about 13-15%, and involves three symmetrical proportions: concordant, half-concordant (i.e. only one component anomaly of multimalformed index patients occurs in the sib), discordant (probably random event) sib-occurrences (6). 3. The Mutation Surveillance System is based on the "indicator conditions" of offspring. Indicator conditions in an individual are (or may be) distant phenotypic man- ifestations of altered DNA in the germ cells of the par- ents. The surveillance of germinal mutations in a popu- lation at large through the help of indicator conditions is an attempt to measure the genetic load and any rise in the rate of new germinal mutations. Three types of indi- cator conditions are followed-up within this surveillance system. Sentinel phenotypes Sentinel phenotypes are indicators of germinal genie dominant mutations. They are clinical disorders that occur sporadically as a consequence of a single, highly penetrant mutant gene; they are autosomal, dominant or X-linked traits of considerable frequency and low fit- ness, and are uniformly expressed and accurately diag- nosable with a minimal clinical effort, but a relatively high probability of ascertainment (18). The advantages of using sentinel phenotypes are considerable: (i) the af- fected persons will enter the health-care systems in a developed country and will most certainly be registered in a recording system; (ii) parents of affected children are motivated to cooperate in additional investigations; (iii) when a sentinel phenotype is recognized in the absence of affected parents, the suspicion of a new ger- minal mutation is strong. According to the method of ascertainment, the following three different groups of sentinel phenotypes are distin- guished. Thirteen sentinel anomalies are easily observed at birth, and thus there is a high expected notification. These sentinel anomalies seem to be reasonably satis- factory for study in the Hungarian programme (guidelines on these anomalies including a short description and photos of each were sent to all obstetric and neonatal inpatient clinics). Two sentinel childhood tumours, reti- noblastoma and Wilms' tumour, occur in a well-defined age group and are treated (thus recorded) in centralized medical institutions. The record of the HCMR is com- pleted by the data from the Hungarian Registry of Child- hood Tumours and the Hungarian Institute of Ophthal- mology. Genetic disorders with delayed onset which are readily diagnosed and recorded in various types of genetic disease registries are also sentinel phenotypes. An attempt has been made recently to establish some specific genetic disease registries for patients affected by haemophilia A and B, Huntington's chorea, neurofi- bromatosis and intestinal polyposis. - 224 - The evaluation of 13 sentinel anomalies and two child- hood tumours together as sentinel anomalies consists of the following three steps: The parents of indexed patients are invited to bring the affected child and all other siblings to the Centre. Parents that cooperate and their children are then examined and documented at the Centre, with the help of consultant specialists, if necessary, to con- firm (or exclude) the nosological diagnosis. The other families are visited at home. In the case of deceased index patients, detailed autopsy records are evalu- ated. Family history, including data on the four grand- parents (to decide whether the case is sporadic or familial), environmental history and personal data (e.g. parental age) are obtained by personal interview and recorded by means of a questionnaire. The envi- ronmental histories of familial cases are used as con- trols. If the parents plan to have other children, genetic counselling is offered. The data thus obtained are evaluated annually. An average of 46 possible new mutations per year were found over the five-year study period (1980-1985), i.e. about 35 new mutations per 100 OOO live births (19). Additionally, this programme has highlighted some ano- malies of autosomal origin and has resulted in changes in medical services for treating these defects. Down's syndrome Down's syndrome is an indicator of germinal numerical chromosomal mutations. There are several arguments in favour of Down's syndrome as an indicator condition. Among germinal chromosomal mutations, a much higher proportion is numerical rather than structural (5 : 1 at birth and 30: 1 among feta I deaths). Down· s syndrome is the only common anomaly (1 per 1 OOO births) caused by a chromosome aberration. Two-thirds of patients with Down· s syndrome are liveborn and have a relatively good chance of survival. In general, Down's syndrome is recognizable by its clinical symptoms. Karyotyping pro- vides a reliable confirmation. The validity of the diagno- sis was found to be about 94% in the HCMR. From the standpoint of mutation surveillance, the most important characteristic of Down's syndrome is that it nearly always (98%) occurs as a de novo event. Prenatally diagnosed and selectively terminated fetuses affected by Down's syndrome are recorded in the HCMR and provide a useful complementary source of data within this population-based system. So far, surveillance of Down's syndrome has yielded the following information (20). Firstly. over the past 15 years the true birth prevalence of this syndrome has been in the range of 1.2-1.4 per 1 OOO total births. Secondly, a cyto- genetic analysis was performed on only 23% of the fetuses of mothers aged 40. Thirdly, a slight reducing effect of prenatal diagnosis has only been demonstrated in recent years. Unidentified multiple congenital anomalies Unidentified multiple congenital anomalies (MCAs) are useful as indicators of germinal dominant genie and chro- mosomal mutations. MCAs can be considered to be the most sensitive indicators of environmental factors, i.e. teratogens and mutagens. It is a reasonable postulate that new environmental factors can cause new, undeli- neated MCA entities. Thus, a sensitive surveillance mechanism for the detection of unidentified MCAs is of vital importance. For evaluation purposes, all component CAs within un- identified MCAs involving three, four, or more compo- nents are reduced to pairs, and combined with MCAs involving only two component CAs (21 ). Thus, a pair is a set of two independent component CAs in unidentified MCA cases involving two or more components. The baseline occurrences of CA pairs studied per 100000 total births between 1973 and 1982 ranges from Oto 93, presumably because this is determined by birth prevalence and the different ascertainment of com- ponent anomalies in addition to the liability of each com- bination. Some annual rates of congenital anomaly pairs exceed their 10-year averages considerably but these devia- tions seem to be relatively rare: 4. 1 % of pairs studied at the Centre and 0.4% of those in the 10-year data base. The crucial point is to separate the biologically important clusters from the statistically significant changes. After significant increase is detected, it is possible to deter- mine whether the pair is a component of a given combi- nation of CAs or whether this increase reflects a general trend. The former may indicate a causal effect. while the latter may demonstrate a change in diagnosis and noti- fication. The annual occurrence of pairs within unidentified MCA cases born after 1982 is compared with the baseline rates of CA pairs studied. The surveillance of component CA pairs within unidentified MCA cases may help detect any cluster of CA pairs due to environmental factors. Diagnostic, preventive and social rehabilitation functions The HCCAC fulfils its diagnostic function mainly through specific programmes, such as the Case-Control Surveil- lance System of Congenital Anomalies and the Mutation Surveillance of Sentinel Anomalies. However, the na- tionwide follow-up investigation of multiple anomalies merits a short description (22). The programme has a threefold purpose: (i) to increase the rate of identification of CA syndromes and associa- tions; (ii) to identify additional recognizable CA patterns because their recognition and identification can be useful for estimating prognosis and sib-occurrence; (iii) to pro- vide a special surveillance mechanism for multiple ano- malies since these are the most sensitive indicators of teratogenic and mutagenic factors. Owing to the pleio- tropic effect of mutant genes and the generalized impact of microscopically visible chromosomal aberrations, the majority of genie and autosomal chromosomal muta- tions cause congenital anomaly syndromes, i.e. multiple anomalies. In man, nearly all known teratogens (rubella, alcohol, hydantoin, etc.) also produce multiple anoma- lies. The follow-up investigation of multiple anomalies in- volves the following procedure: • Clinically diagnosed MCA entities are registered on the basis of notification (29.1 %). • New or supplementary information is requested in the case of unspecified MCAs (4.5%); as a result of this, their proportion has been reduced to 2.6%. • On the basis of component CAs, registry-diagnosis is attempted for 22 well-defined CA entities ( 12.8%); apparent but not true instances of MCAs have been excluded (5. 7%). • Surviving infants with unidentified multimalformations are referred to one of eight regional multiple congenital abnormality examination centres established within Rapp. trimest. statist. sanit. mond .. 41 (1988) - 225 - this programme, to be examined by specialists. Each of these centres is staffed with paediatricians inter- ested in syndromology and is equipped with labora- tory facilities for chromosome analysis and other tests; 2.4% of multiple anomalies have been identified and 0. 7% of cases excluded. • The remaining unidentified MCAs (47.3%) involving two or three components are listed annually according to a hierarchical order of 45 anomaly groups in a single table. Cases with four or more anomalies are listed separately. Results The main results of this population-based programme involving 7 049 notified multimalformed cases born be- tween 1973 and 1983 are as follows: • The final count of cases with MCAs was 6 643; this is equivalent to a birth prevalence of 4.0 per 1 OOO total births and to 10% of CA cases recorded during the study period. • The proportion of identified multiple anomaly entities increased from 29% to 4 7 % , and 54 % of the expected number of unbalanced chromosome aberrations was ascertained. • Some new MCA entities were identified, e.g. schisis- association (23). • The tabulation of the component CAs within uniden- tified MCAs reveals clusters of specific combinations, e.g. the association of congenital cardiovascular mal- formations and cataract increased significantly 6-8 months after a large rubella epidemic (Fig. 2). Prevention programmes A national prevention programme established in the late 1970s was initially based on the data collected by the HCMR within the framework of its prenatal screening activities, namely amniocentesis and amniotic alfa-feto- protein determination for women whose previous preg- nancies resulted in babies with anencephaly or spina bifida (24). However, since maternal serum alfa-fetopro- tein (MS-AFP) screening techniques have been devel- oped, amniotic examinations have lost their importance. The ongoing vitamin study seeks to reduce the recur- rence of neural tube defects through periconceptional multivitamin supplementation. In 1987, the number of participants exceeded 500 because information was mailed to all Hungarian women recorded as having had a baby affected by a neural tube defect. As a WHO Collaborating Centre, the Centre is mandated to develop approaches for evaluating the relevance, effect and impact of the hereditary diseases control pro- gramme. In addition, there are four prevention programmes con- cerned with genetically-determined disorders. (i) Maternal serum AFP (M-AFP) is screened at week 16 of gestation and completed by ultrasound examination; if necessary, after repeated positive MS-AFP values with negative ultrasound scanning, an amniotic AFP exami- nation is suggested. This programme became popu- lation-based in 1985, but the efficacy is low. In 1984- 1985, for example, 331 cases with isolated neural tube defects were recorded in the HCMR but the data neces- sary for further analysis were available in 279 of the cases. Only 79 were prenatally diagnosed before week 22 of pregnancy and 74 of these were terminated. The remaining 200 births were examined either by MS-AFP or by ultrasound, or by a combination of both. MS-AFP was positive in 94 cases out of 188 (50%) while ultra- sound demonstrated a neural tube defect in 22 cases out of 161 . The lack of selective abortion was explained by a discrepancy between the results of the two exami- nations, by mothers refusing to terminate their pregnan- cies or by the fact that examinations were performed too late. (ii) Prenatal chromosome examination in fetuses of mothers over 40 is offered, but only 23.4% of eligible pregnant women were examined in 1984-1985 and 16 numerical chromosome aberrations detected (in 2.6% of all fetal chromosome analyses). This is far from the desired goal. FIG. 2. INCREASED NUMBER OF CASES WITH CONGENITAL RUBELLA SYNDROME ASCERTAINED BY THE REGISTRY-DIAGNOSIS METHOD BASED ON THE COMBINATION OF CATARACT AND HEART DEFECTS AFTER A RUBELLA EPIDEMIC, HUNGARY, 1974 AUGMENTATION DU NOMBRE DES CAS ATTEINTS DU SYNDROME DE RUBIEOLE CONGENITALE ET CONFIRMES PAR LA METHODE DU DIAGNOSTIC FONDE SUR LE REGISTRE A. PARTIR DE L'ASSOCIATION CATARACTE-MALFORMATIONS CARDIAQUES APRES UNE IEPIDIEMIE DE RUBIEOLE, HONGRIE, 1974 Rubella epidemic - Rubeole epidemique 50 40 "' ! ~ 30 "' .., c: 20 .. "' :, 0 .t: ... 10 0 ........ ........ Wld hlth statist. quan., 41 (1988) 1974 Congenital rubella syndrome /", Syndrome de ruMole cong<§nitale , ' , ' , ' , ', , \ , \ , \ , \ Ouaners - Trimestres \ \ \ \ 1975 10 8 6 4 2 0 - 226 - (iii) Screening of newborns for phenylketonuria (PKU) became a population-based programme in 1973. Screening for galactosemia and hypothyroidism were added later. These components of the programme seem to work well and the recorded birth prevalence of PKU is 0.9 per 1 OOO. (iv) Orthopedic screening of newborns has resulted in the almost total elimination of Hungary's most common CA, dislocation of the hip. As a matter of fact, intensive screening has resulted in overdiagnosis, which has greatly increased the notified birth prevalence of this anomaly (28 per 1 OOO compared with a prevalence of manifest cases of about 10 per 1 OOO in the 1940s). At present, an effort is being made to improve the reliability of diagnosis. Rehabilitation programme The social rehabilitation programme involves surviving children with major CAs registered in the data base of the HCMR; this concerned about 110 OOO children in 1988. This programme offers a range of services for school- age children with major CAs: (i) a medical examination to provide supplementary treatment if necessary (e.g. bet- ter prothesis in children with limb reduction, or additional logopaedic education for those with oral clefts) ; (ii) psy- chological counselling to overcome a number of prob- lems, including inferiority complex; (iii) employment counselling ; (iv) special counselling for family planning; (v) the organization of special clubs or societies where these individuals can have independent but joint social activity and can help one another. In addition, the Hungarian Twin Society and the Twin Club, under the direction of the HCCAC, promote social activities for twins. Besides the activities outlined above, the HCCAC has helped to launch a number of projects and one of the potential benefits to be derived from this research would be a reduction in the occurrence of CAs. However, the main purpose of the Hungarian Centre for Congenital Anomaly Control is to promote and contribute to the improvement of medical services and to an increase in the efficacy of existing preventive measures. SUMMARY The Hungarian Centre for Congenital Anomaly Control manages a number of surveillance programmes based on the Hungarian Congenital Malformation Registry. No- tification of congenital anomalies, which is compulsory. originates exclusively from physicians working in va- rious health institutions and is based on their diagnosis of malformed patients-newborns and infants. In recent years, the total birth prevalence of registered congenital anomalies has exceeded 4 7 per 1 OOO total births. The notified data are critically evaluated and centrally coded according to the unit of notification, the affected individ- ual. Diagnostic accuracy, completeness of notifications and the effect of confounding variables are continuously checked. The purpose of the Case-Control Surveillance System is to obtain etiological information concerning drug consumption, maternal diseases and occupational hazards during pregnancy. The surveillance of mutations through indicator conditions (sentinel anomalies. Down's syndrome and pairs of components of unidenti- fied multiple congenital anomalies) is an attempt to measure the rate and trend of new germinal mutations. The nationwide follow-up of multimalformed infants helps to increase the proportion of identified congenital anomaly entities, to identify new ones, and to detect clusters caused by new environmental teratogenic or mutagenic factors. RESUME Les activites du centre hongrois de lutte contre les anomalies congenitales Le centre hongrois de lutte contre les anomalies conge- nitales gere un certain nombre de programmes de sur- veillance fondes sur le Registre hongrois des malforma- tions congenitales. La notification des anomalies conge- nitales. qui est obligatoire, est du ressort exclusif des medecins attaches aux divers etablissements de sante et repose sur les malformations qu'ils diagnostiquent chez les nouveau-nes et les nourrissons. Ces dernieres annees, la prevalence totale des anomalies congenitales enregistrees a la naissance a depasse 4 7 pour 1 OOO naissances. Les donnees notifiees font I' objet d'une eva- luation critique et sont codees au niveau central selon l'unite de notification, a savoir le sujet affecte. Un controle permanent sert a assurer que le diagnostic est exact et que les notifications sont completes et a verifier l'effet des facteurs d'erreur. Le systeme de surveillance cas-temoins a pour but d'obtenir des renseignements etiologiques sur les medicaments absorbes, les mala- dies de la mere et les risques professionnels pendant la grossesse. La surveillance des mutations par les mala- dies indicatrices (anomalies sentinelles, syndrome de Down et paires d'anomalies composantes d'anomalies congenitales multiples non identifiees) vise a mesurer la frequence des mutations germinales nouvelles et leur evolution. Le suivi a l'echelle du pays des nourrissons atteints de malformations multiples aide a accroitre la proportion des anomalies congenitales identifiees, a identifier certaines anomalies nouvelles et a depister les anomalies groupees dues a de nouveaux facteurs envi- ronnementaux teratogenes et mutagenes. Rapp. trimest. statist. sanit. mond., 41 (1988) - 227 - REFERENCES - REFERENCES 1. CZEIZEI, A. ET AL. The load of genetic and partially genetic disorders in man. I. Congenital anomalies: estimates of detriment in terms of years of life lost and years of impaired life. Mutation research, 128: 73-103 (1984). 2. MILLER, J. R., Some uses of genetic registers. In: Emery, A.E.H. & Miller, J. R. (eds), Registers for the detection and prevention of genetic disease. New York, Stratton International Medical Book Corpora- tion, 1976. 3. Boo, M. ET AL. Congenital malformation surveillance. Teratology, 24: 277-283 (1981). 4. ROSE, G. ET AL. Epidemiology for the uninitiated, London, British Medical Association, 1986. (Sec- ond edition). 5. CZEIZEL, A. Incidence and prevalence as measures of the frequency of birth defects. American journal of epidemiology, 119 : 14 1-142 ( 1984). 6. CZEIZEL, A. ET AL. Multiple congenital abnormality. Budapest, Akademiai Kiad6, 1988. 7. MEHES, K. Minor malformations in the neonate. Budapest, Akademiai Kiad6, 1983. 8. CZEIZEL, A. ET AL. Aetiological studies of isolated common congenital abnormalities in Hungary. Budapest, Akademiai Kiad6, 1984. 9. SPRANGER, J. ET AL. Errors of morphogenesis: concepts and terms. Journal of pediatrics, 100: 160-165 (1982). 10. CZEIZEL, A. ET AL. The Twin Register of Budapest. Acta geneticae medicae et gemellologiae, 22: 219- 223 (1972). 11. METNEKI, J. ET AL. A study of lactose absorption capacity in twins. Human genetics, 67: 296-300 (1984). 12. WORLD HEALTH ORGANIZATION. Guidelines for the study of genetic effects in human populations. Geneva, WHO, 1985. (Environmental Health Crite- ria No. 46). Wld hlth statist. quan., 41 (1988) 13. CZEIZEL, A. ET AL. An aetiological evaluation of in- creased occurrence of congenital limb reduction in Hungary, 19 75-1978. International journal of epide- miology, 12 : 445-449 ( 1983). 14. CZEIZEL, A. Increasing trends in congenital malfor- mations of male external genitalia. Lancet, i: 462- 463 (1985). 15. FLYNT, J. W. ET AL. International clearinghouse for birth defects monitoring systems. Contributions to epidemiology and biostatistics, 1: 44 (1979). 16. CASTILLA, E. ET AL. Methodology for birth defects monitoring. Binh defects original anic/e series, 22 (5): 1-43 (1986). 17. CZEIZEL, A. Case-control surveillance system of congenital anomalies in relation to occupational ex- posure. In: Sorsa, M. & Norppa, H. (eds), Monitoring of occupational genetoxicants. New York, Alan R. Liss. 1986. 18. MULVIHILL, J. J. ET AL. Perspectives in mutation epi- demiology - 6. A 1983 view of sentinel phenoty- pes. Mutation research, 123: 345-361 (1983). 19. CZEIZEL, A. ET AL. Mutation surveillance of sentinel anomalies in Hungary, 1980-1984. Mutation re- search, 186: 73-79 (1987). 20. CZEIZEL, A. Some epidemiological characteristics of Down's syndrome in Hungary. Acta morphologica hungarica: in press (1988). 21. K1svARGA, A. ET AL. A pairwise evaluation of compo- nent congenital abnormalities in unidentified multi- ple congenital abnormalities. Mutation research: in press (1988). 22. CZEIZEL, A. ET AL. A nationwide evaluation of multiple congenital abnormalities in Hungary. Genetic epide- miology: in press (1988). 23. CZEIZEL, A. Schisis-association. American journal of medical genetics, 10: 25-35 ( 1981). 24. CZEIZEL, A. ET AL. National programme for prevention of the recurrence of neural tube defects. Acta pae- diatrica hungarica, 21: 315-319 (1979).