Methodological Approaches to the Assessment of Risk of Congenital Anomaly Due to Environment Pollution
Budapest, Hungary, 6-7 March 2007
The European Context
EUROCAT and Environmental Pollution [Presentation]
Prevention of Birth Defects: A View from WHO Geneva [Abstract] [Presentation]
Child Health and the Environment: The View from WHO Europe [Presentation]
The WHO European Environmental Health Indicators Project
The European Regulatory Framework for Teratogens [Abstract] [Presentation]
Rare Diseases and Environmental Exposures - A Review of Research and Research Needs in Europe [Abstract]
European Biomonitoring [Abstract] [Presentation]
Placental Uptake and Transfer of Environmental Chemicals Relating to Allergy in Childhood Years EU FPV QoL
KA4 (Health & Environment) [Abstract] [Presentation]
The Flanders Environment and Health Study: Biomarkers and Biomonitoring [Abstract] [Presentation]
Exposure Assessment in Environmental Epidemiological Studies Relating to Pregnancy Outcome [Abstract]
Water Cholorination - Exposure Assessment for Epidemiologic Studies [Presentation]
Identifying Chemical Exposures During Pregnancy in Human Hair [Abstract] [Presentation]
The Dioxin Incident in Belgium: Exposure Assessment and the Epidemiological Response [Presentation]
The Use of Biological Monitoring for Assessment of Exposure to Heavy Metals Released from Hazardous Waste
Landfill Sites in Poland [Abstract] [Presentation]
Epidemiological Methods for the Investigation of Environmental Exposures in Relation to Congenital Anomalies
Anomaly Risk Versus Geographical Variation in Exposures: Two Examples [Abstract] [Presentation]
Causal Diagrams and Confounding [Abstract] [Presentation]
Researching Environmental Pollution and Congenital Anomalies in Hungary [Presentation]
High Birth Prevalence of Hypospadias and Other Congenital Anomalies in Two Petrochemical Areas of Sicily, Italy
Pregnancy Outcome and Environmental Pollution: A Case Study [Presentation]
Contaminated Land in Helsinki [Presentation]
Welsh Landfill and Gastroschisis Study [Abstract] [Presentation]
Exposure to a Solid-Waste Disposal Site in Cali, Columbia and Congenital Malformations [Abstract]
Current Research in Lithuania: Exposure Assessment in Kaunas Newborns Cohort Study [Abstract]
Current Research on Environmental Pollution and Congenital Anomalies in Belarus [Abstract] [Presentation]
The Detection and Investigation of Clusters of Congenital Anomalies in Time and/or Space
The Current EUROCAT Strategy and Potential Future Directions [Presentation]
Scan Statistics [Abstract] [Presentation]
The EDMP Software for Detection of Time Clusters - User Friendly Software for Congenital Anomaly Registers
Two Years Experience of Routine Cluster Detection and Investigation by EUROCAT Registers [Abstract]
Use of Satscan to Detect Spatial Clusters in Five Regions of Britain [Abstract] [Presentation]
Prevention of Birth Defects: A View from WHO, Geneva
WHO provides international expertise and technical advice to countries in their efforts to establish and manage national programmes of major health importance. A reduction in the mortality and morbidity caused by communicable diseases and malnutrition has changed the disease spectrum across a number of countries. As a result congenital and genetic disorders are increasingly being recognized as significant health burdens. Taking this into account, WHO has supported activities in the field of medical and human genetics at the international and country levels. WHO's Human Genetics (HGN) team, through strong networks of international collaborating programmes, centres and experts, collects and shares information on the possible application of genetic knowledge at the community level. Several international studies on the prevention and control of birth defects have been carried out with the goal of developing and implementing those approaches deemed suitable for incorporation into health services at country level. These points have been widely discussed at various WHO meetings and related publications are available on our WHO/HGN Genomic Resource Centre website http://www.who.int/genomics.
In 2006 the World Health Assembly adopted the Resolution (WHA59.20) on Sickle-cell anaemia and the Executive Board also adopted the Resolution (EB118.R1) on thalassaemia and other haemoglobinopathies. Both of Resolutions urge Member States to develop, implement and reinforce comprehensive national, integrated programmed for the prevention and management of haemoglobin disorders. They also urge to develop and strengthen medical genetics services, and to promote community education and training. Thus, WHO has supported and raised the haemoglobinopathies to the high international level. We hope to maximize the benefits of this in favour of other disorders and to promote the involvement of country experts, governments and non-governmental organizations in the future work.
The European Regulatory Framework for Teratogens
The process of toxicity testing of environmental chemicals is ruled by a framework of guidelines (OECD, OPPTS, etc.). The presentation will describe the process from the biological tests for environmental chemicals up to potential labelling.
A more detailed description of this process and some critical aspects in this process is provided by Buschmann (2006). That paper also gives an overview of the existing documents and draft documents relevant for this field and available in the internet. Based on the current situation, the following points are discussed as critical issues: life stage considerations and their implications on testing, usefulness of investigations in juvenile animals, requirements for ADME studies, design and endpoints in fertility studies, use and usefulness of developmental milestones, performance of Special Studies vs. One-Study-Design, considerations on transplacental carcinogenicity and early aging, significance of maternal / parental toxicity, application of triggers to justify studies, inclusion of new endpoints into test guidelines, and test strategies applied. Based on this situation, the usefulness of risk considerations in the current EU classification system for toxicity to reproduction as well as potency considerations are discussed and suggestions are made to improve the basic requirements for chemical testing.
Buschmann, J. Critical Aspects in Reproductive and Developmental Toxicity Testing of Environmental Chemicals. Reproductive Toxicology 22 (2006) 157-163
Rare Diseases and Environmental Exposures - A Review of Research and Research Needs in Europe
Manuel Posada de la Paz
Papers in the 1980s began to discuss a new notion concept of grouping diseases, based, not upon aetiology or on the damaged organ, but rather upon prevalence of the disorder. The US Food & Drug Administration (FDA) identified these pathologies under the umbrella concept of “orphan diseases”, since their low prevalence prevented researchers from having access to research resources, both private and public (1). Similarly, there was a great difficulty in bringing together sufficient numbers of patients and thus being in a position to conduct scientific studies. Furthermore, the pharmaceutical industry failed to invest in the search for new drugs since this was simply not profitable in view of the scant number of patients and their geographical dispersion. This meant that, in tandem with “orphan diseases”, the term “orphan drugs” was widely accepted in September, 1999. At that time, the European Community launched its own Action Plan for Rare Diseases (2). On 27 April 2000, Regulation (EC) No 141/2000 of the European Parliament and of the Council and the implementing Commission Regulation (EC) No 847/2000 entered into force being the first Community-wide regulations on orphan medicines. The EU orphan legislation aims at providing incentives for the research, development and marketing of orphan medical products that the pharmaceutical industry would be unwilling to develop under normal market conditions (3).
The EC defines a rare disease as being any disease or condition of a frequent disease that has a prevalence of less than 5 cases per 10,000 inhabitants in the community and that, in addition, registers a high rate of morbidity-mortality and disability (2).
Orphan medicinal products are intended for the diagnosis, prevention or treatment of life threatening or very serious conditions that are rare. While only a small number of patients suffer from one of the 5,000 to 7,000 orphan conditions that exist today, according to projections, rare diseases affect a total of some 30 million people in the 25 EU Member States.
A primary reason why sound epidemiological data is often lacking, is the absence of proper classification and coding for the disease and the absence of registration of the patients suffering from rare conditions. Although ICD codes are available for some better-know rare diseases, such as thalassaemia, cystic fibrosis and haemophilia, many diseases are not included in medical registries and databases.
The new European programme of Public Health has amplified the activities on rare diseases (5). Projects supporting cooperation between rare diseases organisations, creating networks of action for rare diseases and organizing European Conferences on Rare Diseases are some of the strategies set up by the DG SANCO. At the same time, a Task Force on Rare Diseases has been commissioned for improving classification and codification of rare diseases, piloting reference networks (centres of reference) for rare diseases and making recommendations on rare diseases research (4). On the other side, DG Research has increased the funds devote to rare diseases in both FP6 and FP7, funded networks of excellence as well as several projects modalities.
Even though, it is said that the majority of rare diseases have a genetic origin, at least 25% of more than 7,000 rare diseases are not linked to genes and most of them are in the group of autoimmune diseases and vasculitis. This 25% of diseases account for the highest number of DALYs among all RD, because most of them show higher prevalence figures than mendelian rare diseases. Environment is one of the most important factors for those rare diseases not directly related with genes. Even in genetic diseases, different phenotypes are not fully justified by known genes.
We still know that genes can be expressed or not, transcriptions are not always determined in the same way, metabolism of xenobiotics can be modified by genes and this is the cause that different clinical phenotypes can be shown. In summary, the epigenetic changes play an important role in the knowledge of the diseases and it becomes a major focus of research. Epigenetic are usually restricted to the internal cell environment and the mechanisms that regulate the genetic information but epigenetic also cover the influence of the environment in this regulation, well in a direct way, or through forcing the expression of other genes (5). Finally, low doses of chemical exposures are a major challenge for researchers because they affect to small proportion of population – those more susceptible - and the effects are not always so clear.
Recent toxic epidemics, such as Eosinophilia Myalgia Syndrome and Toxic Oil Syndrome show some examples on how could have been happened if the toxic doses would have been smaller. Probably, some variants of rare diseases such as eosinophilia fascitis, escleroderma, inflammatory neuropathy, interstitial lung diseases, etc would have been form part of the clinical spectrum of these very well known diseases
Other diseases such as Autistic Spectrum Disorders that contain many rare diseases are being now focus on attention of environment and genes interaction research. Finally, other examples such as Acquired Lipodistrophies, Xeroderma Pigmentosum and ultraviolet light, α-1-Antitrysin deficit and smokers, Alveolar Proteinosis and silica dust exposure clearly show how rare diseases are not a simply field of medicine.
In summary, not all of Rare Diseases have a clear genetic origin and the correlation genotype-phenotype is usually poor even in those diseases with the clearest genetic link. At the same time DALYs are higher in not genetic but still rare diseases but the research on rare health outcomes among environmentalist researchers are still scarce.
Rare diseases need social attention but also more critical mass of researchers and researchers should focus on interaction between genes and environment (Do not forget xenobiotic metabolism) because we should pay attention to the influence of chemical exposures to low doses and rare health effects
1. Report of the National Commission on Orphan Diseases. Commission report Part I: Introduction. J. Rare Diseases. Vol II. March/April. 1996: 21-27
2. DECISION No 1295/1999/EC OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL of 29 April 1999
adopting a programme of Community action on rare diseases within the framework for action in the field of public
health (1999 to 2003) http://europa.eu.int/eur-lex/pri/en/oj/dat/1999/l_155/l_15519990622en00010005.pdf
3. Regulation (EC) No 141/2000 of the European Parliament and of the Council of 16 December 1999 on orphan
medicinal products http://eur-lex.europa.eu/LexUriServ/site/es/oj/2000/l_018/l_01820000122es00010005.pdf
4. Points to consider on the calculations and reporting of prevalence of a condition for orphan designation.
European Medicines Agency (EMEA). London. 26.Mar.2002 COMP/436/01.
5. Programme of Community action in the field of public health (2003-2008).
6. Minutes First meeting of the “Task Force on Rare Diseases” (TFRD). Luxembourg, 21 January 2004
7. Hunter, DJ. Gene-Environment Interactions in Human Diseases. Nature Reviews Genetics 2005 (6): 287-298
The European health action plan 2004-2010 has identified the need for developing a coherent approach to human biomonitoring in Europe. HBM measures the presence of pollutants in blood samples, urine or other body fluids (biomarkers of exposure) and the associated biological effects (biomarkers of effect). As such HBM provides an estimate of the internal dose which is crucial information to understand the causal link between sources of pollutants and potential health effects.
HBM monitors temporal and spatial trends of pollutants in the population. Trends in time may indicate the success of preventive measures to reduce the body burden of pollutants in the population. Geographical differences may allow to identify high exposure groups for which further follow up is needed.
HBM makes pollution personal and raises awareness on environmental health.
However in order to make HBM an effective instrument for regulatory and policy decision making in Europe much more efforts are needed.
Despite a large number of ongoing HBM studies in different member states the data are fragmented and not comparable. No policy interpretation framework exists to link the data to environment and health information. No base line values are available for the European population. For most biomarkers no health based action levels exist.
The European implementation group for biomonitoring has recommended to initiate a pilot project in order to harmonize protocols, make results comparable and elaborate how biomonitoring results can be integrated most efficiently with environmental monitoring and registered health data. Also an effective communication plan is needed.
These recommendations are based on work that has been done in the EU-ESBIO project (www.eu-humanbiomonitoring.org). Biomarker development and validation are high priorities in other frame work projects such as PHIME (www.phime.org) which focuses on heavy metals and their effects, and NEWGENERIS which develops the “omics “ approach and has a focus on effect markers for genotoxicity (www.newgeneris.org). The EU INTARESE project (www.intarese.org/ ) investigates how to apply biomonitoring data in a risk assessment framework for policy decision making .
Several projects and networks are currently set up at different levels within the EU which promise to develop HBM as a powerful instrument for the environmental health action plan.
Placental Uptake and Transfer of Environmental Chemicals Relating to Allergy in Childhood Years EU FPV QoL KA4 (Health & Environment)
The possible link between environmental exposure and adverse health effects in the human population is an area requiring increasing attention within Europe. Prenatal exposure is of particular concern due to increased fetal sensitivity and environmental agents may interfere directly with placental function thus having an adverse effect on the fetus despite long-term low level exposure. Thus, Plutocracy focused on the evaluation of a birth cohort in order to be able to determine exposure to environmental factors during fetal development and examine the risk of allergy development in early childhood following maternal exposure to xenobiotics (XB) such as heavy metals and persistent organic pollutants (POPs).
Pregnant women were recruited from 5 regions of varying environmental characteristics from within the EU (Bratislava and Stara Lubovna in the Slovak Republic, Bucharest and Giurgiu in Romania and Mol, Belgium). Biological samples were collected from mother and neonate and a detailed questionnaire completed. Children underwent a clinical examination at the age of 18 months, a blood sample was collected and a second postnatal questionnaire was completed. The atopic status of the mothers within the cohort was assessed and highlighted significant differences between both regions and countries ranging from 10% in Slovakia to 39% in Belgium. Recent studies have suggested that elevated cord blood (CB) IgE may be an early marker of allergic disease in children. The highest levels of CB IgE positivity (> 0.35 kU/l) were found amongst Romanian neonates and the lowest amongst Slovakian neonates. Mothers of CB IgE positive neonates were more likely to have a higher education level, reside in the city, reside in a house, near heavy traffic, near agricultural fields and animal ownership and smoking patterns were also found to differ significantly between CB IgE positive and negative outcomes. In contrast, the allergic disease status of immediate family members, with the exception of siblings of neonates, did not differ according to CB IgE outcome. The placental concentrations of DDT, DDE, γ-HCH, α-HCH and 1,3,5-TCB were significantly increased in mothers with neonates positive for CB IgE levels. Placental transfer of selected organochlorines was demonstrated in laboratory models and elevated concentrations were found in key fetal organs compared with maternal levels which may have implications for health in later life.
Exposure to toxic metals and organochlorine insecticides was shown to influence immune function as shown by alteration in cytokine production profiles and IgE levels. However, there was no clear evidence of polarisation of the response perhaps due to restricted sample numbers. Sensitisation in children at 18 months of age as assessed by sIgE to specific allergens was lower than expected compared with other studies, perhaps due either to age or geographic location (majority in Eastern Europe). In contrast, the highest incidence of allergic symptoms offspring was found in Belgium which correlates with maternal atopy prevalence and there were significant regional differences. Predictors for atopic eczema included allergy to cow’s milk or other foods, recent ill health such as ‘flu or recent fever, cough, runny nose and daycare outside the home. Th1 and Th2 CK production by cord blood cells was elevated in children in Belgium and Slovakia with atopic eczema by 18 months of age but only Th2 CK production was elevated in peripheral blood cells which may suggest a postnatal bias. Asthma symptoms do not generally appear until later in childhood so follow-up of the cohort is important.
This study provides valuable data which will inform future biomonitoring, prenatal exposure and health intervention studies. Further detailed analysis is expected to highlight other significant outcomes but it can be concluded that in utero exposure to XB is an area of concern and worthy of further investigation.
The Flanders Environment and Health Study: Biomarkers and Biomonitoring
From 2002 to 2006 biomarkers of environmental pollution and of biological effect were analysed in the blood and/or urine of more than 4400 participants to the different campaigns of the Flanders biomonitoring program. Age groups that participated were mother-newborn pairs, adolescents (14-15 years old) and adults (50-65 years old). Aim of the program was to develop a tool for surveillance of environmental health for policy purpose, develop reference values, look at differences between areas. The measured pollutants are persistent pollutants such as: heavy metals (lead, cadmium), chlorinated compounds (dioxin like substances, PCB’s, Hexachlorobenzene and DDE) and metabolites of the volatile compounds benzene and PAHs. Effect markers investigated were on asthma and allergy, growth and development, fertility, endocrine disruptive effects, tumormarkers and DNA damage.
The area of residence determines exposure. People living in the rural areas of Flanders have a high exposure to persistent chlorinated compounds. Cadmium is problematic in some regions. Although use of DDT is forbidden for several years, metabolites are still detected in the human body in considerable amounts. The levels of pollutants reveal no alarming trends. Factors such as age, sex, smoking and nutritional intake are important determinants of exposure.
Asthma and allergy occur frequently in Flanders; regional differences are present between cities and rural areas. Some endocrine and genotoxic markers differ between regions, but the clinical relevance is probably low. Several dose-response relationships were detected: asthma and allergy vs. heavy metals and chlorinated persistent pollutants; endocrine effects vs. lead and chlorinated persistent pollutants; genotoxic markers vs. heavy metals and PAHs.
Exposure Assessment in Environmental Epidemiological Studies Relating to Pregnancy Outcome
Tanya Pless Mulloli
The city of Newcastle upon Tyne in NorthEast England is the setting for the PAMPER (PArticulate Matter and Perinatal Events Research) study which will link air pollution, landuse, socio-economic and weather data to pregnancy outcomes for the period 1961 to 1992 for 90,000+ births. Here we describe the approaches taken to estimate individual black smoke (BS) levels for the time of pregnancy using monitored data alongside information about the location of domestic and industrial chimneys, introduction of smokeless zones, and slum demolition.
Black smoke levels decreased dramatically from around 400mg/m3 in the 1960s to around 20mg/m3 in the 1990s thus providing opportunity to study perinatal outcomes across a very wide range of exposure estimates. The seasonal variation was very pronounced during the earlier years with winter peaks in the order of 600mg/m3 whereas in the 1990s 100mg/m3 was exceeded rarely, consequently seasonal differentials in maternal exposures were pronounced during earlier years and disappeared in later years. However, the seasonal variation also varied between years depending on temperatures in a given year. A dynamic modelling approach was developed to create exposure contours across the study area for each week on the 30year study period allowing both the long term downward trend and seasonal variation to be considered.
Our exposure estimation makes use of very rich data on black smoke, temperature, landuse and domestic chimney density in a novel and unique way thus improving the precision and clarity of error structure of exposure estimates.
PAMPER study team: Tanja Pless-Mulloli, Svetlana Gliniania, Steven Rushton, Judith Rankin, Mark Pearce, Mark Shirley, Roy Sanderson, Rakesh Ghosh, Newcastle University UK, Peter Diggle, Tom Farnshawe, Lancaster University UK, Martin Charlton, National University of Ireland, Louise Parker Dalhousie University, Canada
Identifying Chemical Exposures During Pregnancy in Human Hair
This paper will provide an introduction to the identification of chemical exposures during pregnancy using hair analysis. A summary of the development of hair analysis will be provided including methodological issues. The paper will then concentrate on the detection of recreational drugs using the example of a case control study of gastroschisis where maternal hair analysis was used to validate interview data. Problems encountered during the study will be discussed and solutions provided.
The Use of Biological Monitoring for Assessment of Exposure to Heavy Metals Released from Hazardous Waste Landfill Sites in Poland
2,165 hazardous waste landfills have been registered in Poland. From the priority list of 210 hazardous waste sites, 45 landfills were identified with heavy metals prevailing over other waste materials. The aim of the study was to assess potential risk of exposure to cadmium, chromium and lead released from selected hazardous waste landfills. The study was directly related to exposure of general population living in areas contaminated by the waste sites.
On the basis of the ranking scores and on the identification of exposure pathways for metals released from the sites, three hazardous waste landfills containing cadmium, chromium and lead in accumulated wastes were selected. Taking into account the results of environmental sampling and analysis of metals in groundwater (private wells), soil (fields) and in the air, three impact zones around the landfills were identified. Assessment of environmental exposure to metals released from the sites was based on the measurements of levels of metals in the contaminated media and in biological material (Pb, Cd in blood and Cr, Cd in urine) collected from 230 inhabitants in respective impact zones around the landfills. Heavy metals in biological material were determined using AAS technique. The concentrations of metals in biological material were compared with reference values, i.e. with concentrations typical for occupationally non-exposed populations living in areas with no industrial emitters of heavy metals. The differences between mean values of concentrations of metals in biological material from people inhabiting different impact zones were assessed by analysis of variance with multiple range tests. Potential confounding factors - smoking and occupational exposure to metals, was taken into account.
The results indicate that individuals living in the vicinity of the landfills may be exposed to heavy metals released from wastes sites. Mean cadmium concentration in urine and blood from people inhabiting areas around the particular landfills and mean chromium concentration in the urine, were found to be significantly higher than the reference value (p<0.05). The findings did not confirm the assumption that exposure to metals released from wastes reaches its highest level at the closest proximity to the landfill when it is a primary emission source of the metals in that area.
Selected landfills can create the risk of exposure to metals for general population living in surrounding areas through defined exposure pathway in the environment.
Anomaly Risk Versus Geographical Variation in Exposures: Two Examples
A recently-completed study in the UK examined geographical heterogeneity in congenital anomaly rates in five registry areas comprising about 20% of the UK population (Dolk et al. www.eurocat.ulster.ac.uk/pdf/Geo-Het/Full-Report.pdf). These data were used in particular to investigate the association of risk with two geographically varying exposures: proximity to hazardous waste landfill sites and air pollution. For both studies we used Poisson regression allowing for geographical clustering by inclusion of a random effect at electoral ward level. We also controlled for hospital catchment to allow for variation in assessment, maternal age, and area deprivation. Results confirmed excess risk close to landfill sites but not in areas of high air pollution. Sensitivity analyses showed the importance of good ascertainment and/or allowing for variation in ascertainment. General epidemiological methodology needed only modest adaptation for the geographical context.
Causal Diagrams and Confounding
Based on the work of Greenland, Hernan and colleagues, the objective of this presentation is to provide a brief introduction to the counterfactual model of causation and causal diagrams for epidemiological research. The utility of causal diagrams (directed acyclic graphs) for addressing issues related to confounding and selection bias are outlined. Topics introduced include the notion of sufficient control, and the "backdoor criterion" for its assessment, in making adjustments aimed at minimizing bias due to confounding. Hypothetical examples illustrating the potential application of causal diagrams for analyzing data on the effects of environmental exposures are included.
High Birth Prevalence of Hypospadias and Other Congenital Anomalies in Two Petrochemical Areas of Sicily, Italy
Two of the largest petrochemical areas in Europe are located in Sicily-Italy: the Augusta-Priolo site (33,5 km2) along the east coast (province of Siracusa) and the Gela site (4,7 km2) along the South coast (province of Caltanissetta). Both areas host refineries, oil tanks, pipelines, chemical (petroleum derivatives, polymers) and cement-asbestos factories; in addition, the first site comprises electric power stations, shipyards, incinerators and landfill sites. These two areas have been defined by the Italian Ministry of the Environment as “at high risk of environmental crisis” and are included in the national list of reclamation sites.
Accounts by local physicians, media and scientific reports1 on increasing trends or clusters of congenital anomalies (CAs), cancer incidence and mortality have raised a lot of concern in the communities living in these areas.
Recently a prevalence study was conducted in the municipality of Gela. As the Sicilian registry, active since 1991 over a great part of Sicily, has started lately to operate in Gela, information on CAs over 1991-2002 was collected by reviewing nearly 50.000 Gela hospital discharge records and integrated with Sicilian registry data for newborns of mothers residing in Gela who delivered outside the municipality. Five hundred and twenty malformed newborns were ascertained out of 13,060 births over 1991-2002 (including terminations of pregnancy since 1995), which represents a prevalence rate of 398/10.000. This value is statistically significant and is two times higher than the one observed by the Sicilian registry (182/10,000) and other Italian registries (205/10,000) as showed in table 1 and figure 1. Also, the hypospadias rate turned out to be 56.7/10,000, which is a very highly significant value if compared to that of the Sicilian registry and to the mean of the Italian registries (O/E 2.9, 2.7, respectively) (table 1 and figure 1). Other statistically significant excesses were observed for malformations of cardiac septa, great vessels and central nervous system, due to a high rate of microcephaly. Newborns and termination of pregnancies diagnosed with neural tube and upper limb reduction defects were significantly higher compared with the mean of Italian registries (O/E 2.2 and 2.6).
These data confirm results of an epidemiological investigation on newborns with CAs residing in the Siracusa province (21 municipalities including Augusta and Priolo) conducted over 44,586 births occurred from 1991 to 20002. The study revealed a high rate of hypospadias, particularly in the Augusta-Priolo area (40.6/10,000 including all degrees of the condition), significantly higher if compared to the one of the Sicilian registry and other Italian regional registries (observed/expected ratio [O/E] 2.4, 1.9 respectively)2. The municipality of Augusta registered the highest hypospadias rate: 58.1/10,000. This value, very close to the one of Gela, is rarely reported in published literature. Statistically significant excesses of newborns with digestive system and cardiovascular malformations in the Augusta-Priolo area emerged as well.
Another study by Bianca et al3 found a significant elevated incidence of hypospadias in Augusta and Vittoria - two Sicilian towns - although it must be stressed that this article considered exposure both to industrial and agricultural pollutants.
Hypospadias is known to be susceptible to reporting misclassification problems, and the role of genetic and other environmental and occupational factors cannot be excluded4. However, the very high rates of hypospadias found in two industrial areas in which the presence of similar teratogens, mutagens and endocrine disruptors is documented or suspected, claim attention and call for further investigations not only to address public concern but also for scientific purposes.5
The potential complex exposure at several different pollutants documented in the areas imply to design advanced epidemiological studies using biomarkers and environmental monitoring data to improve the exposure assessment both at ecological and individual level.
Two case-control studies on selected malformations to investigate associations with residential and occupational exposure, diet habits and other potential confounders have been carried out in Augusta-Priolo and Gela areas: significant risk associations resulted for consumers of fish, fruit and vegetables if purchased at street vendors or for fishing and growing own food plants (OR from 3.0 to 50.0).
Two surveys on mercury and PCBs in puerperal hair and milk respectively have been conducted in Augusta and in a control area: both showed higher average intake in subjects living in the polluted areas compared with residents in the control area. At the moment only summary results of these investigations can be communicable since they are part of a Public Prosecutions procedure.
1. Martuzzi M, Mitis F, Biggeri A, Terracini B, Bertollini R. Environment and health status in the population of the areas at high risk of environmental crisis in Italy. Epidemiol Prev 2002; 26 (6) (Supplement).
2. Bianchi F, Bianca S, Linzalone N, Madeddu A. Surveillance of congenital malformations in Italy: an investigation in the province of Siracusa. Epidemiol Prev 2004;28(2):87-93.
3. Bianca S, Li Volti G, Caruso-Nicoletti M, Ettore G, Barone P, Lupo L, et al. Elevated incidence of hypospadias in two Sicilian towns where exposure to industrial and agricultural pollutants is high. Repro Tox 2003;17:539-45.
4. Vriijheid M, Armstrong B, Dolk H, van Tongeren M, Botting B. Risk of hypospadias in relation to maternal occupational exposure to potential endocrine disrupting chemicals. Occup Environ Med 2003;60:543-550.
5. Laurence S. Baskin. Hypospadias and Genital Development. Philadelphia (PA): Kluwer Academic Plenum Publishers; 2004:258.
Welsh Landfill and Gastroschisis Study
Concern that living near a particular landfill site in Wales caused increased risk of births with congenital malformations led to the examination of whether residents living close to 24 landfill sites in Wales experienced increased rates of congenital anomalies after the landfills opened compared with before they opened. Expected rates of congenital anomalies in births to mothers living within 2 km of the sites, before and after opening of the sites, were estimated from a logistic regression model fitted to all births in residents living at least 4km away from these sites, adjusting for hospital catchment area, year of birth, sex, maternal age, and socioeconomic deprivation score. All births from 1983 through 1997 with at least one recorded congenital anomaly (ICD-9 codes 7400-7599; ICD-10 codes Q000-Q999) were investigated. The ratio of the observed to expected rates of congenital anomalies before landfills opened was 0.87 (95%CI 0.75, 1.00), and this increased to 1.21 (95%CI 1.04, 1.40) after opening, giving a standardized risk ratio of 1.39 (95%CI 1.12, 1.72). Enhanced congenital malformation surveillance data collected from 1998 through 2004 showed a standardised risk ratio of 1.02 (95%CI 0.96, 1.08). Causal inferences are difficult because of possible biases from incomplete case ascertainment, lack of data on individual-level exposures, and other socioeconomic and lifestyle factors that may confound a relationship with area of residence1. Furthermore, with multisite studies there are problems with heterogeneous exposures. Whilst concentric circular regions centred on a site are usually used as a surrogate for defining exposed and unexposed populations, this approach does not take into account the actual spatial pattern associated with the population or toxicant dispersion. A kernel density technique to map risk contours for disease, which is not influenced by the coordinates of any putative environmental hazard and which could be married to actual spatial exposure patterns is described2.
In recent years there have been three clusters of gastroschisis in Wales. Investigations into these clusters identified some risk factors such as young maternal age but no common cause has been identified. These investigations highlighted difficulties in obtaining information on individual exposures in a timely manner. In response to this we have set up an enhanced surveillance programme using methods employed in communicable disease epidemiology for the investigation of disease outbreaks, to identify and collect information prospectively about modifiable lifestyle, dietary and environmental factors, with the shortest possible recall period. We identify mothers of babies diagnosed with gastroschisis as soon as possible after diagnosis (usually after the 20 week anomaly scan). We developed a questionnaire that includes information on diet, infections, exposure to chemicals through occupation, hobbies, home improvement and lifestyle activities, and also open questions to explore what the women’s perceptions were about the cause of their baby’s condition, what places they visited, how they spent their leisure time and what activities they took part in. Particular emphasis is placed on establishing in depth and detail exposures that occurred during the window of the 5th to 10th week of gestation. The aim is to generate hypothesis about the cause of gastroschisis that can be tested in future case-control studies. For more information about this enhanced surveillance programme and proposed case-control studies please contact Dr. Shantini Paranjothy (ParanjothyS@cf.ac.uk).
1. Palmer SR, Dunstan FDJ, Fielder H, Fone DL, Higgs G, Senior ML. Risk of Congenital Anomalies after the Opening of Landfill Sites. Environmental Health Perspectives. Vol 113 (10) October 2005 p1362-1365.
2. James L, Matthews I, Nix B. Spatial Contouring of Risk. A Tool for Environmental Epidemiology. Epidemiology. Vol 15 (3) May 2004 p287-292.
Exposure to a Solid-Waste Disposal Site in Cali, Columbia and Congenital Malformations
Exposure to a solid waste deposit may contribute to the occurrence of congenital malformations in Cali, Colombia. In particular, cadmium and lead have been found in water samples from monitoring wells around the municipal deposit. There is a possibility that women get exposed to those metals, because lixiviates reach a river that is used by the city water supply system.
The hospital based surveillance system in the city (part of the Latin American network ECLAMC) not only has called the attention for the elevated incidence of neural tube defects and other major malformations, but identified an unparalleled cluster of 4 sirenomelias (2.4 cases/1,000 births) and 4 cyclops (3.4 cases/1,000 births) in a small area of the city and during a 16 months period between 2004-2005. Unfortunately, study of the cluster was retrospective and data analysis was not conclusive.
A study will be conducted to evaluate routes of exposure to heavy metals, and other pollutants potentially associated with major congenital malformations in the area. Initially, we will assess routes of exposure by monitoring and modeling toxic agents in the environment, and will determine habits and practices of women by using questionnaires and diaries in the field.
In a second phase, we will select most frequent defects and will carry out a case control study in the city to evaluate presence of pollutants in the mother. We are planning to include measurement of cadmium and lead in hair.
Current Research in Lithuania: Exposure Assessment in Kaunas Newborns Cohort Study
We conduct a population-based newborns cohort study and apply a nested case-control design to study relationship between of birth outcomes and drinking water disinfection by-products in genetic susceptible subjects. We measure spatial variation of disinfection by-products concentrations and use questionnaire information to assess individual cumulative exposure. The study deliverables are: databases of drinking water disinfection by-products levels and air pollutants concentrations, its spatial variation, databases of congenital malformations, premature birth, small for gestation age and allergic infants by genetic susceptibility and risk estimates, including exposure-response relationships for indices of fetal development and allergy.
Current Research on Environmental Pollution and Congenital Anomalies in Belarus
Population-based registry of congenital malformations has been operating in Belarus since 1979 covering entire population of the republic and providing a good opportunity to study possible genetic and teratogenic effects of environmental pollution . The most severe ecological accident in Belarus requiring thorough epidemiological consideration was Chernobyl catastrophe resulting in radionuclide contamination of ~1/3 of its territory .
In order to study possible aftermath of the Chernobyl accident we performed a time trend analysis of the prevalence at birth of congenital anomalies defined for registration since the registry foundation – so called “mandatory registered congenital malformations” easily diagnosed within neonatal period: (1) anencephaly, (2) spina bifida, (3) cleft lip and (or) palate, (4) polydactyly, (5) limb reduction defects (6) oesophageal atresia (stenosis), (7) rectal atresia (stenosis), (8) Down syndrome and (9) group of multiple congenital malformations. Comparison of prevalence rates between the regions contrasting by radionuclide contamination was done basing on large (oblast) and small (district) administrative divisions of Belarus.
Since the middle of 80-s a steadily increasing trend was traceable in the republic with a tendency to slowdown during the last decade. Similarity of the time-trend for the areas with high and low radionuclide contamination indicates that the Chernobyl accident could not be considered as a main cause of the finding . Nevertheless, a prominent excess of congenital anomalies under study was marked in highly contaminated areas during the first years after the accident when registered prevalence rates significantly exceeded as pre-accidental values as the rates marked in non-contaminated areas during the same time period .
The main contributors to the marked increase were polydactyly, reduction defects of limbs and the group of multiple congenital malformations. Correlation analysis basing on average for district population effective doses revealed a positive association for this group of congenital anomalies . No similar association was found for other anomalies analyzed as well as for all types of congenital malformations within the further post-accidental time period. All 119 districts of the republic were included in the analysis. Districts were aggregated into 5 exposure groups. Weighted average effective dose was calculated for each exposure group.
To confirm the obtained results a correlation with radionuclide contamination in 1299 settlements of 17 highly-contaminated districts was studied within the period of 1987-1989 . All settlements of the districts, where children were born within the mentioned period, were included in the study. Aggregation into 4 exposure groups was performed and weighted average contamination level was calculated. Positive association was confirmed for skeletal anomalies, but not for the group of multiple congenital malformations.
No substantial changes in the contribution of syndromes with dominant inheritance were observed for the group of verified multiple congenital malformations registered in 1987-1989 in 17 contaminated districts as compared to the group of probands born in the same area in pre-accidental period (1981-1986). Only slight increase was marked for systemic skeletal dysplasias (6.4% v.s. 2.2%, respectively) and for autosomal recessive syndromes (8.5 v.s. 6.5%) being actually non-significant because of small numbers. Majority of cases (up to 60%) had a diagnosis of unclassified multiple congenital malformations making difficult interpretation of the results.
The analysis of monthly prevalence revealed a Down syndrome cluster in January of 1987 . The time of appearance and the spatial distribution of cluster cases assume an association with the exposure during the first weeks after accident due to substantial release of short-lived radionuclides when exposure dose rates exceeded natural background levels by 1-3 orders of magnitude on considerable part of Republic of Belarus . Contribution of main risk factor (maternal age effect) was excluded.
Several biases and confounding could affect our results obtained using descriptive epidemiological approach. Misclassification could be a potential cause of the changes in the prevalence rates of multiple congenital malformations, but unlikely to play any substantial role for polydactyly and limb reduction defects. No special screening program was initiated during first years after Chernobyl accident, but some better ascertainment could be possible. Nevertheless, our estimate for reporting completeness exceeds 85% for all the regions and periods analyzed . Stochastic deviations seems doubtful explanation for 3-year increase of skeletal and multiple congenital anomalies, but possible for Down syndrome cluster registered in January of 1987. Probability of observation of a maximal increase at each separate month among 108 analyzed is low, however. Impact of other environmental and social factors associated with the accident and subsequent relocation could not be excluded.
From animal studies it is known that dominant mutations with clear-cut phenotypic effects and full penetrance are induced relatively rarely; dominant effects within multifactorial genetic systems, e.g. mutations affecting the skeleton, appear to be more common . This is in good agreement with obtained results indicating absence of an increase of dominantly inherited syndromes, but demonstrating enhanced prevalence of relatively common skeletal anomalies of multifactorial nature with considerable mutational component.
Another well known effect of ionizing radiation is meiotic disjunction failure resulting in aneuploidy in the offspring of exposed animals. The hours around fertilization are especially susceptible [8,9]. Thus, conception of aneuploid human fetuses during the period of relatively high exposure dose rates could be a possible consequence of irradiation of the vulnerable meiotic stages.
1. No clear-cut long-term effects of chronic low dose ionizing radiation on the prevalence at birth of congenital
malformations were demonstrated in Belarus. Steadily increasing trend observed since the middle of 80-s is likely to
have the same reasons in contaminated and contamination-free areas most probably not associated with radiation.
2. Enhanced prevalence rates of skeletal anomalies and Down syndrome registered in highly contaminated areas
during the first years after the accident does not exclude adverse impact of acute radiation exposure due to
substantial release of short-lived radionuclides.
3. Possible confounding and biases typical for descriptive epidemiological studies prevent from formulation of the
final conclusions. Obtained results need further confirmation in analytical studies performed in different exposed
4. Autosomal trisomies and skeletal anomalies seem to be promising groups for monitoring of exposed populations.
1. Lazjuk G, Verger P, Gagniere B, Kravchuk Z, Zatsepin I, Robert-Gnansia E. The congenital anomalies registry in
Belarus: a tool for assessing the public health impact of the Chernobyl accident. Reprod.Toxicol. 2003;17(6):659
2. The International Chernobyl Project Technical Report: assessment of radiological consequences and evaluation
of protective measures. Report by an International Advisory Committee. Vienna: IAEA; 1992.
3. Lazjuk G.I., Nikolayev D.L., Novikova I.V., Polityko A.D., Khmel R.D. Belarussian population radiation exposure
after Chernobyl accident and congenital malformations dynamics. Int.J.Radiat.Med. 1999; 1(1):63-70.
4. Zatsepin I.O., Naumchik I.V., Babicheva I.L., Khmel R.D. Study of the prevalence of embryonic development
failures among the offspring of people exposed basing on accumulated dose due to the Chernobyl accident:
newsletter N 1. Belarus Research and Healthcare Center “Mother and Child”, Minsk. 2005; 25 pp. (in Russian).
5. Zatsepin I.O., Naumchik I.V., Babicheva I.L., Khmel R.D. Study of possible genetic consequences of the
Chernobyl accident: newsletter N 2. Belarus Research and Healthcare Center “Mother and Child”, Minsk. 2006; 20
pp. (in Russian).
6. Zatsepin I, Verger P, Robert-Gnansia E, Gagniere B, Khmel R, Lazjuk G. Cluster of Down's syndrome cases
registered in January of 1987 in Republic of Belarus as a possible consequence of the Chernobyl accident.
Int.J.Radiat.Med. 2004; 6(1-4):57-71.
7. Vogel F. Risk calculation for hereditary effects of ionizing radiation in humans. Hum.Genet. 1992; 89:127-146.
8. Evans H.J., Lyon M.F., Czeizel A. International Commission for Protection against Environmental Mutagens and
Carcinogens. ICPEMC meeting report no. 3. Is the incidence of Down syndrome increasing? Mutat.Res. 1986; 17
9. Tease ?, Fisher G. Cytogenetic and genetic studies of radiation induced chromosome damage in mouse oocytes. Mutat.Res. 1996; 349:145-153.
As part of the EUROCAT project, it is intended to identify clusters of excess numbers of congenital anomalies. A variable window width scan has been implemented and used for a number of years for this purpose. The scan statistic provides a metric for assessing clusters based on number of cases and temporal proximity. Due to the nature of both the scan statistic and the CA data, the methodology as presented by Nagarwalla is restrictive. Adaptations have been made that enable the identification of multiple clusters that may or may not overlap. A brief discussion of problems relating to population changes is also included.
The EDMP Software for Detection of Time Clusters - User Friendly Software for Congenital Anomaly Registers
The EDMP software has been developed to assist registries in managing their congenital anomalies data and its transmission to the Eurocat Central Registry. Key elements of the software include flexible data entry or import options, data validation and an export function that enables registries to send their data to the Central Registry in the correct format. EDMP also allows users to analyse and report on their data in a number of ways and also provides allows for the statistical surveillance of trends and clusters.
Each case in EDMP is automatically coded as belonging to one or more of the 95 pre-defined congenital anomaly subgroups based on the ICD malformation codes assigned to each case. Each subgroup is defined in terms of ICD 9 and 10 codes so for instance a case with any mention of Q90 or 7580 would be assigned to the Downs subgroup. EDMP also allows users to define their own anomaly subgroups. Trend and cluster surveillance checks each of the anomaly subgroups individually.
Checking for time clusters is simple as all that is required is the start and end date for the period under surveillance and EDMP will output the results directly into an Excel spreadsheet. In addition a more detailed report can be printed for any significant clusters detected.
EDMP imposes a number of rules including:
• Minimum of seven cases
• Chromosomal cases excluded for non-chromosomal anomaly subgroups
• Maximum 18 month cluster length with the last case must falling within 2 years of end date
• Calculated conception date is used in preference to date of birth
• Missing gestations estimated from mean for year, subgroup and type of birth
• Maximum 10% estimated gestations permitted otherwise cluster is checked by date of birth
• Conception ‘end date’ is end date – 9 months
Two Years Experience of Routine Cluster Detection and Investigation by EUROCAT Registers
EUROCAT aims to provide essential epidemiologic information on congenital anomalies in Europe and to co
ordinate the detection of, and response, to clusters and early warning of teratogenic exposures. This presentation
summarises some of the key findings of EUROCAT’s Statistical Monitoring Report 2000-2004 for 27 EUROCAT
registries covering together approx 700,000 births per year.
Statistical monitoring to detect temporal clusters (by scan analysis) was run on the data of 27 registries for 75 EUROCAT congenital anomaly subgroups, requiring a minimum of 7 cases over the 5 year monitoring period, a total of 1067 tests. The majority of cluster tests were based on estimated date of conception rather than date of birth. Clusters occurring wholly or partly during birth years 2003-2004 and of less than 18 months duration were reported. A total of 56 clusters were detected (5.2% of tests). Fifteen of the 27 registries reported the results of preliminary cluster investigations (27 clusters). Outcomes of local registry investigations into the detected clusters found that 11% were judged to be cause for concern with further ongoing investigation; 22% confirmed excess cases, but no further investigation was proposed other than further investigation; 22% were not ‘true’ clusters as they were associated with aetiologic heterogeneity, changes in diagnosis, familial or twin recurrence; 15% reported increases in cases due to changes or improvements in prenatal detection techniques; and 30% were explained by data quality issues.
Analysis of trends 2000-2004 was conducted for 94 congenital anomaly subgroups (including 19 broad heterogeneous subgroups useful for descriptive purposes only) in 27 registries, a total of 2419 trend tests. This revealed 75 significant increasing trends (3.1%), 156 significant decreasing trends (6.4%), and 183 with significant non-linear yearly heterogeneity (7.6%). Eight registries reported investigations. Outcomes of local registry investigations into 58 of the 231 increasing or decreasing trends (25%) found that 31% were detected in the large heterogeneous subgroups; 21% were caused by data quality errors or changes in reporting practices; 19% were identified in subgroups with too few cases to interpret (future analyses will set a lower threshold); 7% disappeared when trend investigations considered the whole time period of the registry or included year 2005 data; 10% were due to increasing prenatal diagnosis; 3% were judged to require continued monitoring; no explanation was offered for the remaining 9%.
After several pilot years leading to refinement of methodology, this is the first year that EUROCAT can report routine monitoring results. Local registry investigation is crucial, and more needs to be done to increase the number of registries reporting investigations, and refining investigation protocols. Of those clusters and trends investigated, few were judged to warrant further investigation, and will be followed up in future years. Further consideration needs to be given to the time period of trend analysis, and a protocol for investigating similar patterns across registries. We find that statistical monitoring has a dual purpose as a data quality control mechanism.
Use of Satscan to Detect Spatial Clusters in Five Regions of Britain
A recently-completed study in the UK examined geographical heterogeneity in congenital anomaly rates in five registry areas (Dolk et al. www.eurocat.ulster.ac.uk/pdf/Geo-Het/Full-Report.pdf). In particular, among other techniques, we used Kulldorf’s SaTScan to investigate whether there was evidence for localised clustering, after allowing for known risk factors. The method seeks unusually high ratios of observed over expected anomalies in all possible circles on a map, allowing for the implicit multiple testing. Application of the method was reasonably straightforward, the principle choice being of the largest radius of the multiple circles to be considered – taken here as the size (in population) of a typical hospital catchment area, because we believed that variation between catchment areas could be due to variation in ascertainment. From investigating thirty anomaly sub-type groupings we found two nominally significant circles. Both contained siblings, possibly reflecting unrecognised familial aetiology. In conclusion, in these data the SaTScan technique was useful to find data errors and possible unrecognised familial anomalies but we found little evidence for clusters likely to be unhypothesised environmental risks, despite a focused study of one environmental exposure (landfill sites) being positive. This highlights limitations of power in general cluster-seeking without prior hypotheses.