Study population

The study sites were selected in part if they had established clinical obstetrical research infrastructure in place and also to represent different geographical regions of the country (Figure?1). Clinical site investigators from the 10 cities across Canada were asked to recruit a total of 2000 pregnant women from the general population who were attending prenatal clinics (ultrasound, midwife and/or doctor’s clinics) during the first trimester of pregnancy (6 to <14 weeks). Recruitment took place between 2008 and 2011. The selected cities range from some of the largest in Canada, Toronto and Vancouver, where only 53% of the population identify English as their only mother tongue, to a smaller northern city such as Sudbury where 66% are English, and to Montreal where 52% identify their first language as French.

Eligibility criteria included ability to consent and to communicate in English or French, age 18 years or older, <14 weeks gestation, willing to provide a sample of cord blood and planning on delivering at a local hospital. Women with the following medical history were excluded from the study:

Data and biospecimen collection

Contacts were made with participants during each trimester, at delivery and in the early postnatal period (up to 10 weeks) to collect data and biospecimens (i.e. blood, urine, cord blood, meconium, breast milk and hair) for planned laboratory analyses (Figure?2). Under a separate consent form, additional biospecimens were collected to be stored in the MIREC Data and Biological Specimens Bank (MIREC Biobank). Questionnaires were administered by trained research staff to participants during the first and third trimesters to collect demographic, life style (e.g. smoking and alcohol), medical history, use of natural health products and medications, and potential sources of exposure data. Where possible, questions used in previous surveys and studies were used for comparability of results. A validated food frequency questionnaire was administered in the second trimester, along with blood and spot urine collection, blood pressure, clinical laboratory tests, and anthropometric measurements. Medical chart information was extracted at the first trimester ultrasound, second trimester and post?delivery time points. Blood pressure, height, weight and a proteinuria test were taken at each trimester.

Figure?2 provides details on participant contacts and biospecimen collection and analyses for environmental chemicals, nutrients, genetic factors, markers of oxidative stress and immunoprotective factors. Maternal urine was collected in Nalgene® containers (Thermo?Fisher Scientific Inc., Rochester NY, USA), and a Mère Hélène® bioliner (Mère Hélène, Quebec, Canada) was used on the diapers to collect meconium. All collection containers were pretested for phthalates and BPA, and field blanks were incorporated for each major chemical analyte. Post delivery, research staff visited the home of participants to collect samples of maternal milk and hair, and to complete a questionnaire on dietary factors. Women were asked to hand?express hind? and fore?milk over multiple days between 2 and 10 weeks post delivery; if they were having difficulty hand?expressing, a Medela® (Medela International, Zug, Switzerland) manual breast pump was provided, along with instructions for either manually expressing or using the breast pump to collect the milk. The milk was collected in 16?oz wide mouth amber I?CHEM® glass jars with fluoropolymer resin?liner polypropylene closure (Thermo Fisher Scientific, Rockwood, TN, USA) and 16?oz wide mouth TraceClean® clear plastic polyethylene jars (VWR International, Radnor, PA, USA). The breast pump was tested for possible contamination with phthalates and BPA.

Chemical analyses of maternal blood and urine, cord blood, and meconium were carried out by the Toxicology Laboratory, located in the Institut national de santé publique du Québec (https://www.inspq.qc.ca/ctq/Default.asp?Page=1&Lg=en), which is accredited by the Standards Council of Canada under ISO 17025 and CAN?P?43. The accuracy and precision of the analyses are evaluated on a regular basis through the laboratory’s participation in external quality assessment programmes. DNA was extracted from maternal blood, and genotyping was performed by the Plateforme de séquençage et de génotypage des génomes at the Centre de recherche du CHUL/CHUQ (http://www.sequences.crchul.ulaval.ca/eng/index.html).

Nutritional analyses of the maternal and cord blood samples, as well as the chemical and nutritional analyses of the breast milk, are being carried out by the laboratories of the Food Directorate at Health Canada. Markers of oxidative stress, inflammation and vasoregulation are being measured by scientists from the Environmental and Radiation Health Science Directorate of Health Canada.

All questionnaires and biospecimens were labelled with a unique ID and barcode, respectively. Data from questionnaires were entered into a database using single data entry with queries and verification of responses dealt with as needed (e.g. values outside range, internal inconsistencies). A separate database which tracked the collection and laboratory analysis of the biospecimens was also created. All blood, urine and breast milk were aliquoted into smaller cryovials and stored at ?20 or ?80°C as required.

Follow?up studies

In subsequent and ongoing follow?up studies, assessments of child health and development are being made on a subset of the cohort at various ages (see Figure?3). In the MIREC?ID Study (MIREC: infant development to 6 months), in?clinic assessments were performed on approximately 400 infants at birth and 6 months of age to measure growth, sensory function, behaviour and potential indicators of reproductive effects (e.g. anogenital distance). MIREC?CD3 (MIREC: child development at age 3), currently underway, is an online parental survey for children at 36 months of age that is designed to examine potential effects of prenatal exposure to phthalates, BPA and organophosphate pesticides on child neurobehaviour [e.g. subscales of the Behavior Rating Inventory of Executive Function (BRIEF?P)]. The MIREC?CD Plus Study, currently in the planning stage, entails a home visit for children 15 months to 5 years of age to collect a sample of the child’s blood and urine, to measure physical growth and only in children older than 2.5 years, to measure important determinants of child neurodevelopment (behaviour, general cognitive abilities, specific executive functions, and language and communications skills). In the MIREC?CD Plus Study, biological specimens will be collected to measure heavy metal exposure for those children under 3 years of age, with additional biospecimens placed in the Biobank. Delays in obtaining funding (which meant some children were too old for some of the assessments), insufficient funding and difficulties in obtaining ethics approvals at some of the recruitment sites have adversely affected our ability to include all of the cohort children in the follow?up studies. In the MIREC?CD Plus Study, to improve efficiency and reduce costs, recruitment is being restricted to six sites with the most MIREC children.

Ethical considerations

The research protocol, questionnaires, consent forms, and recruitment posters and pamphlets were reviewed and approved by human studies research ethics committees, including the Research Ethics Board at Health Canada and the research ethics committee at the coordinating centre at Ste?Justine’s Hospital in Montreal, as well as >10 academic and hospital ethics committees across Canada – a very time?consuming process. Any changes made to the protocol or survey instruments (amendments) were approved by all ethics committees. Separate consent forms were developed for the data and biospecimen repository (Biobank) and for the infant follow?up studies. Participants could partially (data and biospecimens retained) or completely (all data and biospecimens destroyed) withdraw from the study. Ethical considerations did not permit the collection of any information on the non?participants to examine possible selection bias.

The approach to reporting biomonitoring results to participants was a major point of discussion between the study investigators and the hospital ethics committees at the coordinating centre and recruitment sites.7 Our initial proposal to the ethics committees was that individual biomonitoring results would be provided to participants (if they so requested) at the conclusion of the data collection and the analysis phase of the initial cohort study, along with information on potential sources of exposure. The exception would be chemicals for which there were health?based tissue guidelines [i.e. blood lead, mercury and cadmium (in the case of cadmium, the workplace health guideline was used)]. Any participant with blood levels exceeding the guidelines would be informed as soon as possible and provided with information on reducing exposure, including possible follow?up blood measurements. However, the hospital ethics committees required that only biomonitoring results exceeding health?based tissue guidelines where preventive or treatment options were available could be shared with participants through their health care provider. This latter process was, therefore, used to share maternal blood concentrations of lead, mercury and cadmium that exceeded established guidelines with the participant’s physician. The physicians were also provided with guidance documents to help them identify potential sources of exposure for their patients and to help them interpret the individual results. A physician with expertise in environmental epidemiology of these metals was made available to the participant’s physicians to provide further advice, as needed. In the province of Quebec, a number of chemical substances, including lead, mercury, cadmium and manganese, as well as the dialkyl phosphate metabolites, fall under reportable diseases legislation [maladies et intoxications à déclaration obligatoire (MADO)]. Therefore, our Montreal sites were legally obliged to report elevated results to the public health department, according to the MADO guidelines. Recently, we have had a request from an MIREC participant for her individual biomonitoring results for environmental chemicals, and we have been given permission by the ethics committees to provide these results to her with the disclaimer that we cannot provide any interpretation of her results as there is currently no scientific knowledge available to interpret these results at the individual level.

Perspectives

MIREC is a multisite study with a population of obstetric patients from across Canada. One of the unique features of this study is the most extensive assessment of prenatal and lactational exposure to environmental chemicals at multiple time points, but especially in early pregnancy when the fetus is likely most sensitive to toxic chemicals. In addition, the establishment of the Biobank will facilitate future research on additional chemicals and genomic and nutritional susceptibility factors. As such, MIREC is directly responsive to the exposome paradigm,8 measuring endogenous factors such as oxidative stress, specific external exposures such as chemicals and diet, and general external exposures including economic and educational factors. By doing so, the findings from the MIREC Research Platform will contribute to the identification and quantification of exposomes during sensitive windows of reproduction and development, as recently recommended by Buck Louis and colleagues.9

The results of MIREC may not be generalisable to the Canadian population or to each of the recruitment sites as the study is not population?based. Past experience has highlighted the difficulties in trying to assemble a population?based pregnancy/birth cohort that may be less representative as the cohort is followed over time.10–12

The MIREC participation rate of 39% is consistent with participation rates of several large prospective cohort studies.13 Similar to other pregnancy cohort studies (e.g. Bornehag et?al.11), participants in MIREC tended to be older, more educated, born in Canada, married and less likely to be a current smoker than the Canadian population giving birth in 2009 (Table?2).14 The MIREC participants were, however, more similar to a population of women 20–39 years of age participating in the biomonitoring component of the population?based Canadian Health Measures Survey (Cycle 1 2007–09).15, 16

In summary, the multiple contacts with study participants starting in the first trimester, and the collection of data and biospecimen analysis for multiple key chemical risk factors in the MIREC Study, should provide important information on several hypotheses related to prenatal exposure to environmental chemicals and potential adverse health effects for the pregnant woman and her child. The major challenges are funding to maintain the Biobank and to follow the cohort as the child ages, as well as keeping the families actively participating in the various ancillary studies, while limiting biases that might affect the internal validity of the study.